Bacterial strains, plasmids and media

Bacterial strains and plasmids used in this study are listed in Table 1. One Shot^® TOP10 chemically competent E. coli cells were used for cloning reactions following manufacturer’s protocols (Invitrogen, Carlsbad, CA). Pss B728a and mutant strains were cultured from 20% glycerol stocks stored at -80°C onto nutrient broth-yeast extract (NBY) [30], or on King’s B (KB) [31] at 26°C. Bioassays for syringomycin were grown on Hrp minimal medium (HMM) agar [32, 33]. The following antibiotic concentractions (μg/ml) were added to media: rifampicin, 100; kanamycin, 75; tetracycline, 20; ampicillin, 100; gentamycin, 5; spectinomycin, 100.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Strains and plasmids.

https://doi.org/10.1371/journal.pone.0150234.t001

General DNA manipulations

For methodologies that involve the use of Gateway cloning technology ([42] and regulatory aspects of lambda site specific recombination), targeted genes were PCR amplified and cloned into the pENTR/D-TOPO vector following the manufacturer’s protocols (Invitrogen). Recombination between pENTR constructs and Gateway destination vectors was performed employing the use of LR clonase in accordance with the manufacturer’s protocol (Invitrogen). Plasmids were introduced into E. coli by chemical transformation or electroporation [43]. Plasmids were incorporated into P. syringae by tri-parental mating utilizing the helper plasmid pRK2073 [41]. Complementation of Pss B728a derivative mutants was achieved by the electroporation of the complement construct. Standard PCR procedures and cycling conditions were used [32, 36].

Restriction enzymes, and T4 DNA ligase were purchased from New England Biolabs (Beverly, MA). Phusion High-Fidelity DNA polymerase was purchased from Thermo Scientific Inc. (Waltham, MA). In-Fusion^® HD cloning kit was purchased from Clontech Laboratories (Mountain View, CA). The design and purchase of oligonucleotides was acquired using PrimerQuest and OligoAnalyzer applications of Integrated DNA technologies (Coralville, IA). The oligonucleotide sequences are listed in S1 Table. Plasmids were introduced into E. coli by chemical transformation or electroporation [43]. Plasmids were transferred to P. syringae by triparental mating using the helper plasmid pRK2073 [41]. Standard PCR procedures and cycling conditions were used.

Bioinformatic analysis

Protein sequences were retrieved using the Pseudomonas Genome Database [44]. The Conserved Domain Database at NCBI (http://www.ncbi.nlm.nih.gov/Structure/cdd/) was used to identify conserved domains of protein sequences. Additionally, database searches were performed using a Basic Local Alignment Search Tool (BLAST) to identify homologous sequences SyrG and SyrF in pseudomonad genomes. A clustalW alignment of homologous protein sequences was generated using the CLC Genomics Workbench (V5.5, CLC Bio.) [32].

Construction of markerless deletion mutants in Pss B728a

For targeted deletion mutants in Pss B728a, the gene of interest (GOI) along with 3 to 4-kb of flanking DNA was PCR amplified using Phusion^® high fidelity polymerase (ThermoScientific). The purified PCR product was cloned into a Gateway entry vector pENTR/D-TOPO (Invitrogen) and transformed chemically into E. coli One Shot^® TOP10 cells. LR clonase II (Invitrogen) was used to carry out recombination between the pENTR construct and the Pseudomonas suicide vector, pLVC-D [38].

Site-directed mutagenesis occurred by linearization of the pLVC-D plasmid (pLVC-D:flank-GOI-flank) using inverse PCR with primers that exclude the GOI, and purified using a Wizard^® SV Gel and PCR Clean-Up System (Promega, Madison, WI). A linear kanamycin cassette (nptII) flanked by the FLP recognition target sites, was amplified from pKD13 plasmid using primers with 15 bp extensions that were homologous to regions adjacent to the GOI [37]. The kanamycin cassette was cloned into the purified linearized pLVC-D construct using the In-Fusion^® HD cloning kit (Clontech) according to the manufacturer’s protocol and chemically transformed into E. coli One Shot^® TOP10 cells for confirmation of the construct. The resulting pLVC-D construct (pLVC-D:flank-nptII-flank) was moved into Pss B728a by triparental mating with the helper plasmid pRK2073 [41]. Colony PCR and qRT-PCR was used to confirm double recombination of the kanamycin cassette into Pss B728a, replacing the GOI. The kanamycin marker was later removed by the introduction of the pBH474 vector carrying the FLP recombinase gene. FLP recombination resulted in the loss of the nptII marker, giving a markerless deletion mutant in Pss B728a. The Suc^s pBH474 plasmid was cured from the B728a deletion mutant cells by culturing in NBY + 5% sucrose liquid medium.

