Aminolipids elicit functional trade-offs between competitiveness and bacteriophage attachment in Ruegeria pomeroyi

Bacterial strains and cultivation

All marine bacteria used in this study were cultivated using the ½ YTSS (yeast-tryptone-sea salt) medium (DSMZ 974), containing yeast extract 2 g/L, tryptone 1.25 g/L and Sigma sea salts 20 g/L or the defined marine ammonium mineral salt (MAMS) medium (DSMZ 1313) where HEPES (10 mM, pH 8.0) replaced the phosphate buffer [16]. All cultures were grown at 30 °C aerobically in a shaker (150 rpm).

For growth competition assays between the WT and the olsA mutant, cultures of bacteria were grown in 10 mL ½ YTSS medium for the WT strain, or with the addition of 10 µg/mL gentamicin for the olsA mutant since a gentamicin cassette was inserted to construct the mutant [4]. Cells were harvested at mid-late exponential phase and diluted to an optical density measured at 540 nm (OD₅₄₀) of 1.0. These cells were then both inoculated at 1% (v/v) into 250 mL flasks containing 50 mL growth media (either ½ YTSS or MAMS + 0.5 mM Pi) in triplicate and grown at 30 °C with shaking at 140 rpm. At time point 0 h, 100 µL samples were removed in triplicate from each flask. These samples were then ten-fold serially diluted in the same growth media to a dilution of 10⁻⁹. From each serial dilution tube, 10 µL droplets were pipetted in triplicate onto agar plates containing either ½ YTSS agar (to count both the WT and the olsA mutant) or ½ YTSS agar + 10 µg/mL gentamicin (to count just the olsA mutant). Once the droplets were dry, plates were incubated at 30 °C for 3-4 days. Colony forming units (CFU) were determined by counting the number of colonies in the dilution number where single colonies were clearly visible. For the cultures grown in ½ YTSS medium, samples were removed and enumerated using the same method at time points 24 h and 96 h. For the cultures grown in MAMS media + 0.5 mM Pi, samples were removed and enumerated at time points 0 h, 48 h and 96 h.

Membrane separation by sucrose density gradient ultracentrifugation

The WT strain and the olsA mutant were grown in ½ YTSS medium to OD₅₄₀ ~0.8. One litre of culture was then collected by centrifugation at 12,300 × g at 4 °C for 10 minutes, using a JLA 10.5 rotor. Cells were washed and resuspended in 50 mL HEPES buffer (pH 8.0, 10 mM). Cells were then pelleted by centrifugation at 4,500 × g at 4 °C for 10 min, before resuspending the pellet in 3 mL HEPES buffer (pH 8.0, 10 mM), containing 1.6X cOmplete Protease Inhibitor cocktail (Roche), 3X DNAse I buffer (NEB) and 6 units/mL DNase I (NEB). Cells were then lysed using a French Press at 1000 PSI. Cell debris was removed by centrifugation at 4,500 × g at 4 °C for 10 min and the supernatant was transferred to a new Oakridge centrifuge tube for pelleting total membranes by centrifugation at 75,600 × g at 10 °C for 45 min in a JA25.5 rotor. Pelleted membranes were then washed and resuspended in 20% (w/v) sucrose in HEPES buffer (10 mM, pH 8.0). Resuspended membrane samples were then layered on top of a stepwise gradient containing 3.3 mL 73% (w/v) sucrose at the bottom and 6.7 mL 53% (w/v) sucrose in between. Inner (IM) and outer (OM) membranes were separated by centrifugation at 140,000 × g at 4 °C, for 16 hours in a SW40-Ti rotor. The IM resided in the interface between the 53% (w/v) and 20% (w/v) sucrose layers and the OM in the interface between the 53% (w/v) and 73% (w/v) sucrose layers. Both IM and OM samples were removed from the sucrose density interface, diluted with 30 mL HEPES buffer (10 mM, pH 8.0), and pelleted by centrifugation at 75,600 × g for 45 min. IM and OM were then resuspended in 1 mL of the same HEPES buffer before lipid and protein extractions.

