Vibrioferrin production by the food spoilage bacterium Pseudomonas fragi

Vibrioferrin production by the food spoilage bacterium Pseudomonas fragi Abstract Pseudomonas fragi is a meat and milk spoilage bacterium with high iron requirements; however, mechanisms of iron acquisition remain largely unknown. The aim of this work was to investigate siderophore production as an iron acquisition system for P. fragi. A vibrioferrin siderophore gene cluster was identified in 13 P. fragi, and experiments were conducted with a representative strain of this group (F1801). Chromeazurol S assays showed that P. fragi F1801 produced siderophores under iron starvation at optimum growth and refrigeration temperature. Conversely, supplementation of low iron media with 50 μM FeCl3 repressed transcription of the vibrioferrin genes and siderophore production. Disruption of the siderophore receptor (pvuA) caused polar effects on downstream vibrioferrin genes, resulting in impaired siderophore production of the ΔpvuA mutant. Growth of this mutant was compared to growth of a control strain (Δlip) with wild-type vibrioferrin genes in low iron media supplemented with iron chelators 2,2΄-bipyridyl or apo-transferrin. While 25 μM 2,2΄-bipyridyl caused impaired growth of ΔpvuA, growth of the mutant was completely inhibited by 2.5 μM apo-transferrin, but could be restored by FeCl3 addition. In summary, this work identifies a vibrioferrin-mediated iron acquisition system of P. fragi, which is required for growth of this bacterium under iron starvation. Pseudomonas fragi, siderophore, vibrioferrin, food spoilage, iron acquisition, iron chelators INTRODUCTION Iron is required for fundamental cellular processes such as respiration and the synthesis of amino acids and DNA, and is therefore an essential element for the majority of microorganisms (Ilbert and Bonnefoy 2013). However, iron is in short supply in many habitats. For example, in aerobic environments at neutral pH, Fe2+ is rapidly oxidised to Fe3+, leading to the formation of insoluble hydroxide salts and rendering iron scarce in the environment (Colombo et al.2014). Similarly, in the human host and foods such as milk and meat, iron is sequestered by iron-binding proteins including transferrin family proteins, transferrin and lactoferrin, hemeproteins myoglobin and haemoglobin, and ferritin (Skaar 2010). Microorganisms have evolved a variety of mechanisms to obtain their iron. In iron-rich environments, iron acquisition occurs through energy-independent, low-affinity iron uptake systems (Jones and Niederweis 2010). In iron-limited environments, a common strategy involves the excretion of small molecular weight iron-chelating compounds termed siderophores (Noinaj et al.2010). Fe3+-siderophore complexes that form outside the cell are taken up in Gram-negative bacteria by specific outer membrane receptors, which are supplied with energy for transport from the cytoplasmic membrane by the Ton system (Faraldo-Gomez and Sansom 2003). Excess iron causes toxic cellular effects, necessitating the tight regulation of iron uptake in microorganisms. The ferric uptake regulator protein (Fur) regulates iron homoeostasis at the transcription level in many Gram-negative and some Gram-positive organisms (Carpenter, Whitmire and Merrell 2009). When intracellular Fe2+ concentrations exceed a certain threshold level, Fur binds to specific regulatory DNA sequences termed Fur boxes, preventing RNA polymerase binding to promotors (Troxell and Hassan 2013). In contrast, scarcity of intracellular iron causes Fur to lose its ability to bind to Fur boxes, resulting in gene transcription. Thus, the transcription of iron uptake genes is ultimately regulated by the concentration of intracellular iron. Pseudomonas have some of the best studied siderophore-mediated iron uptake systems such as the pyoverdine siderophores of the fluorescent pseudomonads (Meyer and Abdallah 1978; Cox and Adams 1985; Cezard, Farvacques and Sonnet 2015; Trapet et al.2016). In contrast, Pseudomonas fragi has been considered a non-siderophore producing member of this genus and is described as not producing siderophores in detectable amounts in Bergey's Manual (Champomier-Vergès, Stintzi and Meyer 1996; Garrity, Bell and Lilburn 2005). Pseudomonas fragi is a problematic meat and milk spoilage bacterium with high iron requirements, but beyond its ability to scavenge foreign Fe3+-siderophore complexes, very little is known about the strategies this bacterium employs to obtain iron (Champomier-Vergès, Stintzi and Meyer 1996; De Jonghe et al.2011; Casaburi et al.2015). Interestingly, in a recent study on plant growth-promoting bacteria, a putative P. fragi isolate derived from the rhizosphere was shown to produce siderophores (Farh et al.2017), suggesting previous conclusions that this bacterium is a non-siderophore producer may not have been accurate. In this work, siderophore production was investigated as an iron acquisition system for P. fragi. The genomes of 13 P. fragi were examined for the presence of siderophore biosynthetic gene clusters, which led to the identification of vibrioferrin biosynthesis and vibrioferrin-mediated iron acquisition genes. This work describes experiments that determine siderophore production for P. fragi F1801 and show a role for the siderophore in iron metabolism of this bacterium under low iron conditions. MATERIALS AND METHODS Strains, plasmids and growth conditions Strains, plasmids, and genome and gene sequences used in this study are included in Table 1. Table 1. Strains, plasmids, and genome and gene sequences used in this study. Strains, plasmids, and genome and gene sequences  Relevant characteristics  Locus tags of vibrioferrin genes  Accession numbersa  Pseudomonas fragi genome sequences  F1786b    CJU73_02415–CJU73_02445  GCA_002269585.1  F1791b    CJU79_02430–CJU79_02400  GCA_002269515.1  F1792b    CJU72_13900–CJU72_13930  GCA_002269595.1  F1793b    CJU75_07595–CJU75_07565  GCA_002269565.1  F1794b    CJU80_05030–CJU80_05000  GCA_002269445.1  F1813b    CJU77_09910–CJU77_09940  GCA_002269465.1  F1815b    CJU76_02305–CJU76_02275  GCA_002269545.1  F1816b    CJU74_06845–CJU76_06815  GCA_002269485.1  F1818b    CJU78_20565–CJU78_20595  GCA_002269625.1  F1820b    CJF43_16240–CJF43_16270  GCA_002269055.1  F1821b    CJF37_06830–CJF37_06800  GCA_002269155.1  ATCC 4973b    SAMN05216594_0902–SAMN05216594_0896  GCA_900105835.1  F1801b    CJU81_14170–CJU81_14200  GCA_002269505.1  Pseudomonas fragi strains  F1801c  Wild-type    SRX3235903  F1801Δlipc  Disruption of CJU81_12870b at 167th amino acid encoding codon, Kmr    SRX3235904  F1801ΔpvuAc  Disruption of CJU81_14195b at 419th amino acid encoding codon, Kmr    SRX3235905  Escherichia coli strains  Escherichia coli S17–1 λpir  thi, pro, hsd (r− m+) recA::RP4–2-Tcr::Mu Kmr::Tn7 Tpr Smrλpir      Genome/gene sequences of other bacteria  Azotobacter vinelandii CA      GCA_000380335.1  Xanthomonas campestris 8004      GCA_000012105.1  Vibrio parahaemolyticus WP1      AB048250.2 and AB082123.1  Plasmids  pJP5603  Suicide plasmid, Kmr      pJP5603_lip323–783  pJP5603::disruption construct for gene CJU81_12870c, Kmr      pJP5603_pvuA886–1523  pJP5603::disruption construct for gene CJU81_14195c, Kmr      Strains, plasmids, and genome and gene sequences  Relevant characteristics  Locus tags of vibrioferrin genes  Accession numbersa  Pseudomonas fragi genome sequences  F1786b    CJU73_02415–CJU73_02445  GCA_002269585.1  F1791b    CJU79_02430–CJU79_02400  GCA_002269515.1  F1792b    CJU72_13900–CJU72_13930  GCA_002269595.1  F1793b    CJU75_07595–CJU75_07565  GCA_002269565.1  F1794b    CJU80_05030–CJU80_05000  GCA_002269445.1  F1813b    CJU77_09910–CJU77_09940  GCA_002269465.1  F1815b    CJU76_02305–CJU76_02275  GCA_002269545.1  F1816b    CJU74_06845–CJU76_06815  GCA_002269485.1  F1818b    CJU78_20565–CJU78_20595  GCA_002269625.1  F1820b    CJF43_16240–CJF43_16270  GCA_002269055.1  F1821b    CJF37_06830–CJF37_06800  GCA_002269155.1  ATCC 4973b    SAMN05216594_0902–SAMN05216594_0896  GCA_900105835.1  F1801b    CJU81_14170–CJU81_14200  GCA_002269505.1  Pseudomonas fragi strains  F1801c  Wild-type    SRX3235903  F1801Δlipc  Disruption of CJU81_12870b at 167th amino acid encoding codon, Kmr    SRX3235904  F1801ΔpvuAc  Disruption of CJU81_14195b at 419th amino acid encoding codon, Kmr    SRX3235905  Escherichia coli strains  Escherichia coli S17–1 λpir  thi, pro, hsd (r− m+) recA::RP4–2-Tcr::Mu Kmr::Tn7 Tpr Smrλpir      Genome/gene sequences of other bacteria  Azotobacter vinelandii CA      GCA_000380335.1  Xanthomonas campestris 8004      GCA_000012105.1  Vibrio parahaemolyticus WP1      AB048250.2 and AB082123.1  Plasmids  pJP5603  Suicide plasmid, Kmr      pJP5603_lip323–783  pJP5603::disruption construct for gene CJU81_12870c, Kmr      pJP5603_pvuA886–1523  pJP5603::disruption construct for gene CJU81_14195c, Kmr      aRelevant GenBank and Sequence Read Archive accession numbers provided. bPreviously sequenced P. fragi genomes that were examined for the presence of siderophore biosynthetic gene clusters. cGenomes sequenced in this study. dLocus tags of respective coding sequences. View Large Conditions of iron starvation were achieved in modified M9 media (MM9) comprising 10% v/v of MM9 salts (5 g/L NaCl, 10 g/L NH4Cl, 0.59 g/L Na2HPO4.H2O and 0.45 g/L KH2PO4), 2 mM MgSO4, 0.1 mM CaCl2, 0.2% w/v glucose, 0.3% casamino acids w/v, 0.2% w/v succinate and 0.1 M PIPES (pH 6.8, NaOH). Pseudomonas fragi cultures were grown under agitation unless otherwise specified. Kanamycin was added to culture media of gene disruption mutants at a concentration of 50 μg/ml. Identification of siderophore gene cluster Examination of the 13 previously sequenced P. fragi genomes (Table 1) for siderophore biosynthetic gene clusters was performed with antiSMASH version 4 (Blin et al.2017). Generation of disruption mutants, genomic DNA isolation, and genome sequencing and analysis A gene disruption mutant was generated of the TonB-dependent outer membrane siderophore receptor (ΔpvuA) by homologous recombination with a pJP5603 suicide vector construct. This system results in a single cross-over event, whereby the pJP5603 vector (Riedel et al.2013) encoding a kanamycin resistance gene remains integrated in the genome (Penfold and Pemberton 1992). Thus, a control strain with wild-type vibrioferrin genes and a disrupted lipase gene (Δlip) was generated for subsequent experimental comparisons. The lip gene was chosen because the disruption of this gene would not affect iron metabolism of the strain, nor would it be required for growth of the strain in the culture media used in this work. Furthermore, because it is present as single gene rather than in an operon cluster, polar effects on downstream genes were highly unlikely. DNA manipulations were performed using standard protocols (Sambrook and Russell 2001). Plasmid constructs were generated by ligating BamHI-digested pJP5603 and BamHI-restriction fragments of the P. fragi F1801 lip and pvuA genes (Table 1), which were amplified using primers included in Table S1 (Supporting Information). pJP5603 gene disruption constructs were transferred from E. coli S17–1 λpir (Simon, Priefer and Puhler 1983) to P. fragi F1801 by biparental filter matings as previously described (Windgassen, Urban and Jaeger 2000) with some adjustments; recipient cells were grown at 30°C and the ratio of donor to