Construction of complementing and overexpressing plasmids

For the complementation of B728a derivative mutants a copy of the targeted gene and the predicted promoter region was PCR amplified from Pss B728a with a BamHI and SacI restriction enzyme sited on each end of the PCR product using primers listed in S1 Table. The PCR product was digested with BamHI and SacI. Additionally, the broad-host-range promoter-probe vector, pPROBE-KT’ was digested with BamHI and SacI [39]. Digested PCR products and vector were purified using Wizard^® SV Gel and PCR Clean-Up System (Promega). Purified digested products were quantified using micro-spectrophotometry (Nano-Drop Technologies, Inc.). Ligation of the vector and insert was performed using T4 DNA ligase (New England Biolabs) and chemically transformed into E. coli One Shot^® TOP10 cells (Invitrogen) for confirmation of construct. Both constructs, pPROBE-KT’:syrF and pPROBE-KT’:syrG, were introduced into B728a derivative mutants by electroporation.

The overexpression of SyrF, SyrG, and their respective truncated proteins missing portions of the C-terminal region were cloned into the expression vector, pMEKm12 [40]. A total of two overexpression constructs was generated for each gene of interest. For syrF, the targeted gene, and 583-bp of the N-terminal sequence was PCR amplified from Pss B728a with BamHI and XbaI restriction enzyme sites flanking the PCR products. For syrG, the targeted gene, and 583-bp of the N-terminal sequence was PCR amplified from Pss B728a with BamHI and XbaI restriction enzyme sites flanking the PCR products. The PCR products and the expression vector, pMEKm12, were digested with BamHI and XbaI. Digested PCR products and vector were purified using Wizard^® SV Gel and PCR Clean-Up System (Promega). Purified digested products were quantified using micro-spectrophotometry (Nano-Drop Technologies, Inc.). Ligation of the vector and insert was performed using T4 DNA ligase (New England Biolabs) and chemically transformed into E. coli One Shot^® TOP10 cells (Invitrogen) for confirmation of construct. Overexpression constructs were introduced into Pss B728a by electroporation.

Pathogenicity assays

The ability of derivative mutants (B728aΔsalA, B728aΔsyrF, B728aΔsyrG, and B728aΔsyrFΔsyrG) to cause disease and multiply in planta was evaluated by vacuum infiltration on 2-week old Blue Lake 274 (Burpee Seeds, Warminster, PA) bean plants (Phaseolus vulgaris L.) and 4-week old Nicotiana benthamiana. The method for vacuum infiltration was described previously [32, 36]. Pss B728a was used as a positive control and Pss B728aΔgacS served as the negative control. Each strain was evaluated on at least three plants of each species, with triplicate biological replicates. The virulence of derivative mutants was evaluated by measurement of necrotic lesion surface area found on leaves 3 days post inoculation. Lesion surface areas were calculated using ImageJ software (version 1.49e; http://rsbweb.nih.gov/ij/). A total of three leaves were evaluated for each biological replicate.

To evaluate the ability of derivative mutants to replicate in planta, population analysis was performed for B728a, B728aΔsalA, B728aΔsyrF, B728aΔsyrG, B728aΔsyrFΔsyrG, and B728aΔgacS on Day 0 and Day 3 after vacuum infiltration of bean plants. From each infiltrated plant, a trifoliate leaf was detached and infiltrated tissue was removed using the bottom of a sterile 2 mL microcentrifuge tube (Bio Plas Inc., San Franscisco, CA). A total of 10 leaf discs were removed per leaf and rinsed with sterile deionized water. The leaf discs were ground using a sterile mortar and pestle with Silwet Phosphate Magnesium Buffer (SPM) [32]. Serial dilutions were prepared with SPM buffer and plated on KB agar with appropriate antibiotics followed by incubation at 26°C for 48 h. Colonies were counted and calculated as CFU per squared cm.

Syringomycin assays

The production of syringomycin by Pss B728a and derivative mutant strains were evaluated using a bioassay previously described [45] for syringomycin production on HMM agar. Bacterial strains were grown overnight in 2 ml NBY at 26°C with shaking at 180 rpm. Cells were washed and resuspended in sterile deionized water to OD₆₀₀ = 0.3 (~2 x 10⁸ CFU/ml), and 5 μl aliquots of bacterial suspension were spotted on HMM. After an incubation period of 3 days at 26°C the plates were lightly sprayed with a cell suspension of Geotrichum candidum strain F-260 using a sterile chromatography sprayer. After 24 h, quantification of syringomycin production was determined by measuring the diameter of inhibition zones and compared to the parental strain of Pss B728a. This experiment was repeated in triplicate.