Proteomics sample preparation, in-gel digestion and nanoLC-MS analysis

IM and OM samples were carefully dissolved in 100 μL 1X LDS loading buffer (Invitrogen) before loading on a precast Tris-Bis NuPAGE gel (Invitrogen) using 1X MOPS running solution (Invitrogen). SDS-polyacrylamide gel electrophoresis was run for approximately 5 min to purify polypeptides in the polyacrylamide gel by removing contaminants. Polyacrylamide gel bands containing the membrane proteome were excised and digested by trypsin (Roche) proteolysis. The resulting tryptic peptides were extracted using formic acid-acetonitrile (5%:25%, v/v) before resuspension in acetonitrile-trifluoroacetate (2.5%:0.05%, v/v). Tryptic peptides were separated by nano-liquid chromatography (nanoLC) using an Ultimate 3000 LC system with an Acclaim PepMap RSLC C18 reverse phase column (ThermoFisher) at the Proteomics Research Technology Platform (PRTP) at the University of Warwick. MS/MS spectra were collected using an Orbitrap Fusion mass spectrometer (ThermoFisher) in electrospray ionization (ESI) mode. Survey scans of peptides from m/z 350 to 1500 were collected for each sample in a 1.5-hr LC-MS run. This resulted in 12 mass spectra (3 biological replicates of IM and OM of WT and the olsA mutant) with a total of ~ 7.5 G of MS/MS data.

MS/MS data search and statistical analyses

Compiled MS/MS raw files were searched against the genome of Ruegeria pomeroyi DSS-3 using the MaxQuant software package [17, 18]. Default settings were used and samples were matched between runs. The software package Perseus (v1.6.5.0) was used to determine differentially expressed proteins with a false discovery rate (FDR) of 0.01 [19]. The LFQ (label-free quantitation) intensity of each protein was normalized by dividing the total peptide intensity of each sample by the length of each protein. Peptides were retained for further analyses only if they were consistently found in all three biological replicates in at least one set of the four samples (IM_WT, IM_olsA, OM_WT, OM_olsA). Missing values were imputed using the default parameters (width, 0.3; down-shift 1.8) and statistical analyses were performed using a two-sample Student’s t-test. Principle component analysis (PCA) plots and volcano plots were generated using default settings in the Perseus package.

To analyse the pathways of differentially expressed proteins between the wild-type and the mutant, the sequences of those proteins that were significantly overrepresented (FDR < 0.01) in each sample were annotated using the BlastKOALA program at the KEGG pathway mapping server (https://www.kegg.jp/blastkoala/) using the following settings (taxonomy group-Prokaryotes; KEGG genes databases- family_eukaryotes + genus_prokaryotes).

Lipidomics analysis of intact polar membrane lipids

Lipid extraction from bacterial cultures was carried out using the modified Folch extraction protocol as described previously [20]. Before lipid extraction from IM and OM samples, they were spiked with d17:1/12:0 sphingosyl phosphoethanolamine (SPE) to a final concentration of 500 nM. Total lipids were then extracted using methanol-chloroform, dried under nitrogen gas and the precipitate resuspended in acetonitrile: 10 mM ammonium acetate at 95:5 (v/v). Lipids were then analysed by LC-MS using a Dionex 3400RS HPLC with a HILIC BEH amide XP column (Waters) coupled with an amaZon SL ion trap MS (Bruker) via electrospray ionisation. After running on the LC-MS, the intensities of each sample peak were calculated using the Bruker QuantAnalysis program and compared against the intensity for SPE. Using the known concentration of SPE, these intensities were then compared to previously created external calibration curves (Supplementary Fig. S2) for each lipid species, including PG and PE. Where no calibration curve was available due to the lack of authentic reference material for a particular lipid species, intensities were compared to curves for lipid species which were structurally similar (e.g., PE for OL, QL, MMPE, DMPE and lysoPE). Using this information, the relative abundance of molar ratio of the lipid species in each membrane fraction could be calculated.

Minimum inhibitory concentration (MIC) testing

Single colonies of R. pomeroyi DSS-3 were cultured in ½ YTSS overnight at 30 °C with shaking (160 rpm). Fresh ½ YTSS was inoculated at a 2% ratio (v/v) and cultured in non-tissue culture treated 24 well plates (Falcon) in the presence of a serial 2-fold dilution of a series of antimicrobial chemicals. Plates were incubated at 30 °C with shaking (150 rpm) and endpoint readings at A₅₄₀ nm were taken at 24 or 48 h in a Fluostar Omega (BMG Labtech) plate reader.

