Pseudomonas virulence factor controls expression of virulence genes in Pseudomonas entomophila

Abstract

Quorum sensing is a communication strategy that bacteria use to collectively alter gene expression in response to cell density. Pathogens use quorum sensing systems to control activities vital to infection, such as the production of virulence factors and biofilm formation. The Pseudomonas virulence factor (pvf) gene cluster encodes a signaling system (Pvf) that is present in over 500 strains of proteobacteria, including strains that infect a variety of plant and human hosts. We have shown that Pvf regulates the production of secreted proteins and small molecules in the insect pathogen Pseudomonas entomophila L48. Here, we identified genes that are likely regulated by Pvf using the model strain P. entomophila L48 which does not contain other known quorum sensing systems. Pvf regulated genes were identified through comparing the transcriptomes of wildtype P. entomophila and a pvf deletion mutant (ΔpvfA-D). We found that deletion of pvfA-D affected the expression of approximately 300 genes involved in virulence, the type VI secretion system, siderophore transport, and branched chain amino acid biosynthesis. Additionally, we identified seven putative biosynthetic gene clusters with reduced expression in ΔpvfA-D. Our results indicate that Pvf controls multiple virulence mechanisms in P. entomophila L48. Characterizing genes regulated by Pvf will aid understanding of host–pathogen interactions and development of anti-virulence strategies against P. entomophila and other pvf-containing strains.

Introduction

Quorum sensing is a form of bacterial cell-to-cell communication that regulates collective behaviors in response to changes in cell density through the secretion and detection of signaling molecules called autoinducers [1, 2]. Quorum sensing influences biofilm formation [3, 4] as well as the production of small molecules [5] and virulence factors [6]. Some virulence factors are toxins, while others enable pathogens to evade host defenses [7], colonize and persist in the host, and outcompete other microbes [8]. Characterization of virulence factors and their regulation by quorum sensing is vital for understanding how bacteria cause infections and for developing anti-virulence strategies [9–12]. The pathogen Pseudomonas aeruginosa uses multiple quorum sensing systems to regulate gene expression and virulence [13], including those based on N-acyl homoserine lactones [14, 15] and 2-alkyl-4-quinolones [16–18]. However, many Pseudomonas spp., such as Pseudomonas entomophila L48, do not produce these quorum sensing molecules, limiting our understanding of signaling and virulence in these organisms [19].

Pseudomonas entomophila L48 kills Drosophila melanogaster larvae and adults after ingestion by persisting in the Drosophila gut and inflicting irreversible damage [20]. The oral infection route of P. entomophila provides a model for unraveling the mechanisms of Drosophila innate immunity, which can be extended to higher order eukaryotes [21]. P. entomophila uses a global regulatory system to control virulence—the GacS/GacA two-component signaling system (Gac) for which the signal remains unknown [22]. In 2010, a putative biosynthetic gene cluster (BGC), the Pseudomonas virulence factor (pvf), was found to be responsible for synthesizing unknown small molecules that are required for full virulence against Drosophila [23].

The pvf cluster encodes a nonribosomal peptide synthetase (PvfC), a non-heme diiron N-oxygenase (PvfB), and two proteins with unknown function (PvfA and PvfB) (Figure A in S1 Fig). We have shown in P. entomophila L48 that pvf-encoded enzymes synthesize extracellular small molecules that autoinduce pvf expression (PVF autoinducers). Expression of pvf in a reporter strain depends on cell density of the culture or the amount of cell-free supernatant added from cultures that express pvfA-D, which suggests quorum-sensing properties of the small molecules produced by pvf-encoded enzymes [24]. The genome of P. entomophila L48 does not harbor homologs of biosynthetic genes for autoinducers such as N-acyl homoserine lactones (luxI) or 2-alkyl-4-quinolones (pqsABCDE) [23, 25]. P. entomophila L48 is therefore a good model system for studying the Pvf signaling system (Pvf for short). Additionally, since the pvf cluster is conserved in more than 500 strains of proteobacteria, including the plant pathogen Pseudomonas syringae and the opportunistic human pathogen Burkholderia cenocepacia [26, 27], characterizing genes regulated by Pvf in P. entomophila has broad implications for understanding bacterial virulence and host–pathogen interactions in many different species.

The pvf homologs in B. cenocepacia H111 are part of the seven-gene ham cluster and encode enzymes HamA, HamC, HamD, and HamE, which exhibit 39–53% sequence identity to PvfA-D. The ham cluster (hamA-G) is responsible for synthesizing the diazeniumdiolate-containing antifungal compound fragin [28]; however, homologs of hamB, hamF, and hamG are not present in P. entomophila and the signaling molecules produced by pvf-encoded enzymes in P. entomophila are unknown.

Previously, we characterized the protein- and small molecule-secretome altered by Pvf in P. entomophila L48 using proteomic and metabolomic analyses [24]. Deletion of pvfC affected the abundance of many secreted proteins and secondary metabolites, and these effects were reversed by genetic complementation with the full pvfA-D cluster [24]. We found that Pvf regulates the production of nearly 200 secreted and membrane proteins including toxins and proteins involved in type VI secretion, motility, and siderophore transport, as well as the production of secondary metabolites such as the insecticidal labradorins, the antimicrobial pyreudiones, and many uncharacterized small molecules [24]. Pvf has also been linked to the expression of select genes in P. entomophila [23, 24, 29]. The variety of phenotypes [23, 24, 30] influenced by the Pvf signaling system warrants a genome-wide transcriptomic analysis.

Here we present a global analysis of genes regulated by Pvf in P. entomophila L48. We used RNA sequencing to quantify differences in gene expression between wildtype (WT) and a pvf deletion mutant (ΔpvfA-D or KO) (Figure B in S1 Fig) [24]. We found that Pvf upregulates the expression of more than 200 genes involved in a wide variety of pathways implicated in virulence, metabolism, and physiology. Furthermore, out of around 60 biosynthetic genes upregulated by Pvf, we identified seven putative BGCs that could produce novel secondary metabolites. Combined, these broad transcriptional changes propose a role for Pvf as a global virulence regulator in P. entomophila and highlight avenues for further study.

Results

Pvf regulates the expression of 301 genes

We sequenced RNA from wildtype and ΔpvfA-D strains of P. entomophila L48 and performed differential gene expression analysis. RNA samples were extracted from 24-hour stationary-phase cultures, when Pvf signaling is active [24]. The signaling activity of each culture was validated using a β-galactosidase reporter assay that quantifies the expression of monalysin, a pore-forming toxin and virulence factor regulated by Pvf [24, 29]. Expression of monalysin in the reporter strain was induced by the supernatant of wildtype culture but not by the supernatant of ΔpvfA-D culture (S2 Fig), which confirms the signaling activity of the wildtype culture and the lack of signaling activity of the ΔpvfA-D culture. The β-galactosidase reporter strains were not used for RNA-sequencing. Instead, RNA samples were extracted from wildtype and ΔpvfA-D cultures grown on two separate days (“batches”) to account for day-to-day variability [31], and 12 total RNA samples [32] were sequenced. High-quality sequencing reads and good coverage of protein-coding genes were obtained (Table 1, S1 Table, and S1 Dataset), and principal component analysis shows a difference between experimental groups and similarity amongst replicates (S3 Fig).

[Figure omitted. See PDF.]

The genome of P. entomophila L48 contains 5,287 total annotated genes (5,134 protein coding genes, 78 tRNA genes, 22 rRNA genes, 10 miscellaneous RNA genes, 1 ncRNA gene, and 42 pseudogenes; S1 Dataset) of which 301 were differentially expressed between wildtype and ΔpvfA-D (p < 0.05 and fold change > 2; Fig 1). A total of 222 genes were upregulated and 79 were downregulated in wildtype compared with ΔpvfA-D, which demonstrates that Pvf affects about 6% of all genes. Most of these differentially expressed genes are protein-coding genes, although 4 pseudogenes, 1 rRNA-encoding gene, and the regulatory RNA PrrF1 were also differentially expressed. Gene functions were annotated based on information from NCBI RefSeq [33, 34], UniProt [35], the Pseudomonas Genome Database [36], and literature sources. The differentially expressed genes are likely regulated by Pvf and play a role in various processes, including branched chain amino acid biosynthesis, the type VI secretion system, siderophore transport, virulence, and biosynthesis of secondary metabolites (Fig 1).

[Figure omitted. See PDF.]

