Bacterial strains and culture conditions
The experimental strains, plasmids, and PCR primers used for this study are listed in Additional file 1, Additional file 2, and Additional file 3, respectively. R. capsulatus was grown at 35°C under anaerobic photoheterotrophic conditions with YPS medium  or aerobically with RCV medium  supplemented with appropriate antibiotics when necessary: kanamycin (10 μg ml-1), spectinomycin (20 μg ml-1), and tetracycline (0.5 μg ml-1). Escherichia coli was grown using LB medium at 37°C and supplemented with the appropriate antibiotics when necessary: ampicillin (100 μg ml-1), kanamycin (25 μg ml-1), spectinomycin (50 μg ml-1), and tetracycline (10 μg ml-1).
Open reading frames (ORFs) of the Rba proteins and σ factors were amplified by PCR from the genome of R. capsulatus strain SB1003 and cloned into pGEM-T-Easy (Promega, Madison, USA) according to the manufacturer’s guidelines. The genes were disrupted by insertion of a ~1.4-kb SmaI fragment of the KIXX cartridge , which confers resistance to kanamycin and which has been found to rarely create polar mutations in R. capsulatus. The rbaV (rcc03323) and rbaW (rcc03324) ORFs were amplified using the primers VW-F and VW-R. The rbaV gene was disrupted by insertion at an NruI site located 76 bp into the 348-bp ORF. The rbaW gene was disrupted by insertion at a BlpI site blunted with T4 polymerase, located 274 bp into the 492-bp ORF. A disruption of both genes was created by replacing a 535-bp NruI/BlpI segment with the KIXX fragment. The ORF predicted to encode the rsbY homologue (rcc00181) was amplified using the primers Y-F and Y-R. The 1230-bp rbaY ORF was disrupted at an MscI site located 307 bp into the gene.
Amplicons of the R. capsulatus rpoHI (rcc02811) and rpoHII (rcc00458) genes were amplified using primers rpoHI-F and rpoHI-R, and rpoHII-F and rpoHII-R, respectively. The 900-bp rpoHI ORF was disrupted at a BamHI site located 323 bp from the start of the gene. A 507-bp StuI fragment of the 833-bp rpoHII ORF was replaced by the KIXX cartridge. The ORF encoding the putative ECF σ factor-encoding rcc02291 (570 bp) was amplified using primers 2291-F and 2291-R and disrupted by insertion at a StuI site located 133 bp into the gene. Also, the putative phyR orthologue (rcc02289) and potential anti-σ factor to the protein encoded by rcc02291, was amplified using primers phyR-F and phyR-R and subsequently disrupted by a KIXX cartridge insertion at a SmaI site located 150 bp into the 810 bp ORF. The 594-bp ORF rcc02724 encoding another putative ECF σ factor was amplified using primers 2724-F and 2724-R and disrupted by inserting KIXX into a BsaBI site located 221 bp from the start of the gene. The ORFs rcc00699 (545 bp) and rcc02637 (585 bp) encoding putative σ24 ECF sigma factors were amplified using primers 699-F and 699-R, and 2637-F and 2637-R, respectively. The KIXX cartridge was inserted into a StuI site 376 bp into rcc00699 and an AfeI site located 176 bp from the start of rcc02637. Disruptions were not attempted for the major vegetative σ factor, rpoD (rcc03054), or the nitrogen fixation σ factor, rpoN (rcc00568), genes. A separate rpoHI disruption using a 2-kb spectinomycin resistance-encoding omega cassette  was constructed to allow creation of an rpoHI-rpoHII double mutant strain.
All gene disruption constructs were sequenced and RcGTA-mediated transfer of disrupted versions of genes into R. capsulatus SB1003 were carried out as previously described . The resulting kanamycin and kanamycin/spectinomycin resistant strains (Additional file 1) were confirmed to contain the gene disruptions by PCR using the original amplification primers (Additional file 3) whereby replacement of the wild type gene by the disrupted version was indicated by amplification of a single product of the expected size.
In trans complementation was performed using wild type genes with their native upstream sequences placed on the low copy, broad host range plasmid, pRK767 . A wild type fragment of rbaV and rbaW was amplified using primers VcF and VW-R. Primers VcF and Anti-anti-R were used to amplify the wild type rbaV fragment. The rbaW complement sequence contained an in-frame deletion of the majority of rbaV, replacing bp 24 to bp 272 with a KpnI site. This was created by joining 2 fragments, amplified with VcF and VdR, and VdF and VW-R, via a primer-embedded KpnI site. The complementation vectors (Additional file 2) were conjugated into R. capsulatus using E. coli S17-1 .