RNA isolation

Pss B728a and derivative mutant strains were cultured overnight with shaking at 26°C in 5 ml of liquid NBY medium. Cells were harvested by centrifugation, washed and resuspended in sterile deionized water to a concentration of approximately 2 x 10⁸ CFU per ml. Cell suspensions (100 μl) were spread onto HMM agar and incubated at 26°C for 48 h. Total RNA was purified using an RNeasy Mini Kit along with the RNAprotect reagent following the manufacturer’s protocol (Qiagen Inc., Valencia, CA). For studies that evaluated the influence of the apoplast on gene expression, bacterial strains were introduced into bean leaves by vacuum infiltration. Approximately, 48 hours post-inoculation, a total of 40–80 leaves were harvested individually and endophytic bacteria were extracted using an acidic ethanol/phenol solution (9 mL buffer-saturated phenol (pH 6.6), 171 mL absolute ethanol, 420 mL sterile distilled water). The leaves were individualy removed from the plants and cut into squares (~3 x 3 mm²) while submerged in the ethanol/phenol, and the plant tissue and liquid were sonicated for 10 min. The solution was filtered through sterile cheese cloth and centrifuge to remove bacterial cells at 5,500 x g for 10 min. The cell and plant debris pellet was resuspended in approximately 5 mL of the supernatant and filtered through a Luer-Lock syringe packed with sterile cheesecloth and fitted with a Millipore Millex 25 mm Durapore^® PVDF 5 μm filter unit. Cells were pelleted, flash frozen and stored up to 2 weeks at -20C prior to RNA extraction. For these apoplastic conditions, two RNA samples were obtained each biological replicate, with each RNA sample including samples from 40–80 leaves. RNA samples were treated with TURBO^™ DNase (Ambion, Austin, TX) to remove residual DNA. The RNA was tested for DNA contamination using RT-PCR where RNA is used as the template with no reverse transcription reaction. The RNA quality and quantification was evaluated utilizing an Agilent 2100 Bioanlyzer (Agilent Technologies, Inc.), selecting samples with RNA Integrity Number (RIN) above 8.0 [32].

For qRT-PCR analysis, selected RNA samples were converted to cDNA by reverse transcription using Super Script Vilo^™ cDNA Synthesis kit (Invitrogen) as described by Greenwald et al. [32], and diluted to 10 ng/μl. Reverse transcription was performed with the following temperature cycle: 10 min at 25°C, 60 min at 42°C, and 5 min at 85°C.

qRT-PCR analysis

An Applied Biosystems 7500 Fast Real-Time PCR System was used in conjunction with SYBR^® Select Master Mix (Invitrogen) for qRT-PCR analysis. For each 20 μl reaction the following was used: 10 μl SYBR^® Select Master Mix, 8.20 μl nuclease-free water, 0.4 μl of both the forward and reverse primers (200 nM), and 1 μl of template cDNA (10 ng/μl). Primers used for qRT-PCR analysis as listed in S2 Table along with primers specific for recA and 16s-rRNA as internal control genes that were used to normalize gene expression [28]. For each primer pair, the linearity of detection was confirmed to have a correlation coefficient of at least 0.98 (r²>0.98) over the detection area by measuring a 5-fold dilution curve with cDNA generated from bacterial RNA. Conditions for qRT-PCR involved an incubation temperature of 95°C for two minutes, followed by 40 cycles involving 3 seconds at 95°C and 30 seconds at 60°C. A melting curve analysis was used for each qRT-PCR reaction to validate that a single primer product was amplified. qRT-PCR was performed to determine the effects of apoplastic and HMM conditions on the expression of genes that encode LuxR-like proteins in parental strain Pss B728a. The expression of genes associated with syringomycin biosynthesis, epiphytic fitness and secondary metabolism was also determined in parental strain Pss B728a compared to salA, syrF, and syrG deletion mutants in HMM media conditions.

Data was analyzed using the comparative C_t method [46], where an increase or decrease of transcript levels is determined by comparing the C_t values of the samples of interest to the C_t values of a control sample. Fold change in gene expression was calculated using the following equation: 2^-ΔΔCt = (C_{t gene-of-interest}−C_{t internal control}) Treated sample–(C_{t gene-of-interest}−C_{t internal control}) Untreated sample [46]. A 2-fold or more change in C_t for the sample of interest when compared to the control sample was considered to be significant [36]. A decrease in fold change was computed by taking the negative inverse of the fold change value [46].

Operon analysis of salA, syrF, and syrG in Pss B728a using RT-PCR

RT-PCR analysis was performed to define the operons that encompass salA, syrF, and syrG using RNA isolated from Pss B728a. Primers designed for salA, syrF, syrG, and neighboring genes (S1 Table) were used to identify if they were transcribed as monocistronic or polycistronic mRNA. Total RNA was prepared from Pss B728a by growing cells on HMM agar for 48 h at 26°C and harvesting total RNA using an RNeasy Mini Kit along with the RNAprotect reagent following the manufacturer’s protocol (Qiagen Inc., Valencia, CA). RNA samples were treated with TURBO^™ DNase (Ambion, Austin, TX) to remove residual DNA. The RNA was tested for DNA contamination using RT-PCR where RNA is used as the template with no reverse transcription reaction. The RNA quality and quantification was evaluated using an Agilent 2100 Bioanlyzer (Agilent Technologies, Inc.), and selecting samples with an RNA Integrity Number (RIN) above 8.0 [32].