Each point represents an annotated gene in the L48 genome with gene order (PSEEN0001-PSEEN5559) on the x axis and the log₂ fold change of transcript abundance at 24 hrs in WT culture relative to ΔpvfA-D culture on the y axis. Differentially expressed genes are outlined in black (|log₂ fold change| > 1 and p adjusted < 0.05). Point fill color and shape represents function: virulence factors (red circles), type VI secretion system (orange squares), siderophore transport (green diamonds), valine, leucine, and isoleucine biosynthesis (cyan circles), pvf (lilac squares), and small molecule biosynthesis (pink diamonds). Gene clusters with type VI secretion system-related functions further discussed in the text are annotated as (a) P. entomophila type VI secretion system locus (Pent-T6SS) and (b-d) group 1 orphan vgrG-containing clusters: PSEEN0861-0864 (b), PSEEN3592-3595 (c), and PSEEN5271, 5272, 5274–5276 (d). Putative small molecule biosynthetic gene clusters discussed in the text are labeled with the same cluster number as in Fig 5: cluster 1 (PSEEN0131-0134), cluster 2 (PSEEN3042-3045; PSEEN3332; PSEEN3335), cluster 3 (PSEEN5520-5522), cluster 4 (PSEEN0621-0624), cluster 5 (PSEEN0961-0972), cluster 6 (PSEEN0973-0987), cluster 7 (PSEEN2014-2021), cluster 8 (PSEEN2139-2144, 2146–2156), cluster 9 (PSEEN2466-2482), and cluster 10 (PSEEN3319-3331).

Pvf differentially regulates branched chain amino acid biosynthetic genes

We observed different expression levels of genes required for the biosynthesis of the branched chain amino acids, l-valine, l-leucine, and l-isoleucine, between wildtype and ΔpvfA-D (Fig 2). The biosynthetic pathways of these amino acids are intertwined in bacteria and share four core enzymes—the acetolactate synthase IlvB/N, the keto acid reductoisomerase IlvC, the dihydroxyacid dehydratase IlvD, and the aminotransferase IlvE—which are sufficient to produce l-valine from two molecules of pyruvate [37]. The biosynthesis of l-isoleucine begins with an additional enzyme, the l-threonine dehydratase/deaminase IlvA, which converts l-threonine to 2-ketobutyrate [37]. A molecule of 2-ketobutyrate and a molecule of pyruvate are then processed by IlvB/N, IlvC, IlvD, and IlvE to generate l-isoleucine [37]. The biosynthesis of l-leucine parallels l-valine biosynthesis, then branches off by the action of the isopropylmalate synthase LeuA [37]. In wildtype compared with ΔpvfA-D, we observed higher expression (2.7-fold) of a gene shared in all three pathways–ilvB (PSEEN4350; Fig 2). Expression of leuA (PSEEN4399), only involved in l-leucine biosynthesis, was also higher (2.3-fold) in the wildtype (Fig 2). In contrast, expression of ilvA (PSEEN5013), the first gene in l-isoleucine biosynthesis from l-threonine, was lower (2.1-fold) in the wildtype than the mutant (Fig 2). These results suggest that Pvf upregulates l-valine and l-leucine biosynthesis but downregulates l-isoleucine biosynthesis.

[Figure omitted. See PDF.]

The general biosynthetic pathway for l-valine, l-leucine, and l-isoleucine in bacteria is shown with Escherichia coli genes (in parentheses) and their corresponding P. entomophila homologs. Boxes below genes represent fold change (FC) of transcript levels in WT relative to ΔpvfA-D, and FC values > 2 are shown. Transcripts expressed at higher levels in WT are depicted with increasingly dark shades of red, while transcripts expressed at higher levels in KO are depicted with increasingly dark shades of blue (bottom right). Negative fold change indicates transcripts expressed at higher levels in KO than WT and is calculated as fold change of KO relative to WT with a negative sign added. The same color scheme for fold change of transcript levels is used for all figures.

Pvf highly upregulates genes in the type VI secretion system

Type VI secretion system (T6SS) genes constitute the largest category of genes upregulated by Pvf (Fig 1). The T6SS is a nano syringe in Gram-negative bacteria that injects toxic effector proteins into target cells [38] (Fig 3E). A large variety of effectors can be delivered by the T6SS [39, 40], including antibacterial proteins for interbacterial competition [8, 41] and eukaryotic-targeting virulence factors [42]. Pathogens typically encode multiple T6SS loci, which contain all 13 core genes (tssA−M) [43]. However, P. entomophila contains only one locus (PSEEN0522−0537, 0539−0542; Pent-T6SS; Fig 3A), which was highly expressed in the wildtype compared to ΔpvfA-D (Fig 3A). For example, transcripts of tssM (PSEEN0535) and tssL (PSEEN0534) that encode membrane complex proteins were more abundant in the wildtype (3.1- and 3.4-fold, respectively), and transcripts of genes that encode wedge proteins (tssEFGK; PSEEN0525−0527 and PSEEN0533) were between 2.4 and 4.7-fold higher in the wildtype (Fig 3A). The membrane complex and wedge together orient the needle, which consists of a spike complex formed by the trimeric valine-glycine repeat protein (VgrG or TssI) and a hollow tube formed by hexameric rings of haemolysin co-regulated protein (Hcp or TssD). We observed 2.2-fold higher gene expression of the T6SS locus vgrG (PSEEN0540) and 7.6-fold higher expression of the T6SS locus hcp gene (PSEEN0539) in the wildtype (Fig 3A). A sheath composed of proteins encoded by tssB (PSEEN0523) and tssC (PSEEN0524) wraps the shaft and upon contraction launches the needle into the host cell. Transcript levels of tssB and tssC were 3.8-fold and 2.3-fold higher in the wildtype, respectively (Fig 3A). The wildtype also exhibited 2.9-fold higher expression of tssA (PSEEN0522), which encodes a cap protein that mediates polymerization of the Hcp tube and sheath, and 2.9-fold higher expression of tssH (PSEEN0528), which encodes an ATPase involved in recycling the contracted sheath (Fig 3A). Additionally, transcripts of three type VI secretion system regulatory genes were more abundant in the wildtype—PSEEN0529 (sigma factor activator, sfa, 3.1-fold higher), PSEEN0536 (phosphoprotein phosphatase stp1, 3.6-fold higher), and PSEEN0537 (serine-threonine protein kinase stk1, 2.1-fold higher; Fig 3A). Overall, higher transcript levels of all T6SS core components in the wildtype suggest significant upregulation of the Pent-T6SS by Pvf.

[Figure omitted. See PDF.]

Arrows represent gene length and direction of transcription and are color-coded based on function: tssA (brown), tssBC (light blue), tssEFGK (green), tssJLM (pink), hcp/tssD (light purple), vgrG/tssI (dark purple), RHS effector (dark blue), lipase gene (yellow), regulation (orange), unknown function (grey), and other function (tan). Fold changes > 2 (with p < 0.05) of transcript levels in WT relative to ΔpvfA-D are displayed above the corresponding gene. Fold change values ≥ 3 are bolded and values ≥ 4 are red. Fold changes and p values for all genes shown in the figure can be found in Tab 4 of S1 Dataset. (A) The only T6SS locus in P. entomophila, Pent-T6SS, which encodes essential structural proteins. (B-D) Orphan vgrG and hcp genes are grouped based on phylogeny (S4 Fig) as follows: (B) group 1, (C) group 2, and (D) other. (E) Model of T6SS adapted from Cherrak et al. [38]. Proteins are colored to match their corresponding genes (bottom left).

In addition to the vgrG and hcp genes in the T6SS locus, multiple additional “orphan” vgrG and hcp genes outside of the main T6SS locus (Fig 3B–3D) were expressed at a higher level in the wildtype than ΔpvfA-D. These orphan vgrG genes can be separated into 2 distinct groups based on sequence identity and neighboring genes [44, 45]. Group 1 vgrG genes share greater than 81% pairwise amino acid sequence identity with each other and with the PSEEN0540 vgrG in the main T6SS locus (S4 Fig). The high sequence similarity suggests that group 1 vgrGs are inparalogs [44], or copies, of the PSEEN0540 vgrG and might perform similar functions. Group 1 vgrG genes have an adjacent hcp gene and an adjacent RHS-repeat containing effector gene (Fig 3B). Furthermore, the C-terminus of all group 1 vgrG genes includes DUF2345, which has been linked to recruitment of effectors in E. coli [46] as well as antibacterial and antieukaryotic activity in Klebsiella pneumoniae [47]. Among the group 1 vgrG genes, we observed higher expression of PSEEN0862 and PSEEN3594 (3.0- and 15.7-fold, respectively) in the wildtype (Fig 3B). Transcripts of multiple genes that are adjacent to group 1 vgrG genes were also more abundant in the wildtype (Fig 3B), including the hcp genes PSEEN0861, PSEEN3595, and PSEEN5276 (8.2-, 22.9-, and 3.0-fold higher, respectively) and the RHS effector gene PSEEN3593 (3.3-fold higher). Four other highly expressed genes adjacent to group 1 vgrG genes (PSEEN0864, PSEEN3592, PSEEN5271, and PSEEN5272; 2.7-, 7.0-, 2.2-, and 2.1-fold higher in WT than ΔpvfA-D, respectively) are not annotated and may encode T6SS effectors or immunity proteins. Therefore, Pvf upregulates group 1 vgrGs and their associated genes.