Gene transfer bioassays
Gene transfer bioassays were used as previously described  to measure production and release of RcGTA particles. Stationary phase cultures were filtered using 0.45-μm PVDF syringe filters and filtrates assayed for RcGTA activity using the R. capsulatus puhA strain, DW5 , as the recipient cells. The samples were plated on YPS agar and incubated in anaerobic phototrophic conditions and colony numbers were counted after 48 hours. RcGTA activities in mutant strains were determined as ratios relative to SB1003 in 3 replicate experiments. Statistically significant differences in RcGTA activities were identified by one-way analysis of variance (ANOVA) in R .
Western blots targeting the ~32 kDa RcGTA major capsid protein were performed on the same cultures used for RcGTA activity assays as described previously . Samples contained 5 μl of cells pelleted from cultures and re-suspended in an equal volume of TE buffer or 10 μl of the culture supernatants mixed with 3× SDS-PAGE sample buffer and heated for 5 minutes at 98°C. The proteins were separated on a 10% SDS-PAGE gel and transferred to a nitrocellulose membrane by electro-blotting in transfer buffer [48 mM Tris Base, 39 mM glycine, 20% methanol (v/v)]. Total protein levels within supernatant and cell sample groups were verified to be approximately equivalent by staining the membranes with Ponceau-S. The membranes were rinsed and blocked with a 5% (w/v) skim milk solution in TBST [20 mM Tris, 137 mM NaCl, 0.1% Tween-20 (v/v); pH 7.5] and incubated overnight at 4°C with an anti-Rhodobacterales GTA major capsid protein primary antibody (Agrisera, Vännäs, Sweden)  as a 1:1000 dilution in TBST. The membranes were then washed with TBST and incubated with peroxidase-conjugated anti-rabbit IgG (Santa Cruz Biotechnology, Dallas, USA) as a 1:5000 dilution in TBST for 1 hour at room temperature. The membranes were washed again in TBST and the bands were detected by chemiluminescence using the SuperSignal West Femto Reagent Kit (Thermo Fisher Scientific, Ottawa, Canada). Images were captured on an Alpha Innotech U400 camera, and then inverted and adjusted for brightness and contrast with image processing software.
Viable cell counts
Each culture used for gene transfer assays and western blotting was also assayed for viable cells as previously described . Serial dilutions were plated and colony-forming units (cfu) were calculated for the 3 biological replicates to determine the number of viable cells. The data were converted to a ratio relative to the parental strain. Statistically significant differences in viable cell numbers were identified by one-way ANOVA in R .
β-galactosidase reporter fusions
In-frame fusions of RcGTA orfg2 to the E. coli lacZ gene were constructed using PstI/BamHI fragments cloned into the promoter probe vector pXCA601 vector . Fragments 2 (pX2) and 2NP (pX2NP) were amplified by PCR using primers GTA-F1 and GTA-R1, and GTA-F2 and GTA-R1, respectively. Fragments 2.1 and 2.2 were amplified using primers GTA-F1 and GTA-DP-R, and GTA-DP-F and GTA-R1, respectively. Fragment g2Δp (pX2Δp) was created by ligating 2.1 and 2.2 via a primer-embedded KpnI restriction site, resulting in a deletion of the sequence from -129 to -100 5’ of RcGTA orfg1 (Additional file 2). Fragments 2.3 and 2.4 were amplified using GTA-F1 and GTA-DS-R, and GTA-DS-F and GTA-R1, respectively. The fragment g2Δs was made by combining 2.3 and 2.4 via a primer-embedded KpnI restriction site, resulting in a deletion of the sequence from -73 to -46 5’ of orfg1 (Additional file 2). All fusions were confirmed to be in-frame by sequencing, and the plasmids were transferred into R. capsulatus strains by conjugation using E. coli S17-1 .
Strains of R. capsulatus containing the fusion constructs listed in Additional file 2 were grown in conditions identical to those for RcGTA activity assays. Cells were permeabilized for 15 minutes using 15% (v/v) isopropyl alcohol and washed using Z buffer (60 mM Na2HPO4, 40 mM NaH2PO4, 1 mM MgSO4, 10 mM KCl, 50 mM β-mercaptoethanol; pH 7) . The cells were resuspended in Z buffer and substrate, fluorescein di-β-D-galactopyranoside (FDG) (Sigma-Aldrich) dissolved in H2O:DMSO:ethanol (8:1:1), was added at a final concentration of 0.1 mg ml-1. The cells were then incubated for 1 hour at room temperature and diluted 1:200 in Z buffer before analysis by flow cytometry with recording of 105 events. The mean sample fluorescence was obtained from gated cells from two biological replicates.