Using approximately 100 ng of total RNA from Pss B728a, RT-PCR was performed using an Applied Biosystems 7500 Fast Real-Time PCR System with a OneStep RT-PCR kit (Qiagen Inc., Valencia, CA) according to the manufacturer’s instructions. Reverse transcription was performed by incubating at 50°C for 30 min. After reverse transcription, RT-PCR was carried out using the following temperature cycle: 95°C for 15 minutes, followed by 30 cycles involving 30 seconds at 94°C and 30 seconds at 55°C. After RT-PCR, amplified products were subjected to electrophoresis.

Primer extension analysis

Primer extension was performed using the Primer Extension System (Promega, Madison, WI), and a sequence marker that was created using the Sequenase Version 2.0 DNA sequencing kit following the manufacturer’s instructions (Affymetrix, Santa Clara, CA). Oligonucleotides salAPE, syrFPE, and syrGPE were radiolabeled with [γ-³²P]ATP (Perkin Elmer, Inc., Boston, MA) at the 5’ end. Primer extension was performed with 1.0 pmol of the labeled primer and 15 μg of total RNA from Pss B728a. Total RNA from Pss B728a was prepared as described previously [32]. The plasmids pLV2602, and pLV2607 were used as templates to create sequencing ladders of the upstream regions of salA, syrF, and syrG.

Computer analysis

Nucleotide sequences that were 100-bp upstream of identified transcriptional start sites were analyzed using the Softberry Bprom algorithm (http://linux1.softberry.com/berry.phtml) to identify putative σ⁷⁰-dependent promoters. These putative promoter sequences were aligned with T-Coffee [47].

Construction of GFP translational fusions and mutagenesis

To define and characterize the promoter regions of syrF and syrG, promoter fragments were PCR amplified from Pss B728a genomic DNA using primers listed in S1 Table and cloned into a gfp broad-host-range promoter-probe vector, pPROBE-KT’, resulting in translational fusions to gfp. BamHI and SacI restriction enzyme sites flanked all promoter fragments. Amplified promoter fragments were digested with BamHI and SacI along with the broad-host-range promoter-probe vector, pPROBE-KT’[39]. Digested PCR products and vector were purified using Wizard^® SV Gel and PCR Clean-Up System (Promega) and quantified utilizing micro-spectrophotometry (Nano-Drop Technologies, Inc.). Ligation of the vector and insert was performed using T4 DNA ligase (New England Biolabs) and chemically transformed into E. coli One Shot^® TOP10 cells (Invitrogen) for confirmation of constructs. Additionally, salA, syrB1, and syrP promoters were cloned in pPROBE-KT’. Cloned pPROBE-KT’ constructs were introduced into B728a derivative mutants by electroporation.

GFP assays

Quantitative GFP assays were performed as described by Miller et al. [39]. E. coli cells were cultured overnight in LB with the appropriate antibiotics at 37°C with shaking. For Pss B728a, cells were cultured overnight in NBY with the appropriate antibiotics at 26°C with shaking. Cells were harvested, washed, and resuspended in 10 mM phosphate buffer to a concentration of 2 x 10⁹ cells per mL. GFP fluorescence was measured on Tecan SpectraFluor (Tecan) at an excitation wavelength of 485 nm, and an emission wavelength of 525 nm. Intensity readings were represented by arbitrary units and normalized to a cell density of 10⁹ cells per mL.

SalA, SyrF, and SyrG are novel LuxR transcriptional regulators with homologs found exclusively in Pseudomonas syringae genomospecies 2

An investigation of the Pss B728a genome identified 24 genes encoding LuxR-like regulatory proteins, which were also found in the genome of Pss B301D, displaying a high degree of nucleotide conservation between these two bacterial strains. Three genes of interest are salA, syrG, and syrF that are located adjacent to the syringomycin gene cluster. A conserved domain search of salA, syrG, and syrF using NCBI Conserved Domains database (http://www.ncbi.nlm.nih.gov/Structure/cdd/) confirmed the presence of a conserved helix-turn-helix DNA binding motif. The HTH DNA binding motif on the C-terminal region of the proteins is typical of LuxR regulatory proteins, but they lacked an N-terminal autoinducer-binding domain and receiver domain. BLAST analysis was also performed on SalA, SyrF, and SyrG protein sequences to determine the degree of conservation in pseudomonad strains. It was observed that all three regulatory proteins are exclusively found in P. syringae genomospecies 2, with the C-terminal region being highly conserved with regulatory protein SyrG (S1 Fig).