The group 2 vgrG genes share greater than 62% pairwise amino acid sequence identity amongst themselves (S4 Fig). Instead of being adjacent to hcp and RHS effector genes, group 2 vgrGs are nearby genes that encode lipase effectors and uncharacterized proteins (Fig 3C). None of the group 2 vgrG genes or neighboring genes were differentially expressed between wildtype and ΔpvfA-D in our dataset (Fig 3C and S1 Dataset). Outside of group 1 and 2 vgrGs, the genome of P. entomophila contains two additional orphan vgrG genes, which exhibit relatively low sequence identity with each other and with the other vgrG genes (Fig 3D and S6 Fig). One of these vgrG genes, PSEEN3994, was expressed at a 2.2-fold higher level in the wildtype (Fig 3D). Therefore, Pvf differentially upregulates the expression of this orphan vgrG gene.

Pvf upregulates genes involved in siderophore transport

Several genes involved in siderophore transport were expressed at significantly different levels between wildtype and ΔpvfA-D (Fig 4), although no significant changes were observed in expression of siderophore biosynthetic genes (S1 Dataset). P. entomophila produces at least two siderophores, a pyoverdine and pseudomonine [48, 49]. Pseudomonine is structurally similar to acinetobactin from Acinetobacter baumannii [50], and likely uses a TonB-dependent import system that is homologous to bauABDCE in acinetobactin import [49, 51]. Wildtype exhibited higher gene expression levels than ΔpvfA-D of the bauB (PSEEN2493, 2.2-fold), bauC (PSEEN2495, 2.4-fold), bauD (PSEEN2496, 2.2-fold), and bauE (PSEEN2494, 2.7-fold) homologs, suggesting that Pvf upregulates ferric pseudomonine uptake (Fig 4). Furthermore, transcript levels of two ABC transporter-encoding genes (PSEEN2497 and PSEEN2498) that may be involved in pseudomonine export were 2.8-fold higher in the wildtype compared to ΔpvfA-D (Fig 4). We also observed 1.9-fold higher levels of a homolog of prrF1 in the wildtype (Fig 4), which encodes a small regulatory RNA required for iron homeostasis and virulence in P. aeruginosa [52, 53].

[Figure omitted. See PDF.]

Numerical values represent fold change of transcript levels in WT relative to ΔpvfA-D (KO). Only genes with p value < 0.05 are shown. Annotations adapted from NCBI and the Pseudomonas Genome Database.

Pvf upregulates additional virulence factor genes

Many genes that encode virulence factors such as proteases, lipases, chitinases, and toxins are differentially expressed between the wildtype and ΔpvfA-D (Fig 4). These virulence factors are used by bacteria to overcome insect defenses [54]. Expression of the gene encoding the virulence factor monalysin (mnl, PSEEN3174), which causes irreversible damage to the D. melanogaster gut epithelium [29], is significantly higher (32.1-fold; Fig 4) in the wildtype than ΔpvfA-D, making it the 6^th most Pvf-upregulated gene overall (Fig 1 and S1 Dataset). Additionally, transcripts of protease genes aprA (PSEEN1550) and pspB (PSEEN3027) were more abundant in the wildtype than ΔpvfA-D (2.1- and 2.3-fold, respectively; Fig 4). The aprA gene encodes a metalloprotease responsible for activating monalysin by cleaving the N-terminus of the inactive pre-monalysin [29], and further contributes to pathogenicity and persistence by degrading antimicrobial peptides secreted by the D. melanogaster gut epithelium [21]. Transcript abundance of the chitinase-encoding gene chiC (PSEEN2568) was significantly higher (3.5-fold) in the wildtype than ΔpvfA-D (Fig 4). Chitinase activity of chiC, a gene shared amongst highly insecticidal members of the Pseudomonas fluorescens group, has been linked to the oral insecticidal activity of Pseudomonas protegens CHA0 [55]. We also observed higher expression (2.9-fold) in the wildtype of a homolog of cbpD (chitin-binding protein D, PSEEN2567; Fig 4), which encodes a lytic polysaccharide monooxygenase. CbpD oxidizes chitin [56] and promotes survival of P. aeruginosa PA7 during systemic infection in humans [57]. A protective layer of chitin and glycoproteins coats the insect intestinal lumen [58], and the proteins encoded by chiC and cbpD could work together to break down this barrier. Finally, gacA (PSEEN3450), the response regulator in the GacS/A two-component system, was more highly expressed in wildtype than ΔpvfA-D (3.3-fold; Fig 4).

Pvf upregulates the expression of ten biosynthetic gene clusters

Transcripts for many secondary metabolite biosynthetic genes were more abundant in the wildtype than ΔpvfA-D, which suggests that Pvf upregulates their expression (Fig 5). As expected, expression of the pvf biosynthetic gene cluster (Fig 5, cluster 1) is diminished in the ΔpvfA-D strain. Two additional gene clusters that produce known secondary metabolites were also expressed at an elevated level in the wildtype compared to ΔpvfA-D. For example, all genes involved in entolysin biosynthesis, regulation, and transport (Fig 5, cluster 2) were expressed at significantly higher levels in the wildtype (> 4-fold), which indicates that Pvf controls entolysin production at the transcriptional level. Additionally, the genes responsible for hydrogen cyanide biosynthesis (hcnCBA; Fig 5, cluster 3) were expressed at levels 2.3 to 2.6-fold higher in the wildtype, which suggests that Pvf upregulates the production of hydrogen cyanide, which is also a virulence factor.

[Figure omitted. See PDF.]

Arrows represent gene length and direction of transcription and are color-coded based on function: NRPS (dark purple), PKS (pink), other enzyme (green), transport (aqua), regulation (orange), and unknown (light purple). Fold changes > 2 (with p < 0.05) of transcript levels in WT relative to ΔpvfA-D are displayed above the corresponding gene. Fold change values ≥ 3 are bolded and values ≥ 4 are red. Fold changes, p values, and putative enzyme functions for all genes shown in the figure can be found in Tab 7 of S1 Dataset.

We also identified seven putative biosynthetic gene clusters (BGCs) based on increased expression of constituent genes in the wildtype compared to ΔpvfA-D (Fig 5). For example, three of four genes in cluster 4 (PSEEN0621-0624) were expressed 2.9- to 3.5-fold higher in the wildtype than in ΔpvfA-D (Fig 5; S1 Dataset). A search of cluster 4 against the GenBank database using MultiGeneBlast [59] revealed homologous clusters in Yersinia and Klebsiella species, but not in other pseudomonads (S5 Fig). Genes in another putative BGC, cluster 5 (Fig 5; PSEEN0961−PSEEN0972), exhibited 2.2- to 4.7-fold higher expression in the wildtype (S1 Dataset). While other strains of Pseudomonas contain individual genes homologous to those in cluster 5, this cluster as a whole is not well conserved in pseudomonads (S6 Fig). Genes PSEEN0973−PSEEN0987 form another potential BGC (Fig 5, cluster 6) and these genes were expressed between 2.1- and 4-fold higher in the wildtype (S1 Dataset). Similar clusters to cluster 6 exist in the nematode symbiont and insect pathogen Xenorhabdus nematophila (S7 Fig). Cluster 7 (Fig 5; PSEEN2014−PSEEN2021) includes five genes with 2.2- to 3.1-fold higher expression in the wildtype (S1 Dataset). The closest relatives to cluster 7 exist in A. baumannii, and no homologous clusters were found in pseudomonads (S8 Fig).