Membrane permeability and hydrophobicity assays

To measure membrane permeability, crystal violet uptake was monitored using a previously published method [21]. Briefly, cells were grown to an OD₅₄₀ of 0.5–0.6, and 1 mL samples removed for analysis. A solution of crystal violet was prepared in phosphate buffered saline (PBS, pH 7.4) at 10 µg/mL and the OD₅₉₀ measured. Cells were washed by centrifugation at 4500 × g for 10 min, removing the supernatant and resuspending the pellet in the same volume of PBS buffer (pH 7.4). The wash step was repeated twice before resuspending the cell pellet in 1 mL PBS + crystal violet (10 µg/mL). Cell suspensions were then incubated at 30 °C for 10 min. After this, cells were harvested by centrifugation at 13,400 × g for 15 min, the supernatant removed to a 1 mL cuvette and OD₅₉₀ measured. To measure percentage crystal violet uptake, the following equation was used:

N-phenyl-1-naphthylamine (NPN) uptake was measured as described previously [22]. Briefly, cultures were grown in ½YTSS media at 30 °C to an OD₅₄₀ of ~ 0.5. Cells were centrifuged for 5 min at 3,000 × g and room temperature, before resuspending in half volume of HEPES 5 mM, pH 8.0. Reagents were added in triplicate to a Greiner Bio-One Black 96-well microplate as follows: (1) 200 µL HEPES; (2) 100 µL HEPES + 100 µL bacteria; (3) 150 µL HEPES + 50 µL NPN (40 µM); (4) 50 µL HEPES + 50 µL bacteria + 50 µL NPN (40 µM). All components apart from bacteria were added to the wells in advance, then the bacteria were added immediately before measurement and the fluorescence values began recording within 1 minute. Fluorescence was measured using a BMG Labtech Fluostar Omega microplate reader, with filters set to excitation at 355 nm and emission at 460 nm. All measurements were performed at room temperature. Samples were measured in plate mode every 30 seconds for 10 minutes total.

Cell hydrophobicity was measured using a previously published method [15]. Briefly, cells were grown in ½ YTSS media to an OD₅₄₀ of 0.5–0.6. To a 1.2 mL cell sample, 200 µL hexane were added and samples mixed for 1 min. After waiting 5–10 min to allow phase separation, the lower aqueous phase was removed to a 1 mL cuvette and OD₅₄₀ measured. The percentage adhesion of cells to the organic phase was calculated as follows:

({{{{{{{mathrm{Hydrophobicity}}}}}}}} = frac{{({{{{{{{mathrm{OD}}}}}}}}_{540};{{{{{{{mathrm{of}}}}}}}};{{{{{{{mathrm{the}}}}}}}};{{{{{{{mathrm{initial}}}}}}}};{{{{{{{mathrm{bacterial}}}}}}}};{{{{{{{mathrm{suspension}}}}}}}} – {{{{{{{mathrm{OD}}}}}}}}_{540};{{{{{{{mathrm{of}}}}}}}};{{{{{{{mathrm{the}}}}}}}};{{{{{{{mathrm{aqueous}}}}}}}};{{{{{{{mathrm{phase}}}}}}}})}}{{{{{{{{{mathrm{OD}}}}}}}}_{540};{{{{{{{mathrm{of}}}}}}}};{{{{{{{mathrm{the}}}}}}}};{{{{{{{mathrm{initial}}}}}}}};{{{{{{{mathrm{bacterial}}}}}}}};{{{{{{{mathrm{suspension}}}}}}}}}} times 100)