Expression of the syrG and syrF genes in the apoplast of bean relative to HMM medium

Analysis of the 6.09 Mb Pss B728a genome identified 24 genes dispersed throughout the genome ecoding LuxR-like proteins. Some of these proteins have been implicated in virulence and secondary metabolism in Pss B728a. To identify the LuxR-like proteins that are important to the plant-pathogen interaction, qRT-PCR analysis was used to determine transcript abundance of the genes identified as LuxR-like proteins in the apoplast of bean when compared to parental strain B728a in conditions conducive for hrp gene expression. The results indicated that in Pss B728a, change in gene expression between these conditions was the greatest for both syrG and syrF genes in the apoplast of bean relative to HMM liquid medium (Fig 2). This result is consistent with earlier reports of syrG gene expression levels in the apoplast [5, 10]. Higher expression levels of these genes in the apoplast is attributed to the presence of plant signal molecules known to activate expression of syr genes [6]. The relative expression of syrG and syrF when compared to other luxR genes indicate that the SyrG and SyrF proteins are involved in the transcriptional regulation of genes that potentially are critical to plant pathogenesis.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 2. Expression analysis in the apoplast of bean of genes encoding LuxR-like proteins in Pss B728a.

The genes that encode proteins that are classified in the first subfamily of LuxR (Class I) are shown in the green and are typically associated with two component signal transduction systems. Shown in blue are genes that encode LuxR-like proteins implicated in quorum sensing based on domain architecture. Psyr_0993, which is shown as gray, is the only gene that encodes a protein characterized as a LAL or LuxR-like proteins that require ATP for activation. The final subfamily of LuxR-like proteins are encoded on genes shown in orange bars, which lack an N-terminal regulatory domain and are associated with secondary metabolism. Out of all 24 LuxR-like proteins found in the genome of Pss B728a, the genes encoding SyrG and SyrF are the most highly expressed in the apoplast when compared to HMM liquid medium. The values are represented as the average fold change of three technical replicates of three biological samples. Gene expression was normalized to the 16s-rRNA and recA internal control genes. Vertical bars indicate standard errors of the average values over triplicate runs.

https://doi.org/10.1371/journal.pone.0150234.g002

SalA, SyrF, and SyrG influence virulence of Pss B728a on bean plants

The generation of a clean deletion mutant of syrG and syrF in Pss B728a was achieved by using the mutation strategy described in the Materials and Methods. The bacterial strains B728aΔsyrG, B728aΔsyrF, and B728aΔsyrGΔsyrF displayed colony morphologies and growth curve patterns similar to parental strain B728a (data not shown). The derivative mutants B728aΔsyrG, B728aΔsyrF, and B728aΔsyrGΔsyrF were significantly reduced in virulence relative to the parental strain Pss B728a shown in Fig 3. The salA mutant failed to produce watersoaked necrotic lesions typical of Pss B728a. The salA mutant was comparable to B728aΔgacS in regards to virulence by lacking the ability to produce necrotic lesions and cause disease on bean. B728aΔsyrG was reduced in virulence by approximately 95% when compared to the parental strain Pss B728a. Mutants of syrG displayed small, non-spreading lesions with the average surface area of 45.1 mm² on bean leaves. In contrast, the syrF mutant was able to produce large necrotic lesions, but displayed approximately 61% reduction in virulence on bean with the average necrotic lesion surface area of 284.5 mm². The double deletion mutant of syrG and syrF exhibited disease symptoms comparable to the syrG mutant. Virulence of syrG and syrF derivative mutants was partially restored in trans by complementation of syrG and syrF (Fig 3).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 3. Pathogenicity assays to evaluate the contribution of syrG and syrF to virulence on bean.

Bean leaves were inoculated by vacuum infiltration with bacterial suspensions containing 10⁷ CFU/cm² of either B728a, B728aΔsyrG, B728aΔsyrF, B728aΔsyrGΔsyrF, B728aΔgacS, B728aΔsyrG (pKT::syrG), B728aΔsyrF (pKT::syrF), or B728aΔsyrGΔsyrF (pKT::syrG). Plants were maintained at room temperature in a growth chamber for 72 h. Necrotic lesion surface areas were calculated using ImageJ software. This experiment was performed in triplicate, and representative results are shown.

https://doi.org/10.1371/journal.pone.0150234.g003

Bacterial populations of infected bean plants were monitored over a 3-day period. At 3 days post-inoculation, bacterial titers for parental strain Pss B728a was 6.5 x 10⁷ CFU/cm², while B728aΔsalA, B728aΔsyrG and B728aΔsyrF were 2.0 x 10⁶ CFU/cm², 3.2 x 10⁷ CFU/cm² and 1.5 x 10⁷ CFU/cm², respectively. B728aΔgacS, which fails to produce disease on bean, maintained a population equivalent to day 0 of 2.2 x 10⁴ CFU/cm² 3 days post-inoculation. The bacterial population of B728aΔsyrG and B728aΔsyrF were not significantly different from the parental strain Pss B728a, indicating that the syrG and syrF genes are not required for multiplication in planta. In contrast, B728aΔsalA displayed a 10-fold reduction in bacterial titers compared to parental strain B728a. B728aΔsalA is able to replicate in planta, but at a reduce rate when compared to B728a. In the case of B728aΔgacS, the bacterium remains viable but is limited on its ability to multiply in planta.