Transcripts of most genes from cluster 8 (Fig 5; PSEEN2139−2144, 2146−2156) were at least 3-fold more abundant (and many > 4-fold) in the wildtype (S1 Dataset), including genes that encode three nonribosomal peptide synthetases (NRPSs; PSEEN2139, PSEEN2149, PSEEN2154) and one polyketide synthase (PKS; PSEEN2153) (2.3-, 3.1-, 3.7-, and 4.0-fold higher, respectively). NRPSs and PKSs are major enzyme families that synthesize secondary metabolites; NRPS/PKS hybrid clusters can synthesize peptide and polyketide hybrid products, such as lipopeptides [60]. Additional enzyme-encoding genes in the NRPS/PKS hybrid cluster 8 include a putative O-methyltransferase gene (PSEEN2141; 7.5-fold more highly expressed in the wildtype), the uncharacterized gene (PSEEN2146; 9.5-fold higher), and a putative ornithine cyclodeaminase gene (PSEEN2151; 6.4-fold higher; Fig 5 and S1 Dataset). Expression of an efflux pump gene (PSEEN2142), potentially involved in secondary metabolite export, was 7.8-fold higher in the wildtype than ΔpvfA-D (Fig 5 and S1 Dataset). These data suggest that Pvf upregulates the secondary metabolites produced by cluster 8-encoded enzymes. A search of cluster 8 against the GenBank database conducted with MultiGeneBlast [59] revealed only one homologous cluster in the phytopathogen Burkholderia gladioli BSR3 (S9 Fig).

Potential exopolysaccharide cluster 9 (Fig 5; PSEEN2466−PSEEN2482) contains several uncharacterized genes with higher transcript levels in wildtype (2.0- to 2.5-fold; S1 Dataset) and appears unique to P. entomophila (S10 Fig). Finally, we observed significantly higher expression in the wildtype (> 4-fold; S1 Dataset) of four genes in cluster 10 (Fig 5; PSEEN3319−PSEEN3331). A few strains of Pseudomonas contain sets of genes homologous to those in cluster 10, otherwise this cluster as a whole is not well conserved in pseudomonads (S11 Fig). Activities of the enzymes encoded in clusters 4 through 10 and structures of the secondary metabolites produced by these enzymes warrant further investigation. Pvf control of these seven proposed BGCs indicate the untapped biosynthetic potential of P. entomophila that may be explored using Pvf as a guide.

Discussion

Here we characterized the transcriptome of the Pvf signaling system in P. entomophila L48 by comparing gene expression between wildtype and a pvfA-D mutant. We found that deletion of pvfA-D controls the expression of genes in numerous pathways including branched chain amino acid biosynthesis, the type VI secretion system, siderophore transport, virulence, and biosynthesis of secondary metabolites. We did not further analyze genes that exhibited less than 2-fold difference in expression between wildtype and the pvfA-D mutant in our dataset, but expression of some of these genes could also be affected by Pvf.

Our transcriptomics suggest that Pvf upregulates l-valine (ilvBN) and l-leucine (ilvBN and leuA) biosynthesis and downregulates l-isoleucine biosynthesis (ilvA) in P. entomophila L48. Similarly, the ham cluster in B. cenocepacia H111 that contains pvf homologs upregulates an ilvD homologue and two leuA homologs, which also suggests upregulation of l-valine and l-leucine biosynthesis [28]. l-Valine and l-leucine could then serve as starting materials for pvf or ham-encoded enzymes. These amino acids are precursors to several small molecules including the dihydropyrazine N-oxides [27, 61] and leucinazole [62], which are synthesized by pvf-encoded enzymes in P. entomophila and P. fluorescens Pf0-1. l-Valine is the precursor to diazeniumdiolate-containing compounds valdiazen and fragin, which are produced by ham-encoded enzymes in Burkholderia cenocepacia [63].

The T6SS may play an important role in P. entomophila virulence, since unlike many other Gram-negative pathogens it does not encode a type III secretion system (T3SS). T6SS loci in Pseudomonas spp. cluster into six distinct phylogenetic groups that correlate with function [45]. The P. entomophila T6SS (Pent-T6SS) belongs to the Hcp secretion island two (HSI-II) cluster, which has been linked to virulence in Burkholderia thaliandensis [64] and plant and animal virulence in P. aeruginosa [65]. As the sole T6SS locus, the Pent-T6SS is a good model to investigate the role of the HSI-II type T6SS in infection and virulence. Co-regulation of vgrG genes with Pent-T6SS components suggests that the Pent-T6SS may be responsible for secreting the group 1 VgrG proteins and putative effector proteins encoded nearby. In contrast, group 2 vgrGs do not appear to be co-regulated with the Pent-T6SS, which suggests that group 2 vgrGs might perform a T6SS-independent function. Compared to other bacterial secretion systems, only a limited number of T6SS effectors and immunity proteins have been identified to date [66]. Here, we identified a number of putative T6SS effectors or immunity proteins highly upregulated by Pvf, many of which contain uncharacterized domains, like DUF2345. Our work paves the way for further studying the functions of these proteins.

Results from transcriptional analysis mostly agree with previous analysis of the secreted proteome and yielded new insights on Pvf regulation. For example, Pvf upregulates secretion of the virulence factors AprA and monalysin as well as VgrG spike proteins and RHS effector proteins encoded by the Pent-T6SS locus and group 1 vgrG clusters [24]. The corresponding genes also showed higher expression levels in the WT than the pvfA-D mutant. However, Pvf was linked to reduced swarming motility and decreased secretion of flagellar proteins by proteomic and phenotypic analysis [24], but differential expression of the corresponding genes was not captured in the transcriptomic analysis. This is not surprising because transcriptomic and proteomic datasets often only exhibit modest correlation due to differences in methodology and regulatory mechanisms (i.e. changes in protein abundance may result from post-transcriptional or -translational regulation or degradation) [67, 68]. New insights on Pvf regulation from transcription analysis include upregulation of branched chain amino acid metabolism and a putative hybrid sensory box/response regulator, the latter of which has a close ortholog in Pseudomonas putida (83% amino acid sequence identity) that has been linked to increased c-di-GMP levels, decreased motility, and increased biofilm formation [69, 70]. How Pvf regulation is linked to c-di-GMP signaling in P. entomophila will require further investigation. Altogether, transcriptomics and proteomics offered complementary insights on Pvf regulation.

Deletion of pvfA-D reduced the expression of a number of genes involved in pseudomonine transport, which could suggest a preference by Pvf for upregulating pseudomonine uptake over other iron-import systems. Interestingly, expression of genes involved in pseudomonine and pyoverdine biosynthesis do not appear significantly altered by Pvf. Perhaps, to conserve metabolic resources during infection, Pvf increases pseudomonine transport rather than production. Furthermore, P. entomophila has been shown to effectively import exogenous pyoverdine siderophores [49], so the pseudomonine uptake pathway may allow for import of structurally similar siderophores.

P. entomophila produces many different secondary metabolites, including hydrogen cyanide, the pyreudiones, and the labradorins [24]. Our data indicate that Pvf upregulates hydrogen cyanide on the transcriptional level. P. entomophila is one of a few known hydrogen cyanide producers, but Pvf had not been linked to regulation of this activity [71]. Hydrogen cyanide is known to be a virulence factor in Pseudomonas, including P. aeruginosa, where it is transcriptionally regulated by the las and rhl quorum sensing systems [72]. We did not observe statistically significant trends in transcript levels of pyreudione and labradorin biosynthetic genes in wildtype or ΔpvfA-D, although metabolomics results indicate that Pvf upregulates production of these metabolites [24]. It is possible that regulation of these metabolites occurs post-transcriptionally or that Pvf-induced transcriptional changes were not captured in this dataset. Furthermore, our previous metabolomics analysis revealed that the presence of Pvf increases the production of many uncharacterized metabolites [24]. Here, we identified seven putative BGCs that are transcriptionally upregulated by Pvf, which might correspond to increased production of these uncharacterized metabolites. In addition, many of these putative BGCs are unique to P. entomophila and encode uncharacterized proteins. Therefore, we anticipate that Pvf can be used to elicit production of novel secondary metabolites for structural elucidation and enable characterization of new enzymes involved in their biosynthesis.

Genes affected by Pvf exhibit some overlap with those affected by the GacS/A system, a two-component signal transduction system that is conserved in Gram-negative bacteria. In response to an unknown signal at high cell densities, Gac regulates numerous genes involved in virulence, quorum sensing, motility, stress tolerance, and biofilm formation [73, 74]. In pseudomonads, Gac regulates genes involved in virulence factor production, type II/III/VI secretion systems, siderophore biosynthesis and export, and small molecule biosynthesis [75, 76]. In P. entomophila, Gac has been found to upregulate the production of monalysin, the protease AprA, hydrogen cyanide, entolysin, and some T6SS components [22, 29]. Despite considerable overlap in target genes, Pvf and Gac regulatory mechanisms are proposed to operate independently [23]. However, our finding that Pvf upregulates gene expression of gacA raises the possibility of cross talk between the two regulatory systems.