recipient cells was 1:2 (5 × 108 CFU to 109 CFU), respectively. After mating, cells were spread on Pseudomonas agar containing cetrimide-fucidin-cephaloridine selective supplement (Oxoid) for Pseudomonas and 50 μg/mL kanamycin sulphate to select for transconjugants. Plates were incubated at 25°C for 30 h. Strains (wild-type F1801, Δlip and ΔpvuA) were grown for 18 h in tryptone soya broth (Oxoid) at 25°C. Genomic DNA was isolated with the DNeasy Blood and Tissue Kit (QIAGEN) according to the manufacturer's protocol for Gram-negative bacteria. Library preparation and genome sequencing were carried out at Queensland Health (Health Support Queensland) using MiniSeq High Output kits with 150 cycles, and the Illumina MiniSeq System (Illumina). Genome assembly was performed as previously described (Stanborough et al.2017). Confirmation of correct insertion of the vector was achieved by manual inspection of the genomes with Geneious version 9.0.5 (Kearse et al.2012). Snippy (https://github.com/tseemann/snippy) was used to confirm an absence of additional genetic variants in the Δlip and ΔpvuA mutants’ genomes by aligning sequence reads of wild-type F1801, ΔpvuA and Δlip strains to the reference genome of F1801. Sequence reads of wild-type F1801, Δlip and ΔpvuA strains were deposited in the Sequence Read Archive, and the respective accession numbers are provided in Table 1. Phylogenetic tree of vibrioferrin sequences MAFFT multiple sequence alignments (Katoh et al.2002) of the vibrioferrin PvuA and PvsA-PvsE protein sequences were performed using the default settings. The six multiple alignments were concatenated and Gblocks version 0.91b (Castresana 2000; Talavera and Castresana 2007) was run with the default settings to remove poorly aligned positions and divergent regions of the alignments. Bayesian inference (BI), performed with MrBayes version 3.2.6 (Ronquist and Huelsenbeck 2003), was conducted with a run of 1000 000 generations and sampling every 1000. A mixed amino acid model analysis was set, enabling the MCMC sampler to test all of the fixed rate models. MrBayes determined that the Wheeler and Goldman model of amino acid replacement (Whelan and Goldman 2001) had the best likelihood score and was chosen for the analysis. Convergence parameters were assessed using Tracer version 1.6 (Rambaut et al.2014), and the majority rule consensus tree was rendered with Figtree version 1.4.2 (http://tree.bio.ed.ac.uk/software/figtree/). BIs were run three times to ensure reproducibility of the resulting trees. RNA isolation and quantitative reverse transcription-polymerase chain reaction RNA was isolated from 1 ml of mid-log phase iron-starved cultures (1.5 × 108 CFU). Stabilisation of RNA was achieved with RNAprotect Bacteria Reagent (QIAGEN) following the manufacturer's recommendations, and total RNA was extracted with the RNeasy Mini Kit (QIAGEN) according to the supplier's instructions and stored in H2O at –80°C. cDNA was synthesised from 1 μg of total RNA using iScript gDNA Clear cDNA Synthesis Kit (Biorad) as suggested by the manufacturer; however, DNase treatment was increased from the recommended 5 to 30 min. Quantitative reverse transcription-polymerase chain reaction (qRT-PCR) was performed on an AriaMx Real-time PCR System (Agilent) using the iTaq Universal SYBR Green Supermix (Biorad) as a fluorescence source. PCR reaction mixes included 10 μl iTaq Universal SYBR Green Supermix (Biorad), 500 nM forward and reverse primers, and 1 μl template cDNA (diluted 1:10 in H2O) in a total volume of 20 μL. Primer sequences, provided in Table S1 (Supporting Information), were designed with Primer3 version 2.3.4 (Untergasser et al.2012). Acceptable primer efficiency was confirmed for each target gene with standard curves, and melt curve analysis of amplicons of each target gene was performed to ensure amplification of single gene products. PCR conditions were as follows: 30 s at 95°C followed by 40 cycles of 5 s at 95°C and 30 s at 60°C. Melt curve analysis involved an incremental increase of 0.5°C every 2 s from 65°C to 95°C. Each assay included a non-template control, and each sample a control without reverse transcriptase. Raw data were analysed with Agilent AriaMx version 1.0 software and exported to excel for further analysis. Ct values were normalised to the isocitrate dehydrogenase gene (gene locus tag CJU81_09775 in F1801, GenBank accession no. GCA_002269505.1), and relative expression levels were calculated according to the 2ΔΔCt method. Chromeazurol S supernatant assays Chromeazurol S (CAS) assay solution was prepared as previously described (Alexander and Zuberer 1991). Siderophore detection from culture supernatants was performed using the method of Schwyn and Neilands (1987). Briefly, bacterial cultures were spun down at 2500 × g for 5 min and the supernatants were filtered through a 0.20-μm cellulose acetate membrane filter (Sartorius). Siderophores were detected in the supernatants by mixing equal volumes of filtered supernatant and CAS solution. After allowing the solutions to equilibrate for 1.5 h, a visible change in the colour of the mixture from blue to orange was considered a positive reaction. To detect siderophore production at 25°C, strains were grown in 4 ml MM9 for 18 h prior to supernatant collection. To monitor siderophore production at 4°C, F1801 was grown in 50 ml MM9 for 4 days and every 24 h 3 ml of culture were removed to measure growth (A600), and detect siderophores by CAS supernatant assays. To monitor the increase in siderophore levels over the 4 days, absorbance measurements of the CAS reaction mixtures were performed at 630 nm. Growth experiments with 2,2΄-bipyridyl and bovine apo-transferrin Single colonies were used to inoculate 3 ml MM9. Cultures were grown at 25°C for 18 h. Twenty microlitres of these cultures were used to inoculate 180 μl fresh MM9, supplemented as indicated in the text with 2,2΄-bipyridyl (Sigma-Aldrich), bovine apo-transferrin (Sigma-Aldrich) and/or sodium bicarbonate and FeCl3, in Greiner 96-well flat bottom microtitre plates (Sigma-Aldrich). Growth was monitored (A600) using an EON microplate reader (BioTek). RESULTS AND DISCUSSION In this work, 13 previously sequenced P. fragi genomes were examined for the presence of siderophore biosynthetic gene clusters. This search resulted in the identification of a conserved gene cluster, homologues of which were shown to be responsible for vibrioferrin biosynthesis and vibrioferrin-mediated iron uptake in the terrestrial bacteria Azotobacter vinelandii and Xanthomonas spp., and in the marine bacterium Vibrio parahaemolyticus (Yamamoto et al.1994; Pandey and Sonti 2010; Baars et al.2016; Pandey et al.2017) (Fig. 1). Similar to the vibrioferrin gene clusters in A. vinelandii and Xanthomonas campestris, this ∼10 kb region in P. fragi comprised seven open reading frames (Fig. 1A). Previous work in Xanthomonas spp. and V. parahaemolyticus showed that five of these genes (mhpE, pvsA, pvsB, pvsD and pvsE) are involved in the biosynthesis of vibrioferrin, while pvuA is a TonB-dependent outer membrane receptor and pvsC an inner membrane exporter for vibrioferrin (Funahashi et al.2002; Tanabe et al.2006; Pandey and Sonti 2010; Fujita et al.2011). Locus tags of these genes in the 13 strains are shown in Table 1. Analysis of the upstream region of this gene cluster revealed the presence of a 19-bp sequence, matching 14 of the 19 nucleotides of the X. campestris consensus Fur box (Blanvillain et al.2007) and 11 of the 19 nucleotides of the E. coli and P. aeruginosa consensus Fur box. This sequence was located 129 bases upstream of the mhpE start codon. Fur homologues were also identified in the P. fragi genomes with the locus tag CJU81_16820 in F1801. Figure 1. View largeDownload slide Vibrioferrin gene cluster identified in 13 P. fragi. (A) Schematic diagram of the vibrioferrin gene cluster identified in P. fragi and conserved synteny of these genes in known vibrioferrin producing bacteria A. vinelandii, X. campestris and V. parahaemolyticus. Predicted biosynthesis genes are marked with an asterisk in the P. fragi genes. Numbers shown in genes indicate percent sequence identity of proteins to respective homologues in P. fragi F1801, as determined by MAFFT protein alignment. The putative Fur box identified upstream of the first gene in the P. fragi cluster is shown and consensus Fur box sequences of X. campestris and E. coli (identical to the P. aeruginosa consensus Fur box) are included for comparison. (B) Mid-point rooted phylogram of the consensus BI tree of the PvuA and PvsA-PvsE sequences of the 13 P. fragi, A. vinelandii, X. campestris and V. parahaemolyticus. Pseudomonas fragi F1801 that was used in subsequent experiments is marked with an asterisk. The subtree shown in the box represents BI branching on a different scale to the tree. Posterior probability values are shown as percentages at the nodes, and scale bars denote the number of substitutions per site. Figure 1. View largeDownload slide Vibrioferrin gene cluster identified in 13 P. fragi. (A) Schematic diagram of the vibrioferrin gene cluster identified in P. fragi and conserved synteny of these genes in known vibrioferrin producing bacteria A. vinelandii, X. campestris and V. parahaemolyticus. Predicted biosynthesis genes are marked with an asterisk in the P. fragi genes. Numbers shown in genes indicate percent sequence identity of proteins to respective homologues in P. fragi F1801, as determined by MAFFT protein alignment. The putative Fur box identified upstream of the first gene in the P. fragi cluster is shown and consensus Fur box sequences of X. campestris and E. coli (identical to the P. aeruginosa consensus Fur box) are included for comparison. (B) Mid-point rooted phylogram of the consensus BI tree of the PvuA and PvsA-PvsE sequences of the 13 P. fragi, A. vinelandii, X. campestris and V. parahaemolyticus. Pseudomonas fragi F1801 that was used in subsequent experiments is marked with an asterisk. The subtree shown in the box represents BI branching on a different scale to the tree. Posterior probability values are shown as percentages at the nodes, and scale bars denote the number of substitutions per site. To investigate the relatedness of the vibrioferrin genes between the 13 strains and A. vinelandii, X. campestris and V. parahaemolyticus, a phylogenetic tree of the PvuA and PvsA-PvsE sequences was determined by BI (Fig. 1B). Topology of the BI consensus tree showed three clades of P. fragi vibrioferrin sequences. The largest of the three clades, comprising closely related sequences of 8 of the 13 strains, was clearly separated from the other two clades, which share a more recent common ancestral sequence. The sequences of the P. fragi type strain were found in one of these two clades with those of two additional strains. Sequences of F1801, the P. fragi strain used for subsequent experiments, were grouped together in the second of these clades with vibrioferrin sequences of one other strain. In accordance with their phylogeny, sequences of the P. fragi were closest to those of A. vinelandii, while those of X. campestris and V. parahaemolyticus were substantially more distant. Previous work suggested that P. fragi does not synthesise siderophores (Champomier-Vergès, Stintzi and Meyer 1996), and utilising the succinate media (non-deferrated and deferrated forms) presented in the