Deletion mutants of the salA, syrF, and syrG genes in Pss B728a affect syringomycin production

The bioassay used to evaluate the influence deletion of salA, syrG, and syrF has on syringomycin production was determined by measuring zones of antifungal activity to G. candidum as compared to parental strain Pss B728a grown on HMM agar (Fig 4). All the deletion mutants, including the double mutant Pss B728aΔsyrFΔsyrG, displayed no measurable antifungal activity toward G. candidum. Antifungal activity toward G. candidum was partially restored when B728a mutant derivatives were complemented in trans with the vector pPROBE-KT’ carrying an intact copy of the syrF or syrG gene. These results differed from a previous study by Lu et al. [10] with site directed insertional mutants of salA, syrF, and syrG in Pss B301D. It is hypothesized that the insertional mutants of syrF and syrG produce truncated proteins with reduced functional activity displaying low levels of toxin production. In contrast, clean deletion mutants of syrF and syrG were generated in Pss B301D. These derivative mutants displayed a loss of syringomycin production similar to the derivative mutants generated in Pss B728a (data not shown).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 4. Bioassy to evaluate syringomycin production in parental strain B728a and derivative mutants.

Bacterial strains were grown on HMM for 4 days. Plates were oversprayed with Geotrichum candidum and incubated 24 h at 26°C to observe zones of inhibition indicative of syringomycin production. The experiment was repeated in triplicate.

https://doi.org/10.1371/journal.pone.0150234.g004

Overexpression of N-terminal truncated proteins of SyrG and SyrF has an effect on syringomycin production

To test the hypothesis that insertional mutants of syrF and syrG are leaky mutations and to demonstrate the HTH DNA binding domain of SyrG and SyrF are essential for binding to syr-syp promoters, the SyrG and SyrF proteins lacking the C-terminal HTH DNA-binding domain were overexpressed in Pss B728a. The overexpression of SyrG and SyrF had no effect on syringomycin production. However, overexpression of the N-terminal regions lacking the C-terminal HTH DNA-binding domains of SyrG and SyrF in Pss B728a resulted in a marked reduction of syringomycin zones of inhibition to G. candidum from 16 mm to 0.5 mm and 3 mm, respectively (Fig 5). The overexpression of the N-terminal regions of SyrG and and SyrF resulted in a 97% and 81% reduction in syringomycin production, which can be attributed to nonfunctional heterodimers formed between wild-type proteins and the truncated proteins. These nonfunctional heterodimers lack the ability to properly bind to syr-syp promoters, which is essential for the transcriptional activation of genes required for phytotoxin production. The truncation of SyrG displayed the greatest reduction in syringomycin production, which is comparable to the overexpression of an N-terminal truncated SalA in Pss B301D [12].

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 5. Effect of overexpression of N-terminal region (NTR) of SyrG and SyrF on syringomycin production in Pss B728a.

Bacterial strains were grown on HMM for 4 days. Plates were oversprayed with Geotrichum candidum and incubated 24 h at 26°C to observe zones of inhibition indicative of syringomycin production. The experiment was repeated in triplicate.

https://doi.org/10.1371/journal.pone.0150234.g005

The effect of syrG and syrF deletion mutants on genes associated with virulence

Previous experiments have established both syrG and syrF have an influence on syringomycin production and virulence in Pss B728a. Quantitative real-time PCR [48] was used to identify the effect syrG and syrF deletion mutants have on genes associated with syringomycin production and virulence. A total of 23 genes were evaluated using qRT-PCR that included genes involved in the biosynthesis of syringomycin, syringopeptin, achromobactin, alginate, levansucrase, syringolin, syringafactin, and pyoverdine. Both syrG and syrF had an effect on the expression of syringomycin biosynthesis genes (Fig 6). A deletion mutant of syrG resulted in a 4- to 11-fold decrease in transcript abundance of syrB1, syrB2, syrC, syrD, syrE and syrP. In regards to syrF, the deletion mutant resulted in a 2- to 12-fold decrease of transcript abundance of syringomycin biosynthesis genes. The results also indicated that both SyrG and SyrF are involved in the transcriptional regulation of genes associated with syringomycin production. Mutants of syrG and syrF did not appear to have an effect on the expression of genes associated with achromobactin, alginate, levansucrase, syringolin, syringofactin, or pyoverdine biosynthesis (data not shown). Both syrG and syrF mutants failed to have an effect on known virulence genes outside of the syr-syp gene cluster.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 6. Quantitative real-time PCR analysis of syringomycin biosynthesis genes in ΔsalA, ΔsyrG, and ΔsyrF mutants of Pss B728a.