In conclusion, the RNA-sequencing data indicate that Pvf controls the transcription of more than 300 genes. These findings demonstrate the broad regulatory effects of the Pvf signaling system on the physiology and virulence of P. entomophila L48. We identify many putative virulence factors under Pvf control whose functions are unknown and potentially novel. Further characterization of these virulence factors will contribute to understanding the mechanisms of virulence and host–pathogen interactions in P. entomophila and other pseudomonads. Furthermore, the large number of biosynthetic gene clusters upregulated by Pvf suggests that Pvf regulation may be used to identify novel small molecules produced by this bacterium. The receptor and signal transduction pathway of Pvf will be a focus of future studies.

Materials and methods

Source of chemicals and biological reagents

Chemical reagents were purchased from Sigma Aldrich or ThermoFisher Scientific unless otherwise stated.

Bacterial strains and culture conditions

All P. entomophila L48 strains were cultured in Lennox Luria Broth (low salt LB or LSLB) at 30°C with 225 rpm shaking, unless otherwise stated. The markerless pvf deletion mutant in P. entomophila, ΔpvfA-D, was constructed previously [24] and only includes the final 213 bp of pvfD (less than half of the original 468 bp). The previously constructed β-galactosidase assay reporter strains P. entomophila ΔpvfC::attTn7-Pmnl-lacZ and WT::attTn7-Pmnl-lacZ [26] contain a kanamycin resistance marker, and kanamycin was added to cultures of these strains to a final concentration of 50 μg mL^-1.

Cell culture and RNA extraction for RNA sequencing

Flasks of 25 mL cultures of P. entomophila L48 wildtype and ΔpvfA-D were inoculated with 50 μL of overnight culture and grown for 24 hr. Three biological replicates each of wildtype and ΔpvfA-D were grown. After 24 hr, 1 mL of each culture was removed for RNA extraction and the remaining 24 mL of culture was saved and tested for PVF autoinducer signaling activity (see below). For RNA extraction, cells were harvested by centrifugation for 30 min at 4°C, 3500 x g. RNA was extracted with TRIzol (ThermoFisher Scientific), as per the manufacturer’s protocol. DNase treatment was immediately performed with TURBO DNase (Invitrogen) as follows: 42 μL of nuclease-free water (NFW), 5 μL of TURBO DNase buffer, and 2 μL of TURBO DNase were added to each sample. Samples were incubated at 37°C for 30 min, then 1 μL of TURBO DNase was added, and samples were incubated for an additional 30 min at 37°C. RNA was purified with Mag-Bind TotalPure NGS beads (Omega Bio-tek) as per the manufacturer’s protocol. In brief: 90 μL of beads were added to each 50 μL DNase reaction (1.8X ratio). Samples were incubated for 5 min then transferred to a magnetic rack and incubated an additional 5–8 min. Samples were washed three times with 200 μL of 70% ethanol, then air dried for 5–8 min. RNA was eluted into 50 μL of nuclease-free water and stored at −80°C. A second batch of cultures was grown at a later time and RNA extraction repeated, resulting in a second batch of RNA samples. RNA quality was assessed using the Agilent BioAnalyzer2100, and all samples had a RIN > 8 and a 23S/16S ratio ≥ 1.2.

RNA sequencing and data analysis

A total of 12 RNA samples were submitted to BGI Genomics for sequencing (6 wildtype and 6 ΔpvfA-D). Vazyme Ribo-off rRNA removal kit was used for rRNA depletion. The BGI DNBSEQ-G400 platform was used for library construction and sequencing. Libraries were non-stranded and 100 bp paired-end reads were obtained. BGI removed adapters, contamination, and poor-quality reads; read statistics (S1 Table) represent read quality after the filtering by BGI. Reads were aligned to the Pseudomonas entomophila L48 genome using bbmap [77]. The reference genome was obtained from the NCBI RefSeq Database (accession number: GCF_000026105.1) [33] and the GTF annotation file was obtained from the Pseudomonas Genome Database [36]. The program featureCounts from the subread package was used to generate the count table [78]. Differential expression and statistical analysis were performed with DESeq2 [79]. The log₂ fold changes (WT/KO) were mapped onto biological pathways from the Kyoto Encyclopedia of Genes and Genomes (KEGG) Database [80] using the KEGGParser [81] application in Cytoscape [82]. Multiple sequence alignments were performed with Clustal Omega [83]. Gene cluster homology searches were conducted with MultiGeneBlast [59] against the 12/2015 version of the GenBank genome database.

PVF autoinducer signaling activity assay

The remaining 24 mL of each WT and ΔpvfA-D culture (after aliquots were removed for RNA extraction) was centrifuged for 40 min at 4°C, 4500 x g, then the supernatant transferred into a fresh tube and stored at 4°C. Next, the PVF autoinducer signaling activity of the WT and ΔpvfA-D culture supernatants was tested in a β-galactosidase reporter assay (β-gal assay) with the ΔpvfC::attTn7-Pmnl-lacZ and WT::attTn7-Pmnl-lacZ reporter strains [26]. A 1 mL sample of WT or ΔpvfA-D culture supernatant was added to a 3 mL LSLB culture containing kanamycin and then inoculated with 10 μL of an overnight culture of the ΔpvfC::attTn7-Pmnl-lacZ reporter strain. Two biological replicates were prepared from each supernatant sample. A 1 mL sample of LSLB was added as a control to the reporter strains, ΔpvfC::attTn7-Pmnl-lacZ and WT::attTn7-Pmnl-lacZ, instead of supernatant. Assay cultures were grown for 24 hr. At 24 hr, each β-gal assay culture was diluted 1:10 in 1 mL of fresh Z-buffer (60 mM Na₂HPO₄, 40 mM NaH₂PO₄, 10 mM KCl, 1 mM MgSO₄, 50 mM 2-mercaptoethanol, pH 7.0) and the OD₆₀₀ recorded. Samples were then further diluted in 1 mL of Z buffer as follows (two technical replicates were prepared from each sample)– 1:20 for the wildtype reporter strain control, 1:4 for the ΔpvfC reporter strain control, 1:20 for the ΔpvfC reporter strain complemented with wildtype supernatant, and 1:5 for the ΔpvfC reporter strain complemented with ΔpvfA-D supernatant.

The β-gal assay samples were then lysed with the addition of 100 μL chloroform and 50 μL 0.1% (w/v) SDS. Samples were briefly vortexed, then incubated for 5 min at room temperature. The reaction was started with the addition of 200 μL of fresh ortho-nitrophenyl-β-galactoside (ONPG) substrate (4 mg mL^-1 in Z-buffer), then incubated at 28°C with shaking at 600 rpm. As soon as the reaction developed a yellow color, the reaction was moved to the benchtop and 500 μL of stop solution (1 M Na₂CO₃) immediately added. Reaction start and stop time was recorded. Absorbance at 420 nm and 550 nm was recorded using a BMG Labtech CLARIOstar Plus microplate reader. Promoter activity in Miller units (MU) was calculated as follows:

A420 = absorbance at 420 nm (yellow)

A550 = absorbance at 550 nm (background scattering)

A600 = absorbance at 600 nm (cell density)

t = time of reaction (min)

v = volume of cells in reaction (mL)

Supporting information

S1 Fig. Composition and signaling activity of Pseudomonas virulence factor (pvf) in Pseudomonas entomophila L48.

(A) The pvf cluster contains four genes: pvfB (purple) encodes a non-heme diiron N-oxygenase, pvfC (blue) encodes a nonribosomal peptide synthetase, and pvfA (red) and pvfD (green) encode uncharacterized proteins. (B) pvf-encoded enzymes produce quorum sensing molecules (PVF autoinducers) that regulate the expression of many genes. Deletion of pvfA-D alters the expression of these genes.

https://doi.org/10.1371/journal.pone.0284907.s001

(TIF)

S2 Fig. Cultures used for RNA extraction exhibit the expected signaling activity.