work, P. fragi F1801 displayed little to no growth and siderophore production of this strain was not observed (data not shown). Therefore, in this study, a non-deferrated, modified M9 medium (MM9) was used containing glucose that was not present in the media used by Champomier-Vergès et al. (1996). MM9 enabled both satisfactory growth, as demonstrated by growth of F1801 at 25°C, while still ensuring conditions of iron starvation, which can be seen by the strong increase in growth of F1801 in MM9 supplemented with 50 μM FeCl3 (Fig. 2A). Figure 2. View largeDownload slide Siderophore production by P. fragi F1801 at 25°C and 4°C. (A) Growth of F1801 in MM9 or in MM9 that was supplemented with 50 μM FeCl3. Optical density of the cultures was measured at 600 nm every hour in an EON microplate reader (BioTek). Data shown are the means and error bars the standard deviations of three biological replicates. (B) CAS assay results of F1801. Shown from left to right is CAS solution that was added to sterile MM9 (control), filtered culture supernatant of F1801 grown in MM9 at 25°C (F1801), sterile MM9 supplemented with 50 μM FeCl3 (control (+FeCl3)) and filtered culture supernatant of F1801 grown at 25°C in MM9 supplemented with 50 μM FeCl3 (F1801 (+FeCl3)). (C) Relative quantification by qRT-PCR of the vibrioferrin genes in F1801 under iron-replete conditions (MM9 supplemented with 50 μM FeCl3) compared to conditions of iron starvation (MM9). Data shown are the means and error bars the standard deviations of three biological replicates. (D) Growth of and siderophore production by F1801 in MM9 at 4°C. Optical density of the cultures and CAS assay absorbance measurements were carried out with a Novaspec Plus visible spectrophotometer (VWR). Data shown are the means and error bars the standard deviations of three biological replicates. Figure 2. View largeDownload slide Siderophore production by P. fragi F1801 at 25°C and 4°C. (A) Growth of F1801 in MM9 or in MM9 that was supplemented with 50 μM FeCl3. Optical density of the cultures was measured at 600 nm every hour in an EON microplate reader (BioTek). Data shown are the means and error bars the standard deviations of three biological replicates. (B) CAS assay results of F1801. Shown from left to right is CAS solution that was added to sterile MM9 (control), filtered culture supernatant of F1801 grown in MM9 at 25°C (F1801), sterile MM9 supplemented with 50 μM FeCl3 (control (+FeCl3)) and filtered culture supernatant of F1801 grown at 25°C in MM9 supplemented with 50 μM FeCl3 (F1801 (+FeCl3)). (C) Relative quantification by qRT-PCR of the vibrioferrin genes in F1801 under iron-replete conditions (MM9 supplemented with 50 μM FeCl3) compared to conditions of iron starvation (MM9). Data shown are the means and error bars the standard deviations of three biological replicates. (D) Growth of and siderophore production by F1801 in MM9 at 4°C. Optical density of the cultures and CAS assay absorbance measurements were carried out with a Novaspec Plus visible spectrophotometer (VWR). Data shown are the means and error bars the standard deviations of three biological replicates. The sensitive CAS supernatant assay was used to detect the presence of siderophores in culture supernatants of F1801 grown at optimal growth temperature (25°C) in MM9. The CAS assay is based on the removal of iron from the chromogenic dye CAS by siderophores, whereby a colour change is observed from blue to orange (Schwyn and Neilands 1987). Figure 2B shows that siderophores were clearly detected in cell-free supernatants of iron-starved cultures of F1801. Reaction times for the exchange of iron from CAS were within 1.5 h, a time-frame consistent with carboxylate-type siderophores (Schwyn and Neilands 1987). In contrast, when F1801 was grown in MM9 supplemented with 50 mM FeCl3, siderophores were largely absent from culture supernatants (Fig. 2B). Furthermore, expression of the vibrioferrin genes was strongly downregulated (between 12-fold (pvsE) and 120-fold (pvsD) reduction) relative to their expression in MM9 without iron supplementation (Fig. 2C), indicating repression of siderophore production under iron-rich conditions. These results are in agreement with other studies, which showed that vibrioferrin production in A. vinelandii, Xanthomonas and V. parahaemolyticus increased under iron limitation, while under iron-replete conditions vibrioferrin production ceased (Funahashi et al.2000; Pandey and Sonti 2010; McRose et al.2017; Pandey et al.2017). Thus, together with the presence of a putative Fur box upstream of the vibrioferrin gene cluster, these results suggest Fur-mediated regulation of the P. fragi vibrioferrin genes. As a key meat and milk spoilage bacterium, P. fragi F1801 was tested for siderophore production at refrigeration temperature by monitoring growth and the accumulation of siderophores in supernatants at 4°C over 4 days. Blue-coloured CAS solution has a maximum absorption at 630 nm. Orange-coloured CAS reaction mixtures, in which siderophores have removed iron from CAS, have essentially no absorbance at this wavelength and there is a largely linear dependence of A630 values of CAS reaction mixtures versus concentration of the chelator (Schwyn and Neilands 1987). Figure 2D shows a steady decrease in A630 values of CAS reaction mixtures over the 4 days, indicating an accumulation of siderophores in culture supernatants of F1801. This accumulation correlated well with cell growth of the strain, demonstrating siderophore production of P. fragi at temperatures relevant to the storage of meat and milk. To confirm a role for the vibrioferrin gene cluster in siderophore-mediated iron acquisition, a gene disruption mutant of the TonB-dependent outer membrane siderophore receptor (ΔpvuA) was generated, as well as a control strain with wild-type vibrioferrin genes and a disrupted lipase gene (Δlip) for subsequent experimental comparisons. qRT-PCR analysis revealed that expression of genes downstream of pvuA in the ΔpvuA mutant was strongly reduced (between 7-fold (pvsE) and 128-fold (pvsA) reduction) relative to the Δlip control strain (Fig. 3A), indicating polar effects caused by integration of the suicide vector in the pvuA gene. As our aim was to confirm a role for this gene cluster in siderophore-mediated iron uptake, rather than to study the function of the vibrioferrin genes, which has been addressed in other studies (Funahashi et al.2002; Tanabe et al.2003, 2006; Pandey and Sonti 2010; Fujita et al.2011; Pandey et al.2017), the ΔpvuA mutant was used for further investigations. Figure 3. View largeDownload slide ΔpvuA shows impaired siderophore production. (A) Relative quantification by qRT-PCR of expression of the vibrioferrin genes in ΔpvuA compared with the Δlip control strain. Data shown are the means and error bars the standard deviations of three biological replicates. (B) CAS assay results of the ΔpvuA mutant and Δlip control strain. Shown from left to right is CAS solution that was added to sterile MM9 (control), filtered culture supernatant of Δlip grown in MM9 at 25°C and filtered culture supernatant of ΔpvuA grown in MM9 at 25°C. Figure 3. View largeDownload slide ΔpvuA shows impaired siderophore production. (A) Relative quantification by qRT-PCR of expression of the vibrioferrin genes in ΔpvuA compared with the Δlip control strain. Data shown are the means and error bars the standard deviations of three biological replicates. (B) CAS assay results of the ΔpvuA mutant and Δlip control strain. Shown from left to right is CAS solution that was added to sterile MM9 (control), filtered culture supernatant of Δlip grown in MM9 at 25°C and filtered culture supernatant of ΔpvuA grown in MM9 at 25°C. A strong reduction in siderophore levels was detected in culture supernatants of ΔpvuA compared to the Δlip control strain (Fig. 3B). Gene deletion of pvuA, which encodes an outer membrane vibrioferrin receptor, should result in an accumulation of siderophores in culture supernatants. Therefore, the observed reduction was likely due to the polar effects on the downstream biosynthesis genes. Genetic variants were not found in genes or in non-coding regions associated with the vibrioferrin genes or other iron-related genes of the ΔpvuA mutant's genome (Table S2, Supporting Information), demonstrating that phenotypes of the ΔpvuA mutant were solely due to the disruption of the target gene and associated polar effects. A clear growth defect of the ΔpvuA mutant was not observed in MM9 when compared to the Δlip control strain (Fig. 4A). Therefore, to understand the role of the vibrioferrin siderophore under low iron conditions, growth of the ΔpvuA and Δlip strains was tested under more severe iron-depleted conditions by adding 25 μM of the iron chelator 2,2΄-bipyridyl (BP) to MM9. While growth of the Δlip control strain was unaffected at 25 μM BP, addition of the ferrous iron chelator at these concentrations markedly impaired growth of the ΔpvuA mutant, demonstrating that the siderophore is involved in iron acquisition under low iron conditions. Figure 4. View largeDownload slide Vibrioferrin siderophore plays a role under iron starvation and scavenges iron from transferrin. Growth of ΔpvuA and Δlip control strain in MM9, and in MM9 supplemented with 25 μM 2,2΄-bipyridyl (BP) (A); in MM9, and in MM9 supplemented with 20 mM bicarbonate (BC) (B); in MM9 supplemented with 2.5 μM bovine apo-transferrin (apo-Tsf) and 20 mM BC (C); and in MM9 supplemented with 2.5 μM apo-Tsf, 20 mM BC and 100 μM FeCl3 (D). Data shown for growth experiments in (A–D) are the means and error bars the standard deviations of three biological replicates. All experiments were repeated a minimum of three times with similar results observed. Figure 4. View largeDownload slide Vibrioferrin siderophore plays a role under iron starvation and scavenges iron from transferrin. Growth of ΔpvuA and Δlip control strain in MM9, and in MM9 supplemented with 25 μM 2,2΄-bipyridyl (BP) (A); in MM9, and in MM9 supplemented with 20 mM bicarbonate (BC) (B); in MM9 supplemented with 2.5 μM bovine apo-transferrin (apo-Tsf) and 20 mM BC (C); and in MM9 supplemented with 2.5 μM apo-Tsf, 20 mM BC and 100 μM FeCl3 (D). Data shown for growth experiments in (A–D) are the means and error bars the standard deviations of three biological replicates. All experiments were repeated a minimum of three times with similar results observed. To investigate whether the vibrioferrin siderophore could play a role in iron metabolism of P. fragi in foods such as meat and milk, where iron is sequestered by high-affinity iron-binding proteins, growth of ΔpvuA and Δlip strains was compared in MM9, to which bovine apo-transferrin (apo-Tsf), as a model transferrin protein, was added. In all transferrin proteins, the carbonate (or bicarbonate) of the binding cleft is the synergistic anion, without which the protein loses its affinity for iron (Pakdaman and El Hage Chahine 1997). Thus, to ensure the specific binding of Fe3+ by apo-Tsf, the growth media was also supplemented with 20 mM bicarbonate (BC) (Chung 1984). BC at a concentration of 20 mM had no effect on the strains’ growth when compared to their growth in MM9 without BC addition (Fig. 4B). When the growth medium was supplemented with apo-Tsf at a concentration of 2.5 μM in the presence of 20 mM BC, growth, although slowed, was observed for the Δlip