The values represent the average fold change in gene expression from parental strain Pss B728a; the results are the averages of three technical replicates from three biological samples grown in HMM liquid medium. Gene expression levels were normalized to 16s-rRNA and recA internal control genes, and vertical bars indicate standard errors of the average values over triplicate runs. Negative values indicate a decrease in transcript abundance by taking the negative inverse of a fold change value less than 1.

https://doi.org/10.1371/journal.pone.0150234.g006

The effect of syrF and syrG deletion mutants on LuxR-like homologs in Pss B728a

To determine the effect syrF and syrG mutants have on LuxR-like homologs in Pss B728a, qRT-PCR analysis was performed using primers specific for salA, syrG, syrF, sylA, syrR, and Psyr_2578. SylA and SyrR are LuxR-like proteins that have been implicated in the regulation of syringolin and syringofactin [5, 35, 49]. Results from qRT-PCR analysis show that syrG and syrF require a functional salA gene for activation (Fig 7). Mutants of salA displayed a 3.6-, and 4.5-fold decrease in transcript abundance of syrG and syrF, respectively. Mutants of syrG displayed a 27-fold increase in transcript abundance of syrF, and mutants of syrF displayed a 20-fold increase in transcript abundance of syrG. Data obtained from qRT-PCR analysis indicated that both syrG and syrF negatively regulate expression of each other’s gene. SyrG and SyrF did not have an effect on the expression of the LuxR-like genes sylA, syrR, and Psyr_2578 indicating they are not part of the SyrG or SyrF regulatory networks.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 7. Quantitative real-time PCR analysis of LuxR-like genes in ΔsalA, ΔsyrG, and ΔsyrF mutants of Pss B728a.

The values represent the average fold change in gene expression from parental strain Pss B728a; the results are the averages of three technical replicates from three biological samples grown in HMM liquid medium. Gene expression levels were normalized to 16s-rRNA and recA internal control genes, and vertical bars indicate standard errors of the average values over triplicate runs. Negative values indicate a decrease in transcript abundance by taking the negative inverse of a fold change value less than 1.

https://doi.org/10.1371/journal.pone.0150234.g007

The syrF gene is in an operon with oprM, whereas salA and syrG are monocistronic mRNA transcripts

The syr-syp gene cluster consists of genes involved in the biosynthesis, regulation and secretion of syringomycin and syrinopeptin. These genes are organized into two operons that were defined by Wang et al. [50] in Pss B301D. Located adjacent to the syr-syp gene cluster are the LuxR-like regulatory genes salA, syrF, and syrG. RT-PCR analysis (data not shown) revealed that both salA and syrG are transcribed as moncistronic mRNA with their own native promoter regions. There were no salA-syrG products observed with RT-PCR using primers specific to the 3’ and 5’ sequences of these genes, respectively. The genes that are organized in a polycistronic operon are syrF and oprM. A syrF-oprM product was obtained with RT-PCR using primers oprMF and syrFR that were specific for the 3’ and 5’ sequences of these genes, respectively.

Characterization of transcriptional start sites of the syrG and syrF genes

After identifying the monocistronic and polycistronic transcripts of salA, syrF, and syrG, the transcriptional start sites of the respective genes were defined using primer extension analysis. For salA, the salAPE primer was radiolabeled and used to identify the transcriptional start site. Primer extension analysis revealed the transcription start site was at the thymine residue, which was 63 bp upstream to the translational start codon of salA. This result is similar to the transcriptional start site identified for salA in Pss B301D by Wang et al. [50]. However, the prior study in Pss B301D did not identify the transcriptional start sites for syrF and syrG. For syrF, the syrFPE primer was used to identify the transcriptional start site at the cytosine residue, which was revealed to be 498 bp upstream of the translation start codon. The syrGPE primer was used to identify the transcriptional start site at the thymine residue, 235 bp upstream of the translation start codon. The transcriptional start site of syrG suggests a putative promoter region sequence, CTGAGAN₁₇TCTTAT (Fig 8). Similarly, the transcriptional start site of syrF suggests a putative promoter region sequence, TTGTTAN₂₃TGCAAC. In addition, computer analysis of promoter sequences identified conserved sequences observed around the -35 promoter regions of syrG and syrF, illustrated by the nucleotide sequence alignment of the predicted promoter regions (Fig 9). These putative promoter regions were predicted using defined transcriptional start sites and BPROM promoter prediction software [51]. Both promoters share high similarity to the consensus promoter sequence of σ⁷⁰ found in Gram-negative bacteria [52].

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 8. Comparison of putative promoter sequences of salA, syrG and syrF.

Predicted promoter sequences are based on defined transcriptional start sites and using BPROM promoter prediction software. Conserved sequence motifs corresponding to -35 and -10 promoter regions are underlined and compared to a σ⁷⁰ dependent promoter sequence. The defined promoter region of salA is distinctly different from the predicted promoter regions of syrG and syrF.

https://doi.org/10.1371/journal.pone.0150234.g008

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 9. Alignment of syrG and syrF promoter sequences in Pss B728a.