Signaling activity was indicated by the β-galactosidase activity in Miller Units (MU) of the Pmnl-lacZ reporter strains, WT::Pmnl-lacZ (WTrep) and ΔpvfC::Pmnl-lacZ (ΔpvfCrep). Signaling activity of WTrep culture, ΔpvfCrep culture, and ΔpvfCrep cultures that were supplemented with the supernatant of cultures used for RNA extraction (left to right, WT1a-c, KO1a-c, WT2a-c, or KO2a-c).

https://doi.org/10.1371/journal.pone.0284907.s002

(TIF)

S3 Fig. Principal component analysis of RNA sequencing results.

(A) Principal component analysis (PCA) of rlog transformed gene count table for each RNA sample. Group one corresponds to the first batch of RNA samples listed in Table 1 and S1 Table and group two corresponds to the second batch. WT (group one green; group two purple) and ΔpvfA-D (KO; group one red; group two blue) datasets form separate clusters. (B) PCA results for group two samples only. WT samples (purple) and ΔpvfA-D (KO) samples (blue) cluster separately.

https://doi.org/10.1371/journal.pone.0284907.s003

(TIF)

S4 Fig. Sequence alignment of VgrG proteins in P. entomophila L48.

Sequence alignment was performed with Clustal Omega [83]. (A) Matrix of pairwise amino acid sequence identity between all VgrG proteins, shown as a percentage. Pairwise sequence identity comparisons within VgrG groups are color-coded for group 1 (blue), group 2 (pink), and other VgrGs (gold). Pairwise sequence identity comparisons between VgrG groups are color-coded for group 1 to group 2 (purple), group 1 to the VgrG of the Pent-T6SS locus (PSEEN0540, grey), group 1 to other VgrGs (green), group 2 to PSEEN0540 (grey), and group 2 to other VgrGs (orange). (B) Average pairwise amino acid sequence identity within and between VgrG groups. Color-coding is the same as (A). Error bars indicate standard deviation.

https://doi.org/10.1371/journal.pone.0284907.s004

(TIF)

S5 Fig. MultiGeneBlast results for unknown cluster 4 (PSEEN0621−PSEEN0624).

https://doi.org/10.1371/journal.pone.0284907.s005

(TIF)

S6 Fig. MultiGeneBlast results for unknown cluster 5 (PSEEN0961−PSEEN0972).

https://doi.org/10.1371/journal.pone.0284907.s006

(TIF)

S7 Fig. MultiGeneBlast results for unknown cluster 6 (PSEEN0973−PSEEN0987).

https://doi.org/10.1371/journal.pone.0284907.s007

(TIF)

S8 Fig. MultiGeneBlast results for unknown cluster 7 (PSEEN2014−PSEEN2021).

https://doi.org/10.1371/journal.pone.0284907.s008

(TIF)

S9 Fig. MultiGeneBlast results for unknown cluster 8 (PSEEN2139−2144, 2146–2156).

https://doi.org/10.1371/journal.pone.0284907.s009

(TIF)

S10 Fig. MultiGeneBlast results for unknown cluster 9 (PSEEN2466−PSEEN2482).

https://doi.org/10.1371/journal.pone.0284907.s010

(TIF)

S11 Fig. MultiGeneBlast results for unknown cluster 10 (PSEEN3319−PSEEN3331).

https://doi.org/10.1371/journal.pone.0284907.s011

(TIF)

S1 Table. Summary of RNA sequencing results for all samples.

https://doi.org/10.1371/journal.pone.0284907.s012

S1 Dataset. RNAseq complete results.

(Tab 1) RNA sequencing results for all annotated genes in the P. entomophila genome. (Tab 2) All differentially expressed genes (p < 0.05, FC (WT/KO) > 2). (Tab 3) Differentially expressed genes involved in branched chain amino acid biosynthesis. (Tab 4) Differentially expressed genes involved in the type VI secretion system. (Tab 5) Differentially expressed genes involved in siderophore transport. (Tab 6) Differentially expressed virulence factor genes. (Tab 7) Differentially expressed genes involved in small molecule biosynthesis. (Tab 8) Raw results from DeSeq2. (Tab 9) Raw count table from featureCounts.

https://doi.org/10.1371/journal.pone.0284907.s013

(XLSX)

Acknowledgments

The authors thank Qiang Guo (UNC), Drake Crawford (UNC), Ryan Mull (UNC), Isabelle Vallet-Gély (I2BC Université Paris-Saclay), and Tristan De Buysscher and Hemant Kelkar (UNC Center for Bioinformatics) for helpful discussions.

Citation: Acken KA, Li B (2023) Pseudomonas virulence factor controls expression of virulence genes in Pseudomonas entomophila. PLoS ONE 18(5): e0284907. https://doi.org/10.1371/journal.pone.0284907

About the Authors:

Katie A. Acken

Roles: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Validation, Visualization, Writing – original draft, Writing – review & editing

Affiliation: Department of Chemistry, The University of North Carolina at Chapel Hill, Chapel Hill, North Carolina, United States of America

Bo Li

Roles: Conceptualization, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Supervision, Writing – original draft, Writing – review & editing

E-mail: [email protected]

Affiliations: Department of Chemistry, The University of North Carolina at Chapel Hill, Chapel Hill, North Carolina, United States of America, Department of Microbiology and Immunology, The University of North Carolina at Chapel Hill, Chapel Hill, North Carolina, United States of America

ORICD: https://orcid.org/0000-0002-8019-8891

References

1. Papenfort K, Bassler BL. Quorum sensing signal-response systems in Gram-negative bacteria. Nat Rev Microbiol. 2016;14(9):576–88. pmid:27510864

2. Ng WL, Bassler BL. Bacterial quorum-sensing network architectures. Annu Rev Genet. 2009;43:197–222. pmid:19686078

3. Sakuragi Y, Kolter R. Quorum-sensing regulation of the biofilm matrix genes (pel) of Pseudomonas aeruginosa. J Bacteriol. 2007;189(14):5383–6.

4. De Kievit TR, Gillis R, Marx S, Brown C, Iglewski BH. Quorum-sensing genes in Pseudomonas aeruginosa biofilms: their role and expression patterns. Appl Environ Microbiol. 2001;67:1865–73.

5. Gross H, Loper JE. Genomics of secondary metabolite production by Pseudomonas spp. Nat Prod Rep. 2009;26(11):1408–46.

6. Strateva T, Mitov I. Contribution of an arsenal of virulence factors to pathogenesis of Pseudomonas aeruginosa infections. Ann Microbiol. 2011;61(4):717–32.

7. Jesaitis AJ, Franklin MJ, Berglund D, Sasaki M, Lord CI, Bleazard JB, et al. Compromised host defense on Pseudomonas aeruginosa biofilms: characterization of neutrophil and biofilm interactions. J Immunol. 2003;171(8):4329–39.

8. Perault AI, Chandler CE, Rasko DA, Ernst RK, Wolfgang MC, Cotter PA. Host adaptation predisposes Pseudomonas aeruginosa to type VI secretion system-mediated predation by the Burkholderia cepacia complex. Cell Host Microbe. 2020;28(4):534–47.

9. Rutherford ST, Bassler BL. Bacterial quorum sensing: its role in virulence and possibilities for its control. Cold Spring Harbor Perspect Med. 2012;2(11):a012427. pmid:23125205

10. Hentzer M, Wu H, Andersen JB, Riedel K, Rasmussen TB, Bagge N, et al. Attenuation of Pseudomonas aeruginosa virulence by quorum sensing inhibitors. EMBO J. 2003;15:3803–15.

11. Johnson BK, Abramovitch RB. Small molecules that sabotage bacterial virulence. Trends Pharmacol Sci. 2017;38(4):339–62. pmid:28209403

12. Dickey SW, Cheung GYC, Otto M. Different drugs for bad bugs: antivirulence strategies in the age of antibiotic resistance. Nat Rev Drug Discov. 2017;16(7):457–71. pmid:28337021

13. Jimenez PN, Koch G, Thompson JA, Xavier KB, Cool RH, Quax WJ. The multiple signaling systems regulating virulence in Pseudomonas aeruginosa. Microbiol Mol Biol Rev. 2012;76(1):46–65.

14. Chugani S, Kim BS, Phattarasukol S, Brittnacher MJ, Choi SH, Harwood CS, et al. Strain-dependent diversity in the Pseudomonas aeruginosa quorum-sensing regulon. Proc Natl Acad Sci U S A. 2012;109(41):E2823–31.

15. Kostylev M, Kim DY, Smalley NE, Salukhe I, Greenberg EP, Dandekar AA. Evolution of the Pseudomonas aeruginosa quorum-sensing hierarchy. Proc Natl Acad Sci U S A. 2019;116(14):7027–32.