¹⁴C-DMSP synthesis and measurement of radioactive and molar concentration

¹⁴C-labelled dimethylsulfoniopropionate ([1-¹⁴C]-DMSP) was synthesised from dimethylsulfide and ¹⁴C-labelled acrylic acid as described elsewhere [23]. To determine radioactive concentration, 1 μL of synthesized ¹⁴C-DMSP was added to 3 mL scintillation fluid (EcoLume Liquid Scintillation Cocktail, MP Biomedicals) and measured on a liquid scintillation analyzer (Tri-Carb 2800TR, PerkinElmer) after equilibration overnight. The radioactive concentration of ¹⁴C-DMSP was measured independently three times and the activity was found to be 0.448 kBq μL⁻¹. The identity and molar concentration of DMSP was determined by LC-MS according to a previously published method [24]. ¹⁴C-DMSP stock in methanol:chloroform:water (12:5:1, v/v) was spiked with 200 nM of deuterated glycine betaine (d₁₁-GBT, purchased from Cambridge Isotope Laboratories Inc.) as internal standard (ISTD). Samples were separated on a Discovery HS F5 column (Supelco) with guard column (HS F5 Supelguard) (both obtained from Sigma-Aldrich) using a Dionex 3400RS HPLC and masses detected using an amaZon SL ion trap MS (Bruker) via ESI in positive ion mode. Peak areas of DMSP were detected at m/z 135 and of ISTD at m/z 129 and determined in each sample using Bruker QuantAnalysis software. The peak area ratio of DMSP:ISTD for each sample was compared to a calibration curve (Supplementary Fig. S3). Therefore, triplicate DMSP standards (dimethylpropiothetin hydrochloride, Supelco via Merck) covering a concentration range from 0.005–2 μM in methanol:chloroform:water (12:5:1, v/v) were spiked with 200 nM d₁₁-GBT, measured and peak areas determined as described above. The ratio of DMSP to ISTD allowed to quantify the molar concentration of DMSP in the ¹⁴C-DMSP stock, which was 328.8 ± 15.2 mM.

Uptake assays of ¹⁴C-labelled compounds

A single colony of the WT and the olsA mutant was inoculated into 5 mL marine broth 2216 (MB) medium, grown for ~ 24 h at 30 °C with shaking (170 rpm). For growth in ½ YTSS, cells were pelleted by centrifugation (1000 x g, 3 min), washed twice in ½ YTSS and inoculated into 50 mL ½ YTSS medium at a 1% (v/v) ratio. Cells were grown for 4 h at 30 °C with shaking (170 rpm) until an OD₅₄₀ of ~0.2 was reached and used for the choline uptake assay. For growth in MAMS medium with 2 mM Pi, cells were pelleted by centrifugation (1000 x g, 3 min), washed twice in MAMS medium with 2 mM Pi and inoculated at a 1% ratio (v/v) into 50 mL MAMS with 2 mM Pi. Cells were grown for 5 days at 30 °C with shaking, then pelleted by centrifugation (1000 x g, 3 min), washed once in MAMS with 2 mM Pi and inoculated at a 1% ratio (v/v) into 50 mL fresh MAMS with 2 mM Pi. For growth in MAMS medium with 0.5 mM Pi, cells were pelleted, washed and pre-grown in MAMS with 2 mM Pi as described above. Then cells were pelleted by centrifugation (2000 x g, 3 min) and washed once in MAMS medium with 0.5 mM Pi. WT cells were inoculated at a 1% ratio and olsA mutant cells at a 10% ratio (v/v) into 50 mL fresh MAMS with 0.5 mM Pi. Cultures were grown until an OD₅₄₀ of 0.2–0.4 was reached, then used for the choline uptake assay.

For uptake assays using [methyl-¹⁴C] choline (55.2 mCi mmol⁻¹, Perkin Elmer), cells were diluted 1:1 (v/v) in the same fresh medium and 8 concentrations (0.5–25 µM) of choline (¹⁴C:¹²C at 1:1000, v/v) were added to 5 mL culture from three biological replicates. Cells were incubated at 30 °C with shaking (130 rpm) for 5 min, harvested by filtration onto 0.2 µm pore size Supor filters (∅ 25 mm diameter, Pall Corporation) and filters washed 3x with 1 mL of the corresponding medium. Filters were transferred into 6 mL scintillation vials, covered with 3 mL scintillation fluid (EcoLume Liquid Scintillation Cocktail, MP Biomedicals) and measured on a liquid scintillation analyser (Tri-Carb 2800TR, PerkinElmer) after equilibration overnight. For background correction, cells were fixed in a final concentration of 2% (v/v) paraformaldehyde solution for a minimum of 15 min at 4 °C in the dark prior to radioisotope addition. Maximum uptake velocity (V_max) and Michaelis constant (K_m) were calculated based on Lineweaver-Burke transformation.