control strain (Fig. 4C). In contrast, growth of ΔpvuA was completely inhibited in the presence of apo-Tsf and BC, but could be rescued by supplementation of this media with 100 μM FeCl3 (Fig. 4D). These data suggest that the vibrioferrin siderophore may be capable of removing ferric iron from transferrin proteins. Remarkably, despite considerable divergence between the six vibrioferrin protein sequences shared by A. vinelandii and V. parahaemolyticus, the absence of an mhpE gene in V. parahaemolyticus, and the absence of the vibrioferrin-uptake genes pvuA(1)-pvuE in A. vinelandii, both bacteria were shown to produce identical vibrioferrin compounds (Yamamoto et al.1994; Baars et al.2016). Based on percent identities and the evolutionary relationship of the P. fragi vibrioferrin sequences with A. vinelandii and V. parahaemolyticus counterparts, it is likely that the P. fragi siderophore is an α-hydroxycarboxylate type siderophore, similar to or identical to vibrioferrin. CONCLUSION A vibrioferrin gene cluster was identified in 13 strains of the food spoilage bacterium P. fragi, and siderophore production of P. fragi F1801 was observed. Based on the homology of the identified genes, P. fragi likely excretes an α-hydroxycarboxylate type siderophore, similar to or identical to vibrioferrin, which plays a role in iron metabolism of this bacterium under low iron conditions. More experiments are required to determine whether the siderophore contributes to the successful growth and colonisation of milk and meat by P fragi; however, these data suggest that the vibrioferrin siderophore is capable of removing ferric iron from transferrin proteins. SUPPLEMENTARY DATA Supplementary data are available at FEMSLE online. Acknowledgements We thank Steve Petrovski for providing plasmid pJP5603 and Escherichia coli strain S17–1 λpir. FUNDING This work was supported by the Australian Meat Processor Corporation (grant number 2013-5041) and the Commonwealth Scientific and Industrial Research Organisation (grant number OD-106390). Conflicts of interest. None declared. REFERENCES Alexander DB, Zuberer DA. Use of chrome azurol S reagents to evaluate siderophore production by rhizosphere bacteria. Biol Fert Soils  1991; 12: 39– 45. Google Scholar CrossRef Search ADS   Baars O, Zhang X, Morel FMM et al.   The siderophore metabolome of Azotobacter vinelandii. Appl Environ Microb  2016; 82: 27– 39. Google Scholar CrossRef Search ADS   Blanvillain S, Meyer D, Boulanger A et al.   Plant carbohydrate scavenging through TonB-dependent receptors: A feature shared by phytopathogenic and aquatic bacteria. PLoS One  2007; 2: e224. Google Scholar CrossRef Search ADS PubMed  Blin K, Wolf T, Chevrette MG et al.   antiSMASH 4.0-improvements in chemistry prediction and gene cluster boundary identification. Nucleic Acids Res  2017, DOI 10.1093/nar/gkx319. Carpenter BM, Whitmire JM, Merrell DS. This is not your mother's repressor: the complex role of Fur in pathogenesis. Infect Immun  2009; 77: 2590– 601. Google Scholar CrossRef Search ADS PubMed  Casaburi A, Piombino P, Nychas G-J et al.   Bacterial populations and the volatilome associated to meat spoilage. Food Microbiol  2015; 45( Part A): 83– 102. Google Scholar CrossRef Search ADS PubMed  Castresana J. Selection of conserved blocks from multiple alignments for their use in phylogenetic analysis. Mol Biol Evol  2000; 17: 540– 52. Google Scholar CrossRef Search ADS PubMed  Cezard C, Farvacques N, Sonnet P. Chemistry and biology of pyoverdines, Pseudomonas primary siderophores. Curr Med Chem  2015; 22: 165– 86. Google Scholar CrossRef Search ADS PubMed  Champomier-Vergès M-C, Stintzi A, Meyer J-M. Acquisition of iron by the non-siderophore-producing Pseudomonas fragi. Microbiology  1996; 142: 1191– 9. Google Scholar CrossRef Search ADS PubMed  Chung MC-M. Structure and function of transferrin. Biochem Educ  1984; 12: 146– 54. Google Scholar CrossRef Search ADS   Colombo C, Palumbo G, He J-Z et al.   Review on iron availability in soil: Interaction of Fe minerals, plants, and microbes. J Soils Sed  2014; 14: 538– 48. Google Scholar CrossRef Search ADS   Cox CD, Adams P. Siderophore activity of pyoverdin for Pseudomonas aeruginosa. Infect Immun  1985; 48: 130– 8. Google Scholar PubMed  De Jonghe V, Coorevits A, Van Hoorde K et al.   Influence of storage conditions on the growth of Pseudomonas species in refrigerated raw milk. Appl Environ Microb  2011; 77: 460– 70. Google Scholar CrossRef Search ADS   Faraldo-Gomez JD, Sansom MSP. Acquisition of siderophores in Gram-negative bacteria. Nat Rev Mol Cell Bio  2003; 4: 105– 16. Google Scholar CrossRef Search ADS   Farh ME-A, Kim Y-J, Sukweenadhi J et al.   Aluminium resistant, plant growth promoting bacteria induce overexpression of Aluminium stress related genes in Arabidopsis thaliana and increase the ginseng tolerance against Aluminium stress. Microbiol Res  2017; 200: 45– 52. Google Scholar CrossRef Search ADS PubMed  Fujita MJ, Kimura N, Sakai A et al.   Cloning and heterologous expression of the vibrioferrin biosynthetic gene cluster from a marine metagenomic library. Biosci Biotech Bioch  2011; 75: 2283– 7. Google Scholar CrossRef Search ADS   Funahashi T, Fujiwara C, Okada M et al.   Characterization of Vibrio parahaemolyticus manganese-resistant mutants in reference to the function of the ferric uptake regulatory protein. Microbiol Immunol  2000; 44: 963– 70. Google Scholar CrossRef Search ADS PubMed  Funahashi T, Moriya K, Uemura S et al.   Identification and characterization of pvuA, a gene encoding the ferric vibrioferrin receptor protein in Vibrio parahaemolyticus. J Bacteriol  2002; 184: 936– 46. Google Scholar CrossRef Search ADS PubMed  Garrity GM, Bell JA, Lilburn T. Pseudomonadales Orla-Jensen 1921, 270AL. In: Brenner DJ, Krieg NR, Staley JT et al.   (eds.) Bergey's Manual® of Systematic Bacteriology: Volume Two The Proteobacteria Part B The Gammaproteobacteria . Boston, MA: Springer, 2005, 323– 442. Google Scholar CrossRef Search ADS   Ilbert M, Bonnefoy V. Insight into the evolution of the iron oxidation pathways. Biochim Biophys Acta  2013; 1827: 161– 75. Google Scholar CrossRef Search ADS PubMed  Jones CM, Niederweis M. Role of porins in iron uptake by Mycobacterium smegmatis. J Bacteriol  2010; 192: 6411– 7. Google Scholar CrossRef Search ADS PubMed  Katoh K, Misawa K, Kuma K-i et al.   MAFFT: a novel method for rapid multiple sequence alignment based on fast Fourier transform. Nucleic Acids Res  2002; 30: 3059– 66. Google Scholar CrossRef Search ADS PubMed  Kearse M, Moir R, Wilson A et al.   Geneious Basic: an integrated and extendable desktop software platform for the organization and analysis of sequence data. Bioinformatics  2012; 28: 1647– 9. Google Scholar CrossRef Search ADS PubMed  McRose DL, Baars O, Morel FMM et al.   Siderophore production in Azotobacter vinelandii in response to Fe-, Mo- and V-limitation. Environ Microbiol  2017; 19: 3595– 605. Google Scholar CrossRef Search ADS PubMed  Meyer JM, Abdallah MA. The fluorescent pigment of Pseudomonas fluorescens: biosynthesis, purification and physicochemical properties. Microbiology  1978; 107: 319– 28. Noinaj J, Guillier M, Travis J. Barnard et al.   TonB-dependent transporters: regulation, structure, and function. Annu Rev Microbiol  2010; 64: 43– 60. Google Scholar CrossRef Search ADS PubMed  Pakdaman R, El Hage Chahine J-M. Transferrin. Eur J Biochem  1997; 249: 149– 55. Google Scholar CrossRef Search ADS PubMed  Pandey A, Sonti RV. Role of the FeoB protein and siderophore in promoting virulence of Xanthomonas oryzae pv. oryzae on rice. J Bacteriol  2010; 192: 3187– 203. Google Scholar CrossRef Search ADS PubMed  Pandey SS, Patnana PK, Rai R et al.   Xanthoferrin, the alpha-hydroxycarboxylate-type siderophore of Xanthomonas campestris pv. campestris, is required for optimum virulence and growth inside cabbage. Mol Plant Pathol  2017; 18: 949– 62. Google Scholar CrossRef Search ADS PubMed  Penfold RJ, Pemberton JM. An improved suicide vector for construction of chromosomal insertion mutations in bacteria. Gene  1992; 118: 145– 6. Google Scholar CrossRef Search ADS PubMed  Rambaut A, Suchard MA, Xie D et al.   Tracer v1.6 . 2014 http://tree.bio.ed.ac.uk/software/tracer/. Riedel T, Rohlfs M, Buchholz I et al.   Complete sequence of the suicide vector pJP5603. Plasmid  2013; 69: 104– 7. Google Scholar CrossRef Search ADS PubMed  Ronquist F, Huelsenbeck JP. MrBayes 3: Bayesian phylogenetic inference under mixed models. Bioinformatics  2003; 19: 1572– 4. Google Scholar CrossRef Search ADS PubMed  Sambrook J, Russell DW. Molecular Cloning: A Laboratory Manual . Cold Spring Harbor, NY: Cold Spring Harbor Laboratory, 2001. Schwyn B, Neilands JB. Universal chemical assay for the detection and determination of siderophores. Anal Biochem  1987; 160: 47– 56. Google Scholar CrossRef Search ADS PubMed  Simon R, Priefer U, Puhler A. A broad host range mobilization system for in vivo genetic engineering: transposon mutagenesis in gram negative bacteria. Nat Biotech  1983; 1: 784– 91. Google Scholar CrossRef Search ADS   Skaar EP. The battle for iron between bacterial pathogens and their vertebrate hosts. PLoS Pathog  2010; 6: e1000949. Google Scholar CrossRef Search ADS PubMed  Stanborough T, Fegan N, Powell SM et al.   Insight into the genome of Brochothrix thermosphacta, a problematic meat spoilage bacterium. Appl Environ Microb  2017; 83: e02786– 16. Google Scholar CrossRef Search ADS   Talavera G, Castresana J. Improvement of phylogenies after removing divergent and ambiguously aligned blocks from protein sequence alignments. Syst Biol  2007; 56: 564– 77. Google Scholar CrossRef Search ADS PubMed  Tanabe T, Funahashi T, Nakao H et al.   Identification and characterization of genes required for biosynthesis and transport of the siderophore vibrioferrin in Vibrio parahaemolyticus. J Bacteriol  2003; 185: 6938– 49. Google Scholar CrossRef Search ADS PubMed  Tanabe T, Nakao H, Kuroda T et al.   Involvement of the Vibrio parahaemolyticus pvsC gene in export of the siderophore vibrioferrin. Microbiol Immunol  2006; 50: 871– 6. Google Scholar PubMed  Trapet P, Avoscan L, Klinguer A et al.   The Pseudomonas fluorescens siderophore pyoverdine weakens Arabidopsis thaliana defense in favour of growth in iron-deficient conditions. Plant Physiol  2016, DOI 10.1104/pp.15.01537. Troxell B, Hassan HM. Transcriptional regulation by Ferric Uptake Regulator (Fur) in pathogenic bacteria. Front Cell Infect Microbiol  2013; 3: 59. Google Scholar PubMed  Untergasser A, Cutcutache I, Koressaar T et al.   Primer3–new capabilities and interfaces. Nucleic Acids Res  2012; 40: e115. Google Scholar CrossRef Search ADS PubMed  Whelan S, Goldman N. A general empirical model of protein evolution derived from multiple protein families using a maximum-likelihood approach. Mol Biol Evol  2001; 18: 691– 9. Google Scholar CrossRef Search ADS PubMed  Windgassen M, Urban A, Jaeger K-E. Rapid gene inactivation in Pseudomonas aeruginosa. FEMS Microbiol Lett  2000; 193: 201– 5. Google Scholar CrossRef Search ADS PubMed  Yamamoto S, Okujo N, Yoshida T et al.   Structure and iron transport activity of vibrioferrin, a new siderophore of Vibrio parahaemolyticus. J Biochem  1994; 115: 868– 74. Google Scholar CrossRef Search ADS PubMed  © FEMS 2017. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png FEMS Microbiology Letters Oxford University Press