Expression and purification of recombinant proteins from E. coli
For expression of recombinant N-terminal 6×-histidine tagged proteins, rbaW and rbaV were independently cloned as NdeI/BamHI fragments into the pET15b vector (Novagen, Darmstadt, Germany), using primers Anti-S-F and Anti-S-R, and Anti-anti-F and Anti-anti-R, respectively. This resulted in the rbaW and rbaV sequences in-frame with an N-terminal 6x-histidine tag. A C-terminal 6×-histidine tagged sequence of RbaW was also created using the primers Anti-SC-F and Anti-SC-R, with the product cloned as an NcoI/XhoI fragment into the pET26b vector (Novagen). The plasmids, pET15W, pET15V and pET26W (Additional file 2), were sequenced to confirm the R. capsulatus sequences were in-frame with the histidine tags and then transformed into E. coli BL21(DE3) (New England Biolabs, Whitby, Canada).
Overnight starter cultures were used to inoculate 200 ml of LB broth containing either ampicillin (pET15b derivatives) or kanamycin (pET26b derivative), followed by incubation for 1 hour at 37°C with shaking at 250 rpm. Expression of the recombinant proteins was induced by addition of isopropyl-β-D-thiogalactopyranoside (IPTG) to a final concentration of 1 mM followed by growth at 37°C for 4 hours with shaking at 250 rpm. Cell pellets of these induced cultures were resuspended in lysis buffer [50 mM NaH2PO4, 300 mM NaCl, 10 mM imidazole, 0.1% (v/v), Benzonase® nuclease (Qiagen, Toronto, Canada), 1 mg ml-1 lysozyme (w/v); pH 8] and incubated on ice for 30 minutes. The lysates were centrifuged at 14000 × g for 30 minutes at 4°C and supernatants were mixed 4:1 (v:v) with Ni-NTA agarose (Qiagen) and incubated at 4°C with shaking at 200 rpm for 1 hour. The samples were loaded into polypropylene columns, washed twice with wash buffer (50 mM NaH2PO4, 300 mM NaCl, 20 mM imidazole; pH 8) and the fusion proteins eluted in 1 ml aliquots of elution buffer (50 mM NaH2PO4, 300 mM NaCl, 250 mM imidazole; pH 8). The purified proteins were dialyzed into a coupling buffer (20 mM sodium phosphate buffer, 500 mM NaCl; pH 7.5) and quantified using a ND-1000 Nanodrop spectrophotometer.
In-gel digestion and peptide extraction for LC-MS/MS sequencing
Purified recombinant protein samples were mixed with 3× SDS-PAGE sample buffer, heated for 5 minutes at 98°C, and run on a 10% SDS-PAGE gel. The gels were stained with Coomassie Blue [0.25% (w/v) Coomassie Brilliant Blue R-250 in methanol:H2O:acetic acid (5:4:1)] for 30 minutes, destained in methanol:H2O:acetic acid (5:4:1), and recombinant protein bands of predicted sizes were cut out using a clean scalpel. The gel slices were washed first with water, followed by 100 mM NH4HCO3, and finally acetonitrile, with samples being vortexed for 10 minutes, centrifuged at 3000 × g and supernatants decanted after each wash step. The samples were dried in a vacuum centrifuge for 5 minutes before adding a sufficient amount of 10 mM dithiothreitol (DTT) in 100 mM NH4HCO3 to cover the gel slices. After incubation for 45 minutes at 56°C, the samples were centrifuged at 3000 × g and the supernatant decanted. The solution was replaced by 55 mM iodoacetamide in 100 mM NH4HCO3 and the samples incubated in the dark at room temperature for 30 minutes with occasional vortexing. The samples were centrifuged at 3000 × g and all liquid was removed by aspiration. The samples were washed in 100 mM NH4HCO3 with vortexing for 10 minutes followed by centrifugation at 3000 × g and removal of the supernatant. This wash procedure was repeated once with acetonitrile and twice with 50% (v/v) acetonitrile. The samples were vacuum-centrifuged for 15 minutes before the addition of sequencing grade trypsin (12 ng μl-1) in trypsin digestion buffer (Promega). The tubes were sealed and incubated overnight at 37°C. After addition of formic acid (to 5% v/v) and vortexing, the samples were centrifuged at 3000 × g and supernatants collected in a separate tube. This extraction process was repeated sequentially with 1% formic acid-5% acetonitrile (v/v), 1% formic acid-60% acetonitrile (v/v), and 1% formic acid-99% acetonitrile (v/v). The supernatants from each of these extractions were collected together in one tube and vacuum centrifuged. The dried extracts were sequenced by LC-MS/MS at the Genomic and Proteomic (GaP) facility at Memorial University.