16. Webb K, Fogarty A, Barrett DA, Nash EF, Whitehouse JL, Smyth AR, et al. Clinical significance of Pseudomonas aeruginosa 2-alkyl-4-quinolone quorum-sensing signal molecules for long-term outcomes in adults with cystic fibrosis. J Med Microbiol. 2019;68(12):1823–8.

17. Garcia-Reyes S, Soberon-Chavez G, Cocotl-Yanez M. The third quorum-sensing system of Pseudomonas aeruginosa: Pseudomonas quinolone signal and the enigmatic PqsE protein. J Med Microbiol. 2020;69(1):25–34.

18. Rampioni G, Falcone M, Heeb S, Frangipani E, Fletcher MP, Dubern JF, et al. Unravelling the genome-wide contributions of specific 2-alkyl-4-quinolones and PqsE to quorum sensing in Pseudomonas aeruginosa. PLoS Pathog. 2016;12(11):e1006029.

19. Tobias NJ, Brehm J, Kresovic D, Brameyer S, Bode HB, Heermann R. New vocabulary for bacterial communication. ChemBioChem. 2020;21(6):759–68. pmid:31709676

20. Vodovar N, Vinals M, Liehl P, Basset A, Degrouard J, Spellman P, et al. Drosophila host defense after oral infection by an entomopathogenic Pseudomonas species. Proc Natl Acad Sci U S A. 2005;102(32):11414–9.

21. Liehl P, Blight M, Vodovar N, Boccard F, Lemaitre B. Prevalence of local immune response against oral infection in a Drosophila/Pseudomonas infection model. PLoS Pathog. 2006;2(6):e56.

22. Dieppois G, Opota O, Lalucat J, Lemaitre B. Pseudomonas entomophila: a versatile bacterium with entomopathogenic properties. In: Ramos J-L, Goldberg JB, Filloux A, editors. Pseudomonas: Volume 7: New Aspects of Pseudomonas Biology: Springer, Dordrecht, The Netherlands; 2015. p. 25–49.

23. Vallet-Gely I, Opota O, Boniface A, Novikov A, Lemaitre B. A secondary metabolite acting as a signalling molecule controls Pseudomonas entomophila virulence. Cell Microbiol. 2010;12(11):1666–79.

24. Kretsch AM, Morgan GL, Acken KA, Barr SA, Li B. Pseudomonas virulence factor pathway synthesizes autoinducers that regulate the secretome of a pathogen. ACS Chem Biol. 2021;16(3):501–9. pmid:33595276

25. Kuznetsov D, Tegenfeldt F, Manni M, Seppey M, Berkeley M, Kriventseva EV, et al. OrthoDB v11: annotation of orthologs in the widest sampling of organismal diversity. Nucleic Acids Res. 2023;51(D1):D445–D51. pmid:36350662

26. Morgan GL, Kretsch AM, Santa Maria KC, Weeks SJ, Li B. Specificity of nonribosomal peptide synthetases in the biosynthesis of the Pseudomonas virulence factor. Biochemistry. 2019;58(52):5249–54.

27. Morgan GL, Li B. In Vitro reconstitution reveals a central role for the N-oxygenase PvfB in (dihydro)pyrazine-N-oxide and valdiazen biosynthesis. Angew Chem, Int Ed. 2020;59(48):21387–91. pmid:32662921

28. Jenul C, Sieber S, Daeppen C, Mathew A, Lardi M, Pessi G, et al. Biosynthesis of fragin is controlled by a novel quorum sensing signal. Nat Commun. 2018;9(1):1297. pmid:29602945

29. Opota O, Vallet-Gely I, Vincentelli R, Kellenberger C, Lacovache I, Gonzalez MR, et al. Monalysin, a novel β-pore-forming toxin from the Drosophila pathogen Pseudomonas entomophila, contributes to host intestinal damage and lethality. PLoS Pathog. 2011;7(9):e1002259.

30. Vallet-Gely I, Novikov A, Augusto L, Liehl P, Bolbach G, Pechy-Tarr M, et al. Association of hemolytic activity of Pseudomonas entomophila, a versatile soil bacterium, with cyclic lipopeptide production. Appl Environ Microbiol. 2010;76(3):910–21.

31. Manga P, Klingeman DM, Lu TY, Mehlhorn TL, Pelletier DA, Hauser LJ, et al. Replicates, read numbers, and other important experimental design considerations for microbial RNA-seq identified using Bacillus thuringiensis datasets. Front Microbiol. 2016;7:794.

32. Schurch NJ, Schofield P, Gierlinski M, Cole C, Sherstnev A, Singh V, et al. How many biological replicates are needed in an RNA-seq experiment and which differential expression tool should you use? RNA. 2016;22(6):839–51. pmid:27022035

33. O’Leary NA, Wright MW, Brister JR, Ciufo S, Haddad D, McVeigh R, et al. Reference sequence (RefSeq) database at NCBI: current status, taxonomic expansion, and functional annotation. Nucleic Acids Res. 2016;44(D1):D733–D45. pmid:26553804

34. Li W, O’Neill KR, Haft DH, DiCuccio M, Chetvernin V, Badretdin A, et al. RefSeq: expanding the Prokaryotic Genome Annotation Pipeline reach with protein family model curation. Nucleic Acids Res. 2021;49(D1):D1020–D8. pmid:33270901

35. UniProt C. UniProt: the universal protein knowledgebase in 2021. Nucleic Acids Res. 2021;49(D1):D480–D9. pmid:33237286

36. Winsor GL, Griffiths EJ, Lo R, Dhillon BK, Shay JA, Brinkman FSL. Enhanced annotations and features for comparing thousands of Pseudomonas genomes in the Pseudomonas genome database. Nucleic Acids Res. 2016.

37. Amorim Franco TM, Blanchard JS. Bacterial branched-chain amino acid biosynthesis: structures, mechanisms, and drugability. Biochemistry. 2017;56(44):5849–65. pmid:28977745

38. Cherrak Y, Flaugnatti N, Durand E, Journet L, Cascales E. Structure and activity of the type VI secretion system. Microbiol Spectrum. 2019;7(4):PSIB-0031-2019. pmid:31298206

39. Russell AB, Peterson SB, Mougous JD. Type VI secretion system effectors: poisons with a purpose. Nat Rev Microbiol. 2014;12(2):137–48. pmid:24384601

40. Diniz JA, Liu YC, Coulthurst SJ. Molecular weaponry: diverse effectors delivered by the type VI secretion system. Cell Microbiol. 2015;17(12):1742–51. pmid:26432982

41. Hood RD, Singh P, Hsu F, Guvener T, Carl MA, Trinidad RR, et al. A type VI secretion system of Pseudomonas aeruginosa targets a toxin to bacteria. Cell Host Microbe. 2010;7(1):25–37.

42. Jiang F, Waterfield NR, Yang J, Yang G, Jin Q. A Pseudomonas aeruginosa type VI secretion phospholipase D effector targets both prokaryotic and eukaryotic cells. Cell Host Microbe. 2014;15(5):600–10.

43. Boyer F, Fichant G, Berthod J, Vandenbrouck Y, Attree I. Dissecting the bacterial type VI secretion system by a genome wide in silico analysis: what can be learned from available microbial genomic resources? BMC Genomics. 2009;10:104.

44. Sarris PF, Scoulica EV. Pseudomonas entomophila and Pseudomonas mendocina: potential models for studying the bacterial type VI secretion system. Infect, Genet Evol. 2011;11(6):1352–60.

45. Barret M, Egan F, Fargier E, Morrissey JP, O’Gara F. Genomic analysis of the type VI secretion systems in Pseudomonas spp.: novel clusters and putative effectors uncovered. Microbiology. 2011;157(Pt 6):1726–39.

46. Flaugnatti N, Le TT, Canaan S, Aschtgen MS, Nguyen VS, Blangy S, et al. A phospholipase A1 antibacterial type VI secretion effector interacts directly with the C-terminal domain of the VgrG spike protein for delivery. Mol Microbiol. 2016;99(6):1099–118.

47. Storey D, McNally A, Astrand M, Sa-Pessoa Graca Santos J, Rodriguez-Escudero I, Elmore B, et al. Klebsiella pneumoniae type VI secretion system-mediated microbial competition is PhoPQ controlled and reactive oxygen species dependent. PLoS Pathog. 2020;16(3):e1007969.