Vibrioferrin production by the food spoilage bacterium Pseudomonas fragi

Loading next page...
 
/lp/ou_press/vibrioferrin-production-by-the-food-spoilage-bacterium-pseudomonas-0GaUGHulJQ
Publisher
Blackwell
Copyright
© FEMS 2017. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com
ISSN
0378-1097
eISSN
1574-6968
D.O.I.
10.1093/femsle/fnx279
Publisher site
See Article on Publisher Site

Abstract

Abstract Pseudomonas fragi is a meat and milk spoilage bacterium with high iron requirements; however, mechanisms of iron acquisition remain largely unknown. The aim of this work was to investigate siderophore production as an iron acquisition system for P. fragi. A vibrioferrin siderophore gene cluster was identified in 13 P. fragi, and experiments were conducted with a representative strain of this group (F1801). Chromeazurol S assays showed that P. fragi F1801 produced siderophores under iron starvation at optimum growth and refrigeration temperature. Conversely, supplementation of low iron media with 50 μM FeCl3 repressed transcription of the vibrioferrin genes and siderophore production. Disruption of the siderophore receptor (pvuA) caused polar effects on downstream vibrioferrin genes, resulting in impaired siderophore production of the ΔpvuA mutant. Growth of this mutant was compared to growth of a control strain (Δlip) with wild-type vibrioferrin genes in low iron media supplemented with iron chelators 2,2΄-bipyridyl or apo-transferrin. While 25 μM 2,2΄-bipyridyl caused impaired growth of ΔpvuA, growth of the mutant was completely inhibited by 2.5 μM apo-transferrin, but could be restored by FeCl3 addition. In summary, this work identifies a vibrioferrin-mediated iron acquisition system of P. fragi, which is required for growth of this bacterium under iron starvation. Pseudomonas fragi, siderophore, vibrioferrin, food spoilage, iron acquisition, iron chelators INTRODUCTION Iron is required for fundamental cellular processes such as respiration and the synthesis of amino acids and DNA, and is therefore an essential element for the majority of microorganisms (Ilbert and Bonnefoy 2013). However, iron is in short supply in many habitats. For example, in aerobic environments at neutral pH, Fe2+ is rapidly oxidised to Fe3+, leading to the formation of insoluble hydroxide salts and rendering iron scarce in the environment (Colombo et al.2014). Similarly, in the human host and foods such as milk and meat, iron is sequestered by iron-binding proteins including transferrin family proteins, transferrin and lactoferrin, hemeproteins myoglobin and haemoglobin, and ferritin (Skaar 2010). Microorganisms have evolved a variety of mechanisms to obtain their iron. In iron-rich environments, iron acquisition occurs through energy-independent, low-affinity iron uptake systems (Jones and Niederweis 2010). In iron-limited environments, a common strategy involves the excretion of small molecular weight iron-chelating compounds termed siderophores (Noinaj et al.2010). Fe3+-siderophore complexes that form outside the cell are taken up in Gram-negative bacteria by specific outer membrane receptors, which are supplied with energy for transport from the cytoplasmic membrane by the Ton system (Faraldo-Gomez and Sansom 2003). Excess iron causes toxic cellular effects, necessitating the tight regulation of iron uptake in microorganisms. The ferric uptake regulator protein (Fur) regulates iron homoeostasis at the transcription level in many Gram-negative and some Gram-positive organisms (Carpenter, Whitmire and Merrell 2009). When intracellular Fe2+ concentrations exceed a certain threshold level, Fur binds to specific regulatory DNA sequences termed Fur boxes, preventing RNA polymerase binding to promotors (Troxell and Hassan 2013). In contrast, scarcity of intracellular iron causes Fur to lose its ability to bind to Fur boxes, resulting in gene transcription. Thus, the transcription of iron uptake genes is ultimately regulated by the concentration of intracellular iron. Pseudomonas have some of the best studied siderophore-mediated iron uptake systems such as the pyoverdine siderophores of the fluorescent pseudomonads (Meyer and Abdallah 1978; Cox and Adams 1985; Cezard, Farvacques and Sonnet 2015; Trapet et al.2016). In contrast, Pseudomonas fragi has been considered a non-siderophore producing member of this genus and is described as not producing siderophores in detectable amounts in Bergey's Manual (Champomier-Vergès, Stintzi and Meyer 1996; Garrity, Bell and Lilburn 2005). Pseudomonas fragi is a problematic meat and milk spoilage bacterium with high iron requirements, but beyond its ability to scavenge foreign Fe3+-siderophore complexes, very little is known about the strategies this bacterium employs to obtain iron (Champomier-Vergès, Stintzi and Meyer 1996; De Jonghe et al.2011; Casaburi et al.2015). Interestingly, in a recent study on plant growth-promoting bacteria, a putative P. fragi isolate derived from the rhizosphere was shown to produce siderophores (Farh et al.2017), suggesting previous conclusions that this bacterium is a non-siderophore producer may not have been accurate. In this work, siderophore production was investigated as an iron acquisition system for P. fragi. The genomes of 13 P. fragi were examined for the presence of siderophore biosynthetic gene clusters, which led to the identification of vibrioferrin biosynthesis and vibrioferrin-mediated iron acquisition genes. This work describes experiments that determine siderophore production for P. fragi F1801 and show a role for the siderophore in iron metabolism of this bacterium under low iron conditions. MATERIALS AND METHODS Strains, plasmids and growth conditions Strains, plasmids, and genome and gene sequences used in this study are included in Table 1. Table 1. Strains, plasmids, and genome and gene sequences used in this study. Strains, plasmids, and genome and gene sequences  Relevant characteristics  Locus tags of vibrioferrin genes  Accession numbersa  Pseudomonas fragi genome sequences  F1786b    CJU73_02415–CJU73_02445  GCA_002269585.1  F1791b    CJU79_02430–CJU79_02400  GCA_002269515.1  F1792b    CJU72_13900–CJU72_13930  GCA_002269595.1  F1793b    CJU75_07595–CJU75_07565  GCA_002269565.1  F1794b    CJU80_05030–CJU80_05000  GCA_002269445.1  F1813b    CJU77_09910–CJU77_09940  GCA_002269465.1  F1815b    CJU76_02305–CJU76_02275  GCA_002269545.1  F1816b    CJU74_06845–CJU76_06815  GCA_002269485.1  F1818b    CJU78_20565–CJU78_20595  GCA_002269625.1  F1820b    CJF43_16240–CJF43_16270  GCA_002269055.1  F1821b    CJF37_06830–CJF37_06800  GCA_002269155.1  ATCC 4973b    SAMN05216594_0902–SAMN05216594_0896  GCA_900105835.1  F1801b    CJU81_14170–CJU81_14200  GCA_002269505.1  Pseudomonas fragi strains  F1801c  Wild-type    SRX3235903  F1801Δlipc  Disruption of CJU81_12870b at 167th amino acid encoding codon, Kmr    SRX3235904  F1801ΔpvuAc  Disruption of CJU81_14195b at 419th amino acid encoding codon, Kmr    SRX3235905  Escherichia coli strains  Escherichia coli S17–1 λpir  thi, pro, hsd (r− m+) recA::RP4–2-Tcr::Mu Kmr::Tn7 Tpr Smrλpir      Genome/gene sequences of other bacteria  Azotobacter vinelandii CA      GCA_000380335.1  Xanthomonas campestris 8004      GCA_000012105.1  Vibrio parahaemolyticus WP1      AB048250.2 and AB082123.1  Plasmids  pJP5603  Suicide plasmid, Kmr      pJP5603_lip323–783  pJP5603::disruption construct for gene CJU81_12870c, Kmr      pJP5603_pvuA886–1523  pJP5603::disruption construct for gene CJU81_14195c, Kmr      Strains, plasmids, and genome and gene sequences  Relevant characteristics  Locus tags of vibrioferrin genes  Accession numbersa  Pseudomonas fragi genome sequences  F1786b    CJU73_02415–CJU73_02445  GCA_002269585.1  F1791b    CJU79_02430–CJU79_02400  GCA_002269515.1  F1792b    CJU72_13900–CJU72_13930  GCA_002269595.1  F1793b    CJU75_07595–CJU75_07565  GCA_002269565.1  F1794b    CJU80_05030–CJU80_05000  GCA_002269445.1  F1813b    CJU77_09910–CJU77_09940  GCA_002269465.1  F1815b    CJU76_02305–CJU76_02275  GCA_002269545.1  F1816b    CJU74_06845–CJU76_06815  GCA_002269485.1  F1818b    CJU78_20565–CJU78_20595  GCA_002269625.1  F1820b    CJF43_16240–CJF43_16270  GCA_002269055.1  F1821b    CJF37_06830–CJF37_06800  GCA_002269155.1  ATCC 4973b    SAMN05216594_0902–SAMN05216594_0896  GCA_900105835.1  F1801b    CJU81_14170–CJU81_14200  GCA_002269505.1  Pseudomonas fragi strains  F1801c  Wild-type    SRX3235903  F1801Δlipc  Disruption of CJU81_12870b at 167th amino acid encoding codon, Kmr    SRX3235904  F1801ΔpvuAc  Disruption of CJU81_14195b at 419th amino acid encoding codon, Kmr    SRX3235905  Escherichia coli strains  Escherichia coli S17–1 λpir  thi, pro, hsd (r− m+) recA::RP4–2-Tcr::Mu Kmr::Tn7 Tpr Smrλpir      Genome/gene sequences of other bacteria  Azotobacter vinelandii CA      GCA_000380335.1  Xanthomonas campestris 8004      GCA_000012105.1  Vibrio parahaemolyticus WP1      AB048250.2 and AB082123.1  Plasmids  pJP5603  Suicide plasmid, Kmr      pJP5603_lip323–783  pJP5603::disruption construct for gene CJU81_12870c, Kmr      pJP5603_pvuA886–1523  pJP5603::disruption construct for gene CJU81_14195c, Kmr      aRelevant GenBank and Sequence Read Archive accession numbers provided. bPreviously sequenced P. fragi genomes that were examined for the presence of siderophore biosynthetic gene clusters. cGenomes sequenced in this study. dLocus tags of respective coding sequences. View Large Conditions of iron starvation were achieved in modified M9 media (MM9) comprising 10% v/v of MM9 salts (5 g/L NaCl, 10 g/L NH4Cl, 0.59 g/L Na2HPO4.H2O and 0.45 g/L KH2PO4), 2 mM MgSO4, 0.1 mM CaCl2, 0.2% w/v glucose, 0.3% casamino acids w/v, 0.2% w/v succinate and 0.1 M PIPES (pH 6.8, NaOH). Pseudomonas fragi cultures were grown under agitation unless otherwise specified. Kanamycin was added to culture media of gene disruption mutants at a concentration of 50 μg/ml. Identification of siderophore gene cluster Examination of the 13 previously sequenced P. fragi genomes (Table 1) for siderophore biosynthetic gene clusters was performed with antiSMASH version 4 (Blin et al.2017). Generation of disruption mutants, genomic DNA isolation, and genome sequencing and analysis A gene disruption mutant was generated of the TonB-dependent outer membrane siderophore receptor (ΔpvuA) by homologous recombination with a pJP5603 suicide vector construct. This system results in a single cross-over event, whereby the pJP5603 vector (Riedel et al.2013) encoding a kanamycin resistance gene remains integrated in the genome (Penfold and Pemberton 1992). Thus, a control strain with wild-type vibrioferrin genes and a disrupted lipase gene (Δlip) was generated for subsequent experimental comparisons. The lip gene was chosen because the disruption of this gene would not affect iron metabolism of the strain, nor would it be required for growth of the strain in the culture media used in this work. Furthermore, because it is present as single gene rather than in an operon cluster, polar effects on downstream genes were highly unlikely. DNA manipulations were performed using standard protocols (Sambrook and Russell 2001). Plasmid constructs were generated by ligating BamHI-digested pJP5603 and BamHI-restriction fragments of the P. fragi F1801 lip and pvuA genes (Table 1), which were amplified using primers included in Table S1 (Supporting Information). pJP5603 gene disruption constructs were transferred from E. coli S17–1 λpir (Simon, Priefer and Puhler 1983) to P. fragi F1801 by biparental filter matings as previously described (Windgassen, Urban and Jaeger 2000) with some adjustments; recipient cells were grown at 30°C and the ratio of donor to recipient cells was 1:2 (5 × 108 CFU to 109 CFU), respectively. After mating, cells were spread on