For DMSP uptake assays, the WT and the olsA mutant were grown for ~24 h at 30 °C with shaking (170 rpm) before inoculating at a 1% v/v ratio into 50 mL fresh marine broth. Cells were grown until an OD₅₄₀ of ~0.5–0.8 was reached before the DMSP uptake assay was carried out. The relatively low radiochemical concentration of the ¹⁴C-DMSP stock (0.448 kBq μL⁻¹, 329 mM) made it impractical to use less than 50 µM DMSP stock for the uptake kinetic assays since the radioactivity would be below the detection limit of the scintillation analyser (Tri-Carb 2800TR, PerkinElmer). Therefore, three high DMSP concentrations (50, 150, and 300 µM) were added to 3 mL of undiluted culture from three biological replicates. Cells were incubated and processed as described above and counts of ¹⁴C-DMSP for WT and olsA mutant cultures measured on the scintillation counter.

To normalise uptake rates, bacterial cell counts were determined by flow cytometry as described previously [25]. Subsamples of 1 mL from each culture were fixed with glutaraldehyde (Electron microscopy grade, BDH) (0.5% (v/v) final concentration) for 30 min at 4 °C, snap frozen in liquid nitrogen and stored at −80 °C. Samples were melted for 5 min at 37 °C and stained with SYBR gold (Invitrogen) (final concentration 10⁻⁴ of commercial stock) for 10 min at 60 °C in the dark. Samples were diluted between 10- to 50-fold (choline) and 100- to 200-fold (DMSP) in sterile TE buffer (10 mM Tris-HCl, 1 mM EDTA, pH 8.0). Three analytical replicates of each sample were counted on a CytoFLEX flow cytometer (Beckman Coulter) equipped with a 50 mW 488 nm solid-state diode laser and standard filters at a flow rate of 30 µL min^‑1 for 60 s with the discriminator set to 525 nm (green fluorescence).

Bacteriophage adsorption assays

To determine the impact of the olsA mutant on the interaction of the bacterium with bacteriophages, the rate of adsorption (k) of bacteriophage to the R. pomeroyi DSS-3 WT and the olsA mutant was compared. Bacteriophage DSS3Phi2 was used in this experiment, the first known bacteriophage infecting R. pomeroyi DSS-3 [26]. Briefly, culture flasks containing 9 mL bacterial cells at an OD₅₄₀ of 0.5 were kept in a 30 °C water bath, with another flask containing 9 mL ½ YTSS media as a control. After allowing the temperature of the flasks to equilibrate, DSS3Phi2 was added to the flasks at an MOI of 0.1, at which point a timer was started. At time points of 0, 2, 5, 10, 20, and 30 min post-infection, 100 μL bacteria from each flask was removed and added to an Eppendorf tube containing 890 μL ½ YTSS media and 10 μL chloroform, and kept on ice. After the final time point was taken, all samples were centrifuged for 5 minutes at 2000 × g and 4 °C to pellet the bacterial cells and the adsorbed phage. The supernatant was then removed into a fresh Eppendorf tube and kept at 4 °C. To determine the phage titre, 100 µL supernatant was removed and serially diluted in Eppendorf tubes containing 900 µL ½ YTSS media down to 10⁻⁹. 900 µL was removed from the 10⁻⁴ to 10⁻⁹ dilutions and added to a 6-well plate. To each well, 1 mL host culture (wild type R. pomeroyi DSS-3) at an OD₅₄₀ of approximately 0.3 was added, as well as 4 mL soft ½ YTSS agar (0.8% w/v). Plates were allowed to dry before being incubated overnight at 30 °C. The next day, wells containing 10–100 plaques were counted by eye and used to calculate initial phage titre. The phage adsorption constant was calculated using the following equation ln(P/P₀) = - kNt where P is the free phage titre after adsorption, P₀ is the total bacteriophage titre, k is the adsorption constant (ml/ min), N is the bacterial density (cells/ml) and t is time (min) [27].

Source: Ecology - nature.com