48. Vodovar N, Vallenet D, Cruveiller S, Rouy Z, Barbe V, Acosta C, et al. Complete genome sequence of the entomopathogenic and metabolically versatile soil bacterium Pseudomonas entomophila. Nat Biotechnol. 2006;24(6):673–9.

49. Matthijs S, Laus G, Meyer JM, Abbaspour-Tehrani K, Schafer M, Budzikiewicz H, et al. Siderophore-mediated iron acquisition in the entomopathogenic bacterium Pseudomonas entomophila L48 and its close relative Pseudomonas putida KT2440. Biometals. 2009;22(6):951–64.

50. Wuest WM, Sattely ES, Walsh CT. Three siderophores from one bacterial enzymatic assembly line. J Am Chem Soc. 2009;131:5056–7. pmid:19320483

51. Cornelis P, Bodilis J. A survey of TonB-dependent receptors in fluorescent pseudomonads. Environ Microbiol Rep. 2009;1(4):256–62. pmid:23765855

52. Reinhart AA, Powell DA, Nguyen AT, O’Neill M, Djapgne L, Wilks A, et al. The prrF-encoded small regulatory RNAs are required for iron homeostasis and virulence of Pseudomonas aeruginosa. Infect Immun. 2015;83(3):863–75.

53. Djapgne L, Panja S, Brewer LK, Gans JH, Kane MA, Woodson SA, et al. The Pseudomonas aeruginosa PrrF1 and PrrF2 small regulatory RNAs promote 2-alkyl-4-quinolone production through redundant regulation of the antR mRNA. J Bacteriol. 2018;200(10):e00704–17.

54. Vallet-Gely I, Lemaitre B, Boccard F. Bacterial strategies to overcome insect defences. Nat Rev Microbiol. 2008;6(4):302–13. pmid:18327270

55. Flury P, Aellen N, Ruffner B, Pechy-Tarr M, Fataar S, Metla Z, et al. Insect pathogenicity in plant-beneficial pseudomonads: phylogenetic distribution and comparative genomics. ISME J. 2016;10(10):2527–42. pmid:26894448

56. Forsberg Z, Sorlie M, Petrovic D, Courtade G, Aachmann FL, Vaaje-Kolstad G, et al. Polysaccharide degradation by lytic polysaccharide monooxygenases. Curr Opin Struct Biol. 2019;59:54–64. pmid:30947104

57. Askarian F, Uchiyama S, Masson H, Sorensen HV, Golten O, Bunaes AC, et al. The lytic polysaccharide monooxygenase CbpD promotes Pseudomonas aeruginosa virulence in systemic infection. Nat Commun. 2021;12(1):1230.

58. Kuraishi T, Binggeli O, Opota O, Buchon N, Lemaitre B. Genetic evidence for a protective role of the peritrophic matrix against intestinal bacterial infection in Drosophila melanogaster. Proc Natl Acad Sci U S A. 2011;108(38):15966–71.

59. Medema MH, Takano E, Breitling R. Detecting sequence homology at the gene cluster level with MultiGeneBlast. Mol Biol Evol. 2013;30(5):1218–23. pmid:23412913

60. Fisch KM. Biosynthesis of natural products by microbial iterative hybrid PKS–NRPS. RSC Adv. 2013;3(40):18228.

61. Kretsch AM, Morgan GL, Tyrrell J, Mevers E, Vallet-Gely I, Li B. Discovery of (dihydro)pyrazine N-oxides via genome mining in Pseudomonas. Org Lett. 2018;20(16):4791–5.

62. Morgan GL, Li K, Crawford DM, Aube J, Li B. Enzymatic synthesis of diverse heterocycles by a noncanonical nonribosomal peptide synthetase. ACS Chem Biol. 2021;16(12):2776–86. pmid:34767712

63. Sieber S, Daeppen C, Jenul C, Mannancherril V, Eberl L, Gademann K. Biosynthesis and structure-activity relationship investigations of the diazeniumdiolate antifungal agent fragin. ChemBioChem. 2020;21(11):1587–92. pmid:31945256

64. Schwarz S, West TE, Boyer F, Chiang WC, Carl MA, Hood RD, et al. Burkholderia type VI secretion systems have distinct roles in eukaryotic and bacterial cell interactions. PLoS Pathog. 2010;6(8):e1001068. pmid:20865170

65. Lesic B, Starkey M, He J, Hazan R, Rahme LG. Quorum sensing differentially regulates Pseudomonas aeruginosa type VI secretion locus I and homologous loci II and III, which are required for pathogenesis. Microbiology. 2009;155(9):2845–55.

66. An Y, Wang J, Li C, Leier A, Marquez-Lago T, Wilksch J, et al. Comprehensive assessment and performance improvement of effector protein predictors for bacterial secretion systems III, IV and VI. Briefings Bioinf. 2018;19(1):148–61. pmid:27777222

67. Maier T, Guell M, Serrano L. Correlation of mRNA and protein in complex biological samples. FEBS Lett. 2009;583(24):3966–73. pmid:19850042

68. Vogel C, Marcotte EM. Insights into the regulation of protein abundance from proteomic and transcriptomic analyses. Nat Rev Genet. 2012;13(4):227–32. pmid:22411467

69. Matilla MA, Travieso ML, Ramos JL, Ramos-Gonzalez MI. Cyclic diguanylate turnover mediated by the sole GGDEF/EAL response regulator in Pseudomonas putida: its role in the rhizosphere and an analysis of its target processes. Environ Microbiol. 2011;13(7):1745–66.

70. Nie H, Xiao Y, He J, Liu H, Nie L, Chen W, et al. Phenotypic-genotypic analysis of GGDEF/EAL/HD-GYP domain-encoding genes in Pseudomonas putida. Environ Microbiol Rep. 2020;12(1):38–48.

71. Ryall B, Mitchell H, Mossialos D, Williams HD. Cyanogenesis by the entomopathogenic bacterium Pseudomonas entomophila. Lett Appl Microbiol. 2009;49(1):131–5.

72. Pessi G, Haas D. Transcriptional control of the hydrogen cyanide biosynthetic genes hcnABC by the anaerobic regulator ANR and the quorum-sensing regulators LasR and RhlR in Pseudomonas aeruginosa. J Bacteriol. 2000;182(24):6940–9.

73. Ferreiro MD, Gallegos MT. Distinctive features of the Gac-Rsm pathway in plant-associated Pseudomonas. Environ Microbiol. 2021;23(10):5670–89.

74. Lapouge K, Schubert M, Allain FH, Haas D. Gac/Rsm signal transduction pathway of γ-proteobacteria: from RNA recognition to regulation of social behaviour. Mol Microbiol. 2008;67(2):241–53.

75. Cheng X, de Bruijn I, van der Voort M, Loper JE, Raaijmakers JM. The Gac regulon of Pseudomonas fluorescens SBW25. Environ Microbiol Rep. 2013;5(4):608–19.

76. Hassan KA, Johnson A, Shaffer BT, Ren Q, Kidarsa TA, Elbourne LD, et al. Inactivation of the GacA response regulator in Pseudomonas fluorescens Pf-5 has far-reaching transcriptomic consequences. Environ Microbiol. 2010;12(4):899–915.

77. Bushnell B. BBMap: a fast, accurate, splice-aware aligner 2014 [Available from: https://sourceforge.net/projects/bbmap/.

78. Liao Y, Smyth GK, Shi W. FeatureCounts: an efficient general purpose program for assigning sequence reads to genomic features. Bioinformatics. 2014;30(7):923–30. pmid:24227677

79. Love MI, Huber W, Anders S. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biol. 2014;15(12):550. pmid:25516281

80. Kanehisa M, Goto S. KEGG: Kyoto encyclopedia of genes and genomes. Nucleic Acids Res. 2000;28(1):27–30. pmid:10592173

81. Nersisyan L, Samsonyan R, Arakelyan A. CyKEGGParser: tailoring KEGG pathways to fit into systems biology analysis workflows. F1000Research. 2014;3:145. pmid:25383185

82. Shannon P, Markiel A, Ozier O, Baliga NS, Wang JT, Ramage D, et al. Cytoscape: a software environment for integrated models of biomolecular interaction networks. Genome Res. 2003;13(11):2498–504. pmid:14597658

83. Madeira F, Park YM, Lee J, Buso N, Gur T, Madhusoodanan N, et al. The EMBL-EBI search and sequence analysis tools APIs in 2019. Nucleic Acids Res. 2019;47(W1):W636–W41. pmid:30976793