Pseudomonas agar containing cetrimide-fucidin-cephaloridine selective supplement (Oxoid) for Pseudomonas and 50 μg/mL kanamycin sulphate to select for transconjugants. Plates were incubated at 25°C for 30 h. Strains (wild-type F1801, Δlip and ΔpvuA) were grown for 18 h in tryptone soya broth (Oxoid) at 25°C. Genomic DNA was isolated with the DNeasy Blood and Tissue Kit (QIAGEN) according to the manufacturer's protocol for Gram-negative bacteria. Library preparation and genome sequencing were carried out at Queensland Health (Health Support Queensland) using MiniSeq High Output kits with 150 cycles, and the Illumina MiniSeq System (Illumina). Genome assembly was performed as previously described (Stanborough et al.2017). Confirmation of correct insertion of the vector was achieved by manual inspection of the genomes with Geneious version 9.0.5 (Kearse et al.2012). Snippy (https://github.com/tseemann/snippy) was used to confirm an absence of additional genetic variants in the Δlip and ΔpvuA mutants’ genomes by aligning sequence reads of wild-type F1801, ΔpvuA and Δlip strains to the reference genome of F1801. Sequence reads of wild-type F1801, Δlip and ΔpvuA strains were deposited in the Sequence Read Archive, and the respective accession numbers are provided in Table 1. Phylogenetic tree of vibrioferrin sequences MAFFT multiple sequence alignments (Katoh et al.2002) of the vibrioferrin PvuA and PvsA-PvsE protein sequences were performed using the default settings. The six multiple alignments were concatenated and Gblocks version 0.91b (Castresana 2000; Talavera and Castresana 2007) was run with the default settings to remove poorly aligned positions and divergent regions of the alignments. Bayesian inference (BI), performed with MrBayes version 3.2.6 (Ronquist and Huelsenbeck 2003), was conducted with a run of 1000 000 generations and sampling every 1000. A mixed amino acid model analysis was set, enabling the MCMC sampler to test all of the fixed rate models. MrBayes determined that the Wheeler and Goldman model of amino acid replacement (Whelan and Goldman 2001) had the best likelihood score and was chosen for the analysis. Convergence parameters were assessed using Tracer version 1.6 (Rambaut et al.2014), and the majority rule consensus tree was rendered with Figtree version 1.4.2 (http://tree.bio.ed.ac.uk/software/figtree/). BIs were run three times to ensure reproducibility of the resulting trees. RNA isolation and quantitative reverse transcription-polymerase chain reaction RNA was isolated from 1 ml of mid-log phase iron-starved cultures (1.5 × 108 CFU). Stabilisation of RNA was achieved with RNAprotect Bacteria Reagent (QIAGEN) following the manufacturer's recommendations, and total RNA was extracted with the RNeasy Mini Kit (QIAGEN) according to the supplier's instructions and stored in H2O at –80°C. cDNA was synthesised from 1 μg of total RNA using iScript gDNA Clear cDNA Synthesis Kit (Biorad) as suggested by the manufacturer; however, DNase treatment was increased from the recommended 5 to 30 min. Quantitative reverse transcription-polymerase chain reaction (qRT-PCR) was performed on an AriaMx Real-time PCR System (Agilent) using the iTaq Universal SYBR Green Supermix (Biorad) as a fluorescence source. PCR reaction mixes included 10 μl iTaq Universal SYBR Green Supermix (Biorad), 500 nM forward and reverse primers, and 1 μl template cDNA (diluted 1:10 in H2O) in a total volume of 20 μL. Primer sequences, provided in Table S1 (Supporting Information), were designed with Primer3 version 2.3.4 (Untergasser et al.2012). Acceptable primer efficiency was confirmed for each target gene with standard curves, and melt curve analysis of amplicons of each target gene was performed to ensure amplification of single gene products. PCR conditions were as follows: 30 s at 95°C followed by 40 cycles of 5 s at 95°C and 30 s at 60°C. Melt curve analysis involved an incremental increase of 0.5°C every 2 s from 65°C to 95°C. Each assay included a non-template control, and each sample a control without reverse transcriptase. Raw data were analysed with Agilent AriaMx version 1.0 software and exported to excel for further analysis. Ct values were normalised to the isocitrate dehydrogenase gene (gene locus tag CJU81_09775 in F1801, GenBank accession no. GCA_002269505.1), and relative expression levels were calculated according to the 2ΔΔCt method. Chromeazurol S supernatant assays Chromeazurol S (CAS) assay solution was prepared as previously described (Alexander and Zuberer 1991). Siderophore detection from culture supernatants was performed using the method of Schwyn and Neilands (1987). Briefly, bacterial cultures were spun down at 2500 × g for 5 min and the supernatants were filtered through a 0.20-μm cellulose acetate membrane filter (Sartorius). Siderophores were detected in the supernatants by mixing equal volumes of filtered supernatant and CAS solution. After allowing the solutions to equilibrate for 1.5 h, a visible change in the colour of the mixture from blue to orange was considered a positive reaction. To detect siderophore production at 25°C, strains were grown in 4 ml MM9 for 18 h prior to supernatant collection. To monitor siderophore production at 4°C, F1801 was grown in 50 ml MM9 for 4 days and every 24 h 3 ml of culture were removed to measure growth (A600), and detect siderophores by CAS supernatant assays. To monitor the increase in siderophore levels over the 4 days, absorbance measurements of the CAS reaction mixtures were performed at 630 nm. Growth experiments with 2,2΄-bipyridyl and bovine apo-transferrin Single colonies were used to inoculate 3 ml MM9. Cultures were grown at 25°C for 18 h. Twenty microlitres of these cultures were used to inoculate 180 μl fresh MM9, supplemented as indicated in the text with 2,2΄-bipyridyl (Sigma-Aldrich), bovine apo-transferrin (Sigma-Aldrich) and/or sodium bicarbonate and FeCl3, in Greiner 96-well flat bottom microtitre plates (Sigma-Aldrich). Growth was monitored (A600) using an EON microplate reader (BioTek). RESULTS AND DISCUSSION In this work, 13 previously sequenced P. fragi genomes were examined for the presence of siderophore biosynthetic gene clusters. This search resulted in the identification of a conserved gene cluster, homologues of which were shown to be responsible for vibrioferrin biosynthesis and vibrioferrin-mediated iron uptake in the terrestrial bacteria Azotobacter vinelandii and Xanthomonas spp., and in the marine bacterium Vibrio parahaemolyticus (Yamamoto et al.1994; Pandey and Sonti 2010; Baars et al.2016; Pandey et al.2017) (Fig. 1). Similar to the vibrioferrin gene clusters in A. vinelandii and Xanthomonas campestris, this ∼10 kb region in P. fragi comprised seven open reading frames (Fig. 1A). Previous work in Xanthomonas spp. and V. parahaemolyticus showed that five of these genes (mhpE, pvsA, pvsB, pvsD and pvsE) are involved in the biosynthesis of vibrioferrin, while pvuA is a TonB-dependent outer membrane receptor and pvsC an inner membrane exporter for vibrioferrin (Funahashi et al.2002; Tanabe et al.2006; Pandey and Sonti 2010; Fujita et al.2011). Locus tags of these genes in the 13 strains are shown in Table 1. Analysis of the upstream region of this gene cluster revealed the presence of a 19-bp sequence, matching 14 of the 19 nucleotides of the X. campestris consensus Fur box (Blanvillain et al.2007) and 11 of the 19 nucleotides of the E. coli and P. aeruginosa consensus Fur box. This sequence was located 129 bases upstream of the mhpE start codon. Fur homologues were also identified in the P. fragi genomes with the locus tag CJU81_16820 in F1801. Figure 1. View largeDownload slide Vibrioferrin gene cluster identified in 13 P. fragi. (A) Schematic diagram of the vibrioferrin gene cluster identified in P. fragi and conserved synteny of these genes in known vibrioferrin producing bacteria A. vinelandii, X. campestris and V. parahaemolyticus. Predicted biosynthesis genes are marked with an asterisk in the P. fragi genes. Numbers shown in genes indicate percent sequence identity of proteins to respective homologues in P. fragi F1801, as determined by MAFFT protein alignment. The putative Fur box identified upstream of the first gene in the P. fragi cluster is shown and consensus Fur box sequences of X. campestris and E. coli (identical to the P. aeruginosa consensus Fur box) are included for comparison. (B) Mid-point rooted phylogram of the consensus BI tree of the PvuA and PvsA-PvsE sequences of the 13 P. fragi, A. vinelandii, X. campestris and V. parahaemolyticus. Pseudomonas fragi F1801 that was used in subsequent experiments is marked with an asterisk. The subtree shown in the box represents BI branching on a different scale to the tree. Posterior probability values are shown as percentages at the nodes, and scale bars denote the number of substitutions per site. Figure 1. View largeDownload slide Vibrioferrin gene cluster identified in 13 P. fragi. (A) Schematic diagram of the vibrioferrin gene cluster identified in P. fragi and conserved synteny of these genes in known vibrioferrin producing bacteria A. vinelandii, X. campestris and V. parahaemolyticus. Predicted biosynthesis genes are marked with an asterisk in the P. fragi genes. Numbers shown in genes indicate percent sequence identity of proteins to respective homologues in P. fragi F1801, as determined by MAFFT protein alignment. The putative Fur box identified upstream of the first gene in the P. fragi cluster is shown and consensus Fur box sequences of X. campestris and E. coli (identical to the P. aeruginosa consensus Fur box) are included for comparison. (B) Mid-point rooted phylogram of the consensus BI tree of the PvuA and PvsA-PvsE sequences of the 13 P. fragi, A. vinelandii, X. campestris and V. parahaemolyticus. Pseudomonas fragi F1801 that was used in subsequent experiments is marked with an asterisk. The subtree shown in the box represents BI branching on a different scale to the tree. Posterior probability values are shown as percentages at the nodes, and scale bars denote the number of substitutions per site. To investigate the relatedness of the vibrioferrin genes between the 13 strains and A. vinelandii, X. campestris and V. parahaemolyticus, a phylogenetic tree of the PvuA and PvsA-PvsE sequences was determined by BI (Fig. 1B). Topology of the BI consensus tree showed three clades of P. fragi vibrioferrin sequences. The largest of the three clades, comprising closely related sequences of 8 of the 13 strains, was clearly separated from the other two clades, which share a more recent common ancestral sequence. The sequences of the P. fragi type strain were found in one of these two clades with those of two additional strains. Sequences of F1801, the P. fragi strain used for subsequent experiments, were grouped together in the second of these clades with vibrioferrin sequences of one other strain. In accordance with their phylogeny, sequences of the P. fragi were closest to those of A. vinelandii, while those of X. campestris and V. parahaemolyticus were substantially more distant. Previous work suggested that P. fragi does not synthesise siderophores (Champomier-Vergès, Stintzi and Meyer 1996), and utilising the succinate media (non-deferrated and deferrated forms) presented in the work, P. fragi F1801 displayed little to no growth and siderophore production of this strain was not observed (data not shown). Therefore, in this study, a non-deferrated, modified M9 medium (MM9) was used containing glucose that was not present in the media used by Champomier-Vergès et al. (1996). MM9 enabled both satisfactory growth, as demonstrated by growth of F1801 at 25°C, while still ensuring conditions of iron starvation, which can be seen by the