The predicted promoter sequences of syrG and syrF were aligned using T-COFFEE and conserved sites are shown as asterisks. The color code is based on CORE index, using consistency among pairwise alignments for estimating reliability. Sequences shown in red indicate high reliability, where green is indicative of low reliability.

https://doi.org/10.1371/journal.pone.0150234.g009

Identification of essential syrG and syrF promoter regions

The effects of promoter deletion mutants on the expression of syrG::gfp and syrF::gfp fusions are shown in Fig 10. For syrG::gfp fusions, deletion constructs were generated from 752-bp, 552-bp, 452-bp, 352-bp, 252-bp, 202-bp, 152-bp, 102-bp, and 52-bp upstream of the translational start site of syrG. For syrF::gfp fusions, deletion constructs were generated from 1000-bp, 800-bp, 600-bp, 500-bp, 252-bp, 202-bp, 152-bp, 102-bp, and 52-bp upstream of the translational start site of syrF. Results indicated that the 100-bp region of 252- to 352-bp upstream of the start codon of syrG is critical for the expression of the syrG::gfp fusion. In addition, the 100-bp region of 500- to 600-bp upstream of the start codon of syrF is critical for the expression of the syrF::gfp fusion. When the predicted -10 promoter region TCTTAT of syrG and TGCAAC of syrF was substituted with CTGCAG, expression of the syrG::gfp and syrF::gfp reporters in Pss B728a decreased by 71% and 70%, respectively. When the predicted -35 promoter region CTGAGA of syrG and TTGTTA of syrF was substituted with CTGCAG, expression of the syrG::gfp and syrF::gfp reporters in Pss B728a decreased by 80% and 68%, respectively.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 10. Analysis of the promoter regions of syrG (A) and syrF (B) by testing the effect deletion mutants have on the expression of syrG::gfp and syrF::gfp transcriptional fusions.

All gfp constructs were electroporated into cells of Pss B728a to measure GFP fluorescence (AU). All measurements were averages over three technical replicates of three biological samples. Horizontal bars represent the standard error of the average values.

https://doi.org/10.1371/journal.pone.0150234.g010

LuxR-type transcription regulator SalA is a positive regulator of syrF and syrG promoters

The effects of deletion mutants of salA, syrG, and syrF have on salA::gfp, syrG::gfp, and syrF::gfp reporter gene activities are shown in Table 2. In the parental strain Pss B728a, the salA::gfp construct (pPKT::salA) displayed the relative GFP fluorescence of 43044 AU/10⁹ CFU. Fluorescence of a salA::gfp fusion was reduced by 76% in a salA derivative mutant of Pss B728a. Deletion mutants of syrG and syrF did not significantly reduce the GFP fluorescence of a salA::gfp transcriptional fusion. Reporter gene activities of syrG::gfp and syrF::gfp were significantly reduced by 91% and 95% in B728aΔsalA as compared to activities in Pss B728a. This indicates that SalA is a transcriptional activator that functions upstream of syrG and syrF.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Effect of salA, syrG, and syrF mutants on gfp reporter gene activity for salA, syrG, syrF, and syrB1.^a

https://doi.org/10.1371/journal.pone.0150234.t002

SyrG is a positive regulator of syrF and both syrG and syrF are involved in the expression of syrB1

The effect deletion mutants of syrG and syrF have on the syrG::gfp and syrF::gfp reporter gene activities is shown in Table 2. The parental strain Pss B728a harboring the syrG::gfp transcriptional fusion displays a relative GFP fluorescence of 45776 AU/10⁹ CFU. GFP fluorescence of syrG::gfp decreased to 6121 AU/10⁹ CFU in syrG derivative mutants of Pss B728a, where the GFP fluorescence syrG::gfp was not significantly reduced in syrF deletion mutants. These results indicated that SyrG is required to activate its own gene expression, but SyrF does not have an affect on the promoter activities of syrG. The parental strain Pss B728a harboring the syrF::gfp transcriptional fusion displayed a relative GFP fluorescence of 44857 AU/10⁹ CFU. The relative GFP fluorescence of syrF::gfp decreased to 6163 and 7713 AU/10⁹ CFU in deletion mutants of syrG and syrF, respectively. Both SyrG and SyrF are required for activation of the syrF gene.

The parental strain Pss B728a harboring the syrB1::gfp transcriptional fusion displays a relative GFP fluorescence of 47266 AU/10⁹ CFU (Table 2). GFP fluorescence of syrB1::gfp decreased to 9653, 24347, 29458 AU/10⁹ CFU in deletion mutants of salA, syrG, and syrF, respectively. Both SyrG and SyrF are involved in the expression of syrB1.

Overexpression of SyrF restores syringomycin production in syrG deletion mutants of Pss B728a

The derivative mutants of syrG and syrF in Pss B728a displayed a significant loss of syringomycin production when compared to the parental strain Pss B728a. The overexpression of SyrF had the ability to partially restore syringomycin production in syrG derivative mutants. Syringomycin inhibition zones increase from 0 mm to 25 mm, resulting in a 25% increase in syringomycin production when compared to a syrG derivative mutant. However, the overexpression of SyrG failed to restore syringomycin production in syrF derivative mutants (data not shown). These results indicated that SyrG is an upstream activator of syrF.