In vitro protein interaction assays
In vitro interaction assays were carried out by separately conjugating 50 μg of recombinant RbaW protein, carrying a 6x-histidine tag on either the N- or C-terminus, to NHS-activated beads (GE Healthcare Life Sciences, Baie d’Urfe, Canada) according to the manufacturer’s guidelines. The conjugated beads were washed several times with 100 mM Tris-HCl (pH 8.0) then resuspended as a 50% (v/v) slurry in the same solution. A sub-sample of conjugated bead slurry was resuspended in a binding buffer [10 mM Tris-HCl (pH 8.0), 200 mM NaCl, 5% (v/v) glycerol, 0.5 mM DTT, and 0.5% (v/v) triton X-100] and either 6x-His-RbaV or chicken egg white lysozyme control protein (Sigma-Aldrich, Oakville, Canada) was added to a final concentration of ~1 μM. The mixture was incubated on ice for 30 minutes with occasional agitation before adding 0.5 ml of binding buffer. The beads were allowed to sediment by gravity and the supernatant was removed. Washing with 0.5 ml of binding buffer was repeated 3 times to remove all non-bound protein. The beads were resuspended in 30 μl of 3× SDS-PAGE buffer, heated for 5 minutes at 98°C, and 20 μl of the sample run on a 10% SDS-PAGE gel. To confirm specific interaction between recombinant fusion proteins, additional control reactions were performed. First, non-conjugated beads were treated with 100 mM Tris-HCl (pH 8.0) and then incubated with test proteins to ensure adequate blocking of bead active sites. Second, conjugated 6x-His-RbaW and RbaW-6x-His were independently incubated with chicken egg white lysozyme to ensure specific interactions between the experimental test proteins.
Bacterial two-hybrid assays
Bacterial two-hybrid analyses for determining protein interactions were carried out as described  using the bacterial adenylate cyclase-based two-hybrid, or BACTH, system (EUROMEDEX, Souffelweyersheim, France). Primers AS-AF and AS-AR, and AAS-AF and AAS-AR, were used to amplify rbaW and rbaV by PCR, respectively. The rpoD and rpoHI σ factor-encoding genes were amplified using rpoD-F/rpoD-R and rpoH-AF/rpoH-AR, respectively. Putative ECF σ factor-encoding genes rcc02637 and rcc00699 were amplified using 2637-AF and 2637-AR, and 699-AF and 699-AR, respectively. All amplicons were cloned as KpnI fragments into all 4 BACTH vectors: pKNT25, pKT25, pUT18 and pUT18c (Additional file 2). All pair-wise combinations of bait (rbaW) and prey (rbaV, rpoD, rpoHI, rcc02637 and rcc00699) recombinant vectors were co-transformed into cya
E. coli BTH101 and plated on LB agar supplemented with ampicillin, kanamycin, 40 μg ml-1 5-bromo-4-chloro-3-indolyl-β-D-galactopyranoside (X-Gal) and 0.5 mM IPTG. Positive control plasmids encoding interacting fragments of a leucine zipper protein, pKT25-zip and pUT18C-zip (Additional file 2), were also co-transformed. Plates were incubated for 48 hours at 30°C.
For quantitative determination of β-galactosidase activity, 3 replicate co-transformants were picked for each interaction to inoculate fresh LB broth containing antibiotics and 0.5 mM IPTG. Cultures were grown overnight at 37°C and then diluted 1:5 in LB broth and the OD600 was determined. The cells were permeabilized with one drop of 0.1% SDS and 2 drops of chloroform and then mixed in a 1:1 ratio with PM2 buffer (70 mM Na2HPO4, 30 mM NaH2PO4, 1 mM MgSO4, 0.2 mM MnSO4; pH 7) containing 100 mM 2-mercaptoethanol. The cells were incubated for 5 minutes at 28°C and one volume of 0.4% ο-nitrophenol-β-D-galactopyranoside (ONPG) substrate in PM2 buffer was added to 4 volumes of cell suspension. After sufficient colour development, the reaction was stopped by addition of 2 volumes of 1 M NaHCO3. The OD420 and OD550 were obtained for each sample and β-galactosidase activity was calculated as units mg-1 dry weight bacteria .