strong increase in growth of F1801 in MM9 supplemented with 50 μM FeCl3 (Fig. 2A). Figure 2. View largeDownload slide Siderophore production by P. fragi F1801 at 25°C and 4°C. (A) Growth of F1801 in MM9 or in MM9 that was supplemented with 50 μM FeCl3. Optical density of the cultures was measured at 600 nm every hour in an EON microplate reader (BioTek). Data shown are the means and error bars the standard deviations of three biological replicates. (B) CAS assay results of F1801. Shown from left to right is CAS solution that was added to sterile MM9 (control), filtered culture supernatant of F1801 grown in MM9 at 25°C (F1801), sterile MM9 supplemented with 50 μM FeCl3 (control (+FeCl3)) and filtered culture supernatant of F1801 grown at 25°C in MM9 supplemented with 50 μM FeCl3 (F1801 (+FeCl3)). (C) Relative quantification by qRT-PCR of the vibrioferrin genes in F1801 under iron-replete conditions (MM9 supplemented with 50 μM FeCl3) compared to conditions of iron starvation (MM9). Data shown are the means and error bars the standard deviations of three biological replicates. (D) Growth of and siderophore production by F1801 in MM9 at 4°C. Optical density of the cultures and CAS assay absorbance measurements were carried out with a Novaspec Plus visible spectrophotometer (VWR). Data shown are the means and error bars the standard deviations of three biological replicates. Figure 2. View largeDownload slide Siderophore production by P. fragi F1801 at 25°C and 4°C. (A) Growth of F1801 in MM9 or in MM9 that was supplemented with 50 μM FeCl3. Optical density of the cultures was measured at 600 nm every hour in an EON microplate reader (BioTek). Data shown are the means and error bars the standard deviations of three biological replicates. (B) CAS assay results of F1801. Shown from left to right is CAS solution that was added to sterile MM9 (control), filtered culture supernatant of F1801 grown in MM9 at 25°C (F1801), sterile MM9 supplemented with 50 μM FeCl3 (control (+FeCl3)) and filtered culture supernatant of F1801 grown at 25°C in MM9 supplemented with 50 μM FeCl3 (F1801 (+FeCl3)). (C) Relative quantification by qRT-PCR of the vibrioferrin genes in F1801 under iron-replete conditions (MM9 supplemented with 50 μM FeCl3) compared to conditions of iron starvation (MM9). Data shown are the means and error bars the standard deviations of three biological replicates. (D) Growth of and siderophore production by F1801 in MM9 at 4°C. Optical density of the cultures and CAS assay absorbance measurements were carried out with a Novaspec Plus visible spectrophotometer (VWR). Data shown are the means and error bars the standard deviations of three biological replicates. The sensitive CAS supernatant assay was used to detect the presence of siderophores in culture supernatants of F1801 grown at optimal growth temperature (25°C) in MM9. The CAS assay is based on the removal of iron from the chromogenic dye CAS by siderophores, whereby a colour change is observed from blue to orange (Schwyn and Neilands 1987). Figure 2B shows that siderophores were clearly detected in cell-free supernatants of iron-starved cultures of F1801. Reaction times for the exchange of iron from CAS were within 1.5 h, a time-frame consistent with carboxylate-type siderophores (Schwyn and Neilands 1987). In contrast, when F1801 was grown in MM9 supplemented with 50 mM FeCl3, siderophores were largely absent from culture supernatants (Fig. 2B). Furthermore, expression of the vibrioferrin genes was strongly downregulated (between 12-fold (pvsE) and 120-fold (pvsD) reduction) relative to their expression in MM9 without iron supplementation (Fig. 2C), indicating repression of siderophore production under iron-rich conditions. These results are in agreement with other studies, which showed that vibrioferrin production in A. vinelandii, Xanthomonas and V. parahaemolyticus increased under iron limitation, while under iron-replete conditions vibrioferrin production ceased (Funahashi et al.2000; Pandey and Sonti 2010; McRose et al.2017; Pandey et al.2017). Thus, together with the presence of a putative Fur box upstream of the vibrioferrin gene cluster, these results suggest Fur-mediated regulation of the P. fragi vibrioferrin genes. As a key meat and milk spoilage bacterium, P. fragi F1801 was tested for siderophore production at refrigeration temperature by monitoring growth and the accumulation of siderophores in supernatants at 4°C over 4 days. Blue-coloured CAS solution has a maximum absorption at 630 nm. Orange-coloured CAS reaction mixtures, in which siderophores have removed iron from CAS, have essentially no absorbance at this wavelength and there is a largely linear dependence of A630 values of CAS reaction mixtures versus concentration of the chelator (Schwyn and Neilands 1987). Figure 2D shows a steady decrease in A630 values of CAS reaction mixtures over the 4 days, indicating an accumulation of siderophores in culture supernatants of F1801. This accumulation correlated well with cell growth of the strain, demonstrating siderophore production of P. fragi at temperatures relevant to the storage of meat and milk. To confirm a role for the vibrioferrin gene cluster in siderophore-mediated iron acquisition, a gene disruption mutant of the TonB-dependent outer membrane siderophore receptor (ΔpvuA) was generated, as well as a control strain with wild-type vibrioferrin genes and a disrupted lipase gene (Δlip) for subsequent experimental comparisons. qRT-PCR analysis revealed that expression of genes downstream of pvuA in the ΔpvuA mutant was strongly reduced (between 7-fold (pvsE) and 128-fold (pvsA) reduction) relative to the Δlip control strain (Fig. 3A), indicating polar effects caused by integration of the suicide vector in the pvuA gene. As our aim was to confirm a role for this gene cluster in siderophore-mediated iron uptake, rather than to study the function of the vibrioferrin genes, which has been addressed in other studies (Funahashi et al.2002; Tanabe et al.2003, 2006; Pandey and Sonti 2010; Fujita et al.2011; Pandey et al.2017), the ΔpvuA mutant was used for further investigations. Figure 3. View largeDownload slide ΔpvuA shows impaired siderophore production. (A) Relative quantification by qRT-PCR of expression of the vibrioferrin genes in ΔpvuA compared with the Δlip control strain. Data shown are the means and error bars the standard deviations of three biological replicates. (B) CAS assay results of the ΔpvuA mutant and Δlip control strain. Shown from left to right is CAS solution that was added to sterile MM9 (control), filtered culture supernatant of Δlip grown in MM9 at 25°C and filtered culture supernatant of ΔpvuA grown in MM9 at 25°C. Figure 3. View largeDownload slide ΔpvuA shows impaired siderophore production. (A) Relative quantification by qRT-PCR of expression of the vibrioferrin genes in ΔpvuA compared with the Δlip control strain. Data shown are the means and error bars the standard deviations of three biological replicates. (B) CAS assay results of the ΔpvuA mutant and Δlip control strain. Shown from left to right is CAS solution that was added to sterile MM9 (control), filtered culture supernatant of Δlip grown in MM9 at 25°C and filtered culture supernatant of ΔpvuA grown in MM9 at 25°C. A strong reduction in siderophore levels was detected in culture supernatants of ΔpvuA compared to the Δlip control strain (Fig. 3B). Gene deletion of pvuA, which encodes an outer membrane vibrioferrin receptor, should result in an accumulation of siderophores in culture supernatants. Therefore, the observed reduction was likely due to the polar effects on the downstream biosynthesis genes. Genetic variants were not found in genes or in non-coding regions associated with the vibrioferrin genes or other iron-related genes of the ΔpvuA mutant's genome (Table S2, Supporting Information), demonstrating that phenotypes of the ΔpvuA mutant were solely due to the disruption of the target gene and associated polar effects. A clear growth defect of the ΔpvuA mutant was not observed in MM9 when compared to the Δlip control strain (Fig. 4A). Therefore, to understand the role of the vibrioferrin siderophore under low iron conditions, growth of the ΔpvuA and Δlip strains was tested under more severe iron-depleted conditions by adding 25 μM of the iron chelator 2,2΄-bipyridyl (BP) to MM9. While growth of the Δlip control strain was unaffected at 25 μM BP, addition of the ferrous iron chelator at these concentrations markedly impaired growth of the ΔpvuA mutant, demonstrating that the siderophore is involved in iron acquisition under low iron conditions. Figure 4. View largeDownload slide Vibrioferrin siderophore plays a role under iron starvation and scavenges iron from transferrin. Growth of ΔpvuA and Δlip control strain in MM9, and in MM9 supplemented with 25 μM 2,2΄-bipyridyl (BP) (A); in MM9, and in MM9 supplemented with 20 mM bicarbonate (BC) (B); in MM9 supplemented with 2.5 μM bovine apo-transferrin (apo-Tsf) and 20 mM BC (C); and in MM9 supplemented with 2.5 μM apo-Tsf, 20 mM BC and 100 μM FeCl3 (D). Data shown for growth experiments in (A–D) are the means and error bars the standard deviations of three biological replicates. All experiments were repeated a minimum of three times with similar results observed. Figure 4. View largeDownload slide Vibrioferrin siderophore plays a role under iron starvation and scavenges iron from transferrin. Growth of ΔpvuA and Δlip control strain in MM9, and in MM9 supplemented with 25 μM 2,2΄-bipyridyl (BP) (A); in MM9, and in MM9 supplemented with 20 mM bicarbonate (BC) (B); in MM9 supplemented with 2.5 μM bovine apo-transferrin (apo-Tsf) and 20 mM BC (C); and in MM9 supplemented with 2.5 μM apo-Tsf, 20 mM BC and 100 μM FeCl3 (D). Data shown for growth experiments in (A–D) are the means and error bars the standard deviations of three biological replicates. All experiments were repeated a minimum of three times with similar results observed. To investigate whether the vibrioferrin siderophore could play a role in iron metabolism of P. fragi in foods such as meat and milk, where iron is sequestered by high-affinity iron-binding proteins, growth of ΔpvuA and Δlip strains was compared in MM9, to which bovine apo-transferrin (apo-Tsf), as a model transferrin protein, was added. In all transferrin proteins, the carbonate (or bicarbonate) of the binding cleft is the synergistic anion, without which the protein loses its affinity for iron (Pakdaman and El Hage Chahine 1997). Thus, to ensure the specific binding of Fe3+ by apo-Tsf, the growth media was also supplemented with 20 mM bicarbonate (BC) (Chung 1984). BC at a concentration of 20 mM had no effect on the strains’ growth when compared to their growth in MM9 without BC addition (Fig. 4B). When the growth medium was supplemented with apo-Tsf at a concentration of 2.5 μM in the presence of 20 mM BC, growth, although slowed, was observed for the Δlip control strain (Fig. 4C). In contrast, growth of ΔpvuA was completely inhibited in the presence of apo-Tsf and BC, but could be rescued by supplementation of this media with 100 μM FeCl3 (Fig. 4D). These data suggest that the vibrioferrin siderophore may be capable of removing ferric iron from transferrin proteins. Remarkably, despite considerable divergence between the six vibrioferrin protein sequences shared by A. vinelandii and V. parahaemolyticus, the absence of an mhpE gene in V. parahaemolyticus, and the absence of the vibrioferrin-uptake genes pvuA(1)-pvuE in A. vinelandii, both bacteria were shown to produce identical vibrioferrin compounds (Yamamoto et al.1994; Baars et al.2016). Based on percent identities and the evolutionary relationship of the P. fragi vibrioferrin sequences with A. vinelandii and V. parahaemolyticus counterparts, it is likely that the P. fragi siderophore is an α-hydroxycarboxylate type siderophore, similar to or identical to vibrioferrin. CONCLUSION A vibrioferrin gene cluster was identified in 13 strains of the food spoilage bacterium P. fragi, and siderophore production of P. fragi F1801 was observed. Based on the homology of the identified genes, P. fragi likely excretes an α-hydroxycarboxylate type siderophore, similar to or identical to vibrioferrin, which plays a role in iron metabolism of this bacterium under low iron conditions. More experiments are required to determine whether the siderophore contributes to the successful growth and colonisation of milk and meat by P fragi; however, these data suggest that the vibrioferrin siderophore is capable of removing ferric iron from transferrin proteins. SUPPLEMENTARY DATA Supplementary data are available at FEMSLE online. Acknowledgements We thank Steve Petrovski for providing plasmid pJP5603 and Escherichia coli strain S17–1 λpir. FUNDING This work was supported by the Australian Meat Processor Corporation (grant number 2013-5041) and the Commonwealth Scientific and Industrial Research Organisation (grant number OD-106390). Conflicts of interest. None declared. REFERENCES Alexander DB, Zuberer DA. Use of chrome azurol S reagents to evaluate siderophore production by rhizosphere bacteria. Biol Fert Soils  1991; 12: 39– 45. Google Scholar CrossRef Search ADS   Baars O, Zhang X, Morel FMM et al.   The siderophore metabolome of Azotobacter vinelandii. Appl Environ Microb  2016; 82: 27– 39. Google Scholar CrossRef Search ADS   Blanvillain S, Meyer D, Boulanger A et al.   Plant carbohydrate scavenging through TonB-dependent receptors: A feature shared by phytopathogenic and aquatic bacteria. PLoS One  2007; 2: e224. Google Scholar CrossRef Search ADS PubMed  Blin K, Wolf T, Chevrette MG et al.   antiSMASH 4.0-improvements in chemistry prediction and gene cluster boundary identification. Nucleic Acids Res  2017, DOI 10.1093/nar/gkx319. Carpenter BM, Whitmire JM, Merrell DS. This is not your mother's repressor: the complex role of Fur in pathogenesis. Infect Immun  2009; 77: 2590– 601. Google Scholar CrossRef Search ADS PubMed  Casaburi A, Piombino P, Nychas G-J et al.   Bacterial populations and the volatilome associated to meat spoilage. Food Microbiol  2015; 45( Part A): 83– 102. Google Scholar CrossRef Search ADS PubMed  Castresana J. Selection of conserved blocks from multiple alignments for their use in phylogenetic analysis. Mol Biol Evol  2000; 17: 540– 52. Google Scholar CrossRef Search ADS PubMed  Cezard C, Farvacques N, Sonnet P. Chemistry and biology of pyoverdines, Pseudomonas primary siderophores. Curr Med Chem  2015; 22: 165– 86. Google Scholar CrossRef Search ADS PubMed  Champomier-Vergès M-C, Stintzi A, Meyer J-M. Acquisition of iron by the non-siderophore-producing Pseudomonas fragi. Microbiology  1996; 142: 1191– 9. Google Scholar CrossRef Search ADS PubMed  Chung MC-M. Structure and function of transferrin. Biochem Educ  1984; 12: 146– 54. Google Scholar CrossRef Search ADS   Colombo C, Palumbo G, He J-Z et al.   Review on iron availability in soil: Interaction of Fe minerals, plants, and microbes. J Soils Sed  2014; 14: 538– 48. Google Scholar CrossRef Search ADS   Cox CD, Adams P. Siderophore activity of pyoverdin for Pseudomonas aeruginosa. Infect Immun  1985; 48: 130– 8. Google Scholar PubMed  De Jonghe V, Coorevits A, Van Hoorde K et al.   Influence of storage conditions on the growth of Pseudomonas species in refrigerated raw milk. Appl Environ Microb  2011; 77: 460– 70. Google Scholar CrossRef Search ADS   Faraldo-Gomez JD, Sansom MSP. Acquisition of siderophores in Gram-negative bacteria. Nat Rev Mol Cell Bio  2003; 4: 105– 16. Google Scholar CrossRef Search ADS   Farh ME-A, Kim Y-J, Sukweenadhi J et al.   Aluminium resistant, plant growth promoting bacteria induce overexpression of Aluminium stress related genes in Arabidopsis thaliana and increase the ginseng tolerance against Aluminium stress. Microbiol Res  2017; 200: 45– 52. Google Scholar CrossRef Search ADS PubMed  Fujita MJ, Kimura N, Sakai A et al.   Cloning and heterologous expression of the vibrioferrin biosynthetic gene cluster from a marine metagenomic library. Biosci Biotech Bioch  2011; 75: 2283– 7. Google Scholar CrossRef Search ADS   Funahashi T, Fujiwara C, Okada M et al.   Characterization of Vibrio parahaemolyticus manganese-resistant mutants in reference to the function of the ferric uptake regulatory protein. Microbiol Immunol  2000; 44: 963– 70. Google Scholar CrossRef Search ADS PubMed  Funahashi T, Moriya K, Uemura S et al.   Identification and characterization of pvuA, a gene encoding the ferric vibrioferrin receptor protein in Vibrio parahaemolyticus. J Bacteriol  2002; 184: 936– 46. Google Scholar CrossRef Search ADS PubMed  Garrity GM, Bell JA, Lilburn T. Pseudomonadales Orla-Jensen 1921, 270AL. In: Brenner DJ, Krieg NR, Staley JT et al.   (eds.) Bergey's Manual® of Systematic Bacteriology: Volume Two The Proteobacteria Part B The Gammaproteobacteria . Boston, MA: Springer, 2005, 323– 442. Google Scholar CrossRef Search ADS   Ilbert M, Bonnefoy V. Insight into the evolution of the iron oxidation pathways. Biochim Biophys Acta  2013; 1827: 161– 75. Google Scholar CrossRef Search ADS PubMed  Jones CM, Niederweis M. Role of porins in iron uptake by Mycobacterium smegmatis. J Bacteriol  2010; 192: 6411– 7. Google Scholar CrossRef Search ADS PubMed  Katoh K, Misawa K, Kuma K-i et al.   MAFFT: a novel method for rapid multiple sequence alignment based on fast Fourier transform. Nucleic Acids Res  2002; 30: 3059– 66. Google Scholar CrossRef Search ADS PubMed  Kearse M, Moir R, Wilson A et al.   Geneious Basic: an integrated and extendable desktop software platform for the organization and analysis of sequence data. Bioinformatics  2012; 28: 1647– 9. Google Scholar CrossRef Search ADS PubMed  McRose DL, Baars O, Morel FMM et al.   Siderophore production in Azotobacter vinelandii in response to Fe-, Mo- and V-limitation. Environ Microbiol  2017; 19: 3595– 605. Google Scholar CrossRef Search ADS PubMed  Meyer JM, Abdallah MA. The fluorescent pigment of Pseudomonas fluorescens: biosynthesis, purification and physicochemical properties. Microbiology  1978; 107: 319– 28. Noinaj J, Guillier M, Travis J. Barnard et al.   TonB-dependent transporters: regulation, structure, and function. Annu Rev Microbiol  2010; 64: 43– 60. Google Scholar CrossRef Search ADS PubMed  Pakdaman R, El Hage Chahine J-M. Transferrin. Eur J Biochem  1997; 249: 149– 55. Google Scholar CrossRef Search ADS PubMed  Pandey A, Sonti RV. Role of the FeoB protein and siderophore in promoting virulence of Xanthomonas oryzae pv. oryzae on rice. J Bacteriol  2010; 192: 3187– 203. Google Scholar CrossRef Search ADS PubMed  Pandey SS, Patnana PK, Rai R et al.   Xanthoferrin, the alpha-hydroxycarboxylate-type siderophore of Xanthomonas campestris pv. campestris, is required for optimum virulence and growth inside cabbage. Mol Plant Pathol  2017; 18: 949– 62. Google Scholar CrossRef Search ADS PubMed  Penfold RJ, Pemberton JM. An improved suicide vector for construction of chromosomal insertion mutations in bacteria. Gene  1992; 118: 145– 6. Google Scholar CrossRef Search ADS PubMed  Rambaut A, Suchard MA, Xie D et al.   Tracer v1.6 . 2014 http://tree.bio.ed.ac.uk/software/tracer/. Riedel T, Rohlfs M, Buchholz I et al.   Complete sequence of the suicide vector pJP5603. Plasmid  2013; 69: 104– 7. Google Scholar CrossRef Search ADS PubMed  Ronquist F, Huelsenbeck JP. MrBayes 3: Bayesian phylogenetic inference under mixed models. Bioinformatics  2003; 19: 1572– 4. Google Scholar CrossRef Search ADS PubMed  Sambrook J, Russell DW. Molecular Cloning: A Laboratory Manual . Cold Spring Harbor, NY: Cold Spring Harbor Laboratory, 2001. Schwyn B, Neilands JB. Universal chemical assay for the detection and determination of siderophores. Anal Biochem  1987; 160: 47– 56. Google Scholar CrossRef Search ADS PubMed  Simon R, Priefer U, Puhler A. A broad host range mobilization system for in vivo genetic engineering: transposon mutagenesis in gram negative bacteria. Nat Biotech  1983; 1: 784– 91. Google Scholar CrossRef Search ADS   Skaar EP. The battle for iron between bacterial pathogens and their vertebrate hosts. PLoS Pathog  2010; 6: e1000949. Google Scholar CrossRef Search ADS PubMed  Stanborough T, Fegan N, Powell SM et al.   Insight into the genome of Brochothrix thermosphacta, a problematic meat spoilage bacterium. Appl Environ Microb  2017; 83: e02786– 16. Google Scholar CrossRef Search ADS   Talavera G, Castresana J. Improvement of phylogenies after removing divergent and ambiguously aligned blocks from protein sequence alignments. Syst Biol  2007; 56: 564– 77. Google Scholar CrossRef Search ADS PubMed  Tanabe T, Funahashi T, Nakao H et al.   Identification and characterization of genes required for biosynthesis and transport of the siderophore vibrioferrin in Vibrio parahaemolyticus. J Bacteriol  2003; 185: 6938– 49. Google Scholar CrossRef Search ADS PubMed  Tanabe T, Nakao H, Kuroda T et al.   Involvement of the Vibrio parahaemolyticus pvsC gene in export of the siderophore vibrioferrin. Microbiol Immunol  2006; 50: 871– 6. Google Scholar PubMed  Trapet P, Avoscan L, Klinguer A et al.   The Pseudomonas fluorescens siderophore pyoverdine weakens Arabidopsis thaliana defense in favour of growth in iron-deficient conditions. Plant Physiol  2016, DOI 10.1104/pp.15.01537. Troxell B, Hassan HM. Transcriptional regulation by Ferric Uptake Regulator (Fur) in pathogenic bacteria. Front Cell Infect Microbiol  2013; 3: 59. Google Scholar PubMed  Untergasser A, Cutcutache I, Koressaar T et al.   Primer3–new capabilities and interfaces. Nucleic Acids Res  2012; 40: e115. Google Scholar CrossRef Search ADS PubMed  Whelan S, Goldman N. A general empirical model of protein evolution derived from multiple protein families using a maximum-likelihood approach. Mol Biol Evol  2001; 18: 691– 9. Google Scholar CrossRef Search ADS PubMed  Windgassen M, Urban A, Jaeger K-E. Rapid gene inactivation in Pseudomonas aeruginosa. FEMS Microbiol Lett  2000; 193: 201– 5. Google Scholar CrossRef Search ADS PubMed  Yamamoto S, Okujo N, Yoshida T et al.   Structure and iron transport activity of vibrioferrin, a new siderophore of Vibrio parahaemolyticus. J Biochem  1994; 115: 868– 74. Google Scholar CrossRef Search ADS PubMed  © FEMS 2017. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com

Journal

FEMS Microbiology LettersOxford University Press

Published: Mar 1, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 12 million articles from more than
10,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Unlimited reading

Read as many articles as you need. Full articles with original layout, charts and figures. Read online, from anywhere.

Stay up to date

Keep up with your field with Personalized Recommendations and Follow Journals to get automatic updates.

Organize your research

It’s easy to organize your research with our built-in tools.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

Monthly Plan

  • Read unlimited articles
  • Personalized recommendations
  • No expiration
  • Print 20 pages per month
  • 20% off on PDF purchases
  • Organize your research
  • Get updates on your journals and topic searches

$49/month

Start Free Trial

14-day Free Trial

Best Deal — 39% off

Annual Plan

  • All the features of the Professional Plan, but for 39% off!
  • Billed annually
  • No expiration
  • For the normal price of 10 articles elsewhere, you get one full year of unlimited access to articles.

$588

$360/year

billed annually
Start Free Trial

14-day Free Trial