Fig 1.
Model of phospholipid homeostasis systems in N. gonorrhoeae.
(A) In WT N. gonorrhoeae, MlaA, potentially interacting with an unknown partner, participates in retrograde trafficking of phospholipids from the outer leaflet of the outer membrane, through the periplasmic component of the system, MlaC, to the inner membrane MlaBDEF complex. The phospholipase PldA dimerizes to its active form upon detection of mis-localized phospholipids and removes the sn-1 and sn-2 fatty acid side chains. Different phospholipids are represented by lipid tails of different colors (phosphatidylglycerol [PG], yellow; phosphatidylcholine [PC], red; phosphatidylethanolamine [PE], blue). Native N. gonorrhoeae membrane phospholipid composition can be found in Table 4. (B) When MlaA is removed, phospholipids cannot be transported through the Mla system and invade the outer leaflet of the outer membrane. Increased amounts of membrane vesicles are also produced. (C) When the phospholipase PldA is overexpressed in the absence of MlaA, the PE substrate preference of PldA leads to a membrane phospholipid profile that is skewed toward PG, including in the outer leaflet of the outer membrane. OMV, outer membrane vesicle.
Fig 2.
Bioinformatic analysis of MlaA conservation and genome context.
(A) A phylogenetic tree of MlaA was constructed in MEGA using amino acid sequences of MlaA/VacJ homologs downloaded from NCBI. The Jones-Taylor-Thornton model was used to generate a pairwise distance matrix. Neighbor-Join and BioNJ algorithms were subsequently applied to the matrix to obtain the initial tree for a heuristic search. 500 bootstrap iterations were performed to test the phylogenies. The highest log-likelihood tree is presented. Homologs without lipoprotein signal peptides are represented by red branches; blue branches represent homologs with lipoprotein signal peptides. (B) The PubMLST Neisseria database was used to search for nucleotide polymorphic sites in the mlaA (NEIS1933) locus across 44,289 Neisseria isolates. (C) Local genome context of N. gonorrhoeae mlaA. Intergenic distances between open reading frames are noted above the schematic. The operon predicted by biocyc.org is noted below. Schematic and intergenic distances are not to scale.
Table 1.
Amino acid identity of MlaA homologs.
Fig 3.
Purification of a truncated recombinant version of MlaA and polyclonal anti-MlaA serum validation.
(A) Representative schematic of full length MlaA with annotated MlaA domain. A signal peptide (SP) is noted by a grey rectangle. (B) Schematic of truncated MlaA used for purification with first 119 amino acids removed and a 6 × Histidine tag (His; as a blue rectangle) added to the N-terminus. Schematics are not to scale. (C) Truncated MlaA construct was overexpressed in E. coli and purified by nickel affinity chromatography in the presence of 1% Triton-X 100 detergent. Detergent was subsequently removed by incubation with Bio-Rad Bio-Beads SM-2 resin. To assess purity, 1 μg of protein was subjected to 1D SDS-PAGE and visualized by staining with Colloidal Coomassie Blue G-250. Open arrow indicates migration band of MlaA120-277. Migration of a molecular weight marker (in kDa) is indicated to the left of the gel. (D) Immunoblot evaluation of anti-MlaA antiserum. Indicated amounts of purified MlaA120-277 were separated by SDS-PAGE, transferred to a nitrocellulose membrane, and probed with anti-MlaA antiserum. (E) Equivalent OD600 units of WT, isogenic ΔmlaA, and either ΔmlaA/Plac::mlaA or E. coli harboring the pGCC4-ngo2121 complementation plasmid cultured with indicated concentrations of IPTG, were separated by SDS-PAGE, transferred to a nitrocellulose membrane, and probed with anti-MlaA antiserum. Open arrow indicates MlaA. Non-specific cross-reactive band is marked with an asterisk (*). OD600, optical density at 600 nm; IPTG, isopropyl β-D-thiogalactopyranoside; SDS-PAGE, sodium dodecyl sulfate-polyacrylamide gel electrophoresis.
Fig 4.
Truncated, recombinant MlaA elicits broadly cross-reactive antisera that recognize MlaA in Neisseria species.
(A) 37 N. gonorrhoeae isolates, including common laboratory strains; clinical isolates collected in Baltimore between 1991 and 1994 and Seattle between 2011 and 2013; and the 2016 WHO reference strains were grown on solid media for 20 h at 37 °C in 5% CO2. Whole cell lysates were collected and subjected to immunoblotting analysis. (B) Whole cell lysates of different Gram-negative bacteria, including E. coli BL21(DE3); V. cholerae N19691; P. aeruginosa PA01; K. pneumoniae 6069; N. meningitidis MC58; the commensal bacterium N. lactamica NLI83/-01; and the opportunistic pathogen N. weaveri 1032 were subjected to immunoblot analysis. All lysates were standardized by OD600 values, separated in a 4–15% Tris-glycine gel, and probed with polyclonal rabbit antiserum against MlaA. FA1090 and ΔmlaA were included in blots as positive controls. Open arrow indicates MlaA. Non-specific cross-reactivity is marked with an asterisk (*). OD600, optical density at 600 nm; SDS-PAGE, sodium dodecyl sulfate-polyacrylamide gel electrophoresis.
Fig 5.
In vitro fitness assessments and MlaA expression profiling.
(A) WT FA1090, isogenic knockout ΔmlaA, and complementation strain ΔmlaA/Plac::mlaA were cultured aerobically in liquid medium. IPTG was added to 0.1 mM in ΔmlaA/Plac::mlaA cultures. Bacterial growth was monitored every hour by OD600 measurement. (B) Samples of WT FA1090 were collected at the times indicated. Whole cell lysates were separated by SDS-PAGE and probed with polyclonal rabbit anti-MlaA or anti-BamA (as a loading control) antisera. (C) Whole cell lysates of WT FA1090 were collected after 6 h culture in liquid medium containing desferal at concentrations ranging from 5–25 μM. Samples were probed with antisera against MlaA, MtrE, and TbpB. ΔmlaA cultured under standard conditions was included as a reference. (D) Densitometry analyses of MlaA using immunoblots from three independent desferal titration experiments shown as a representative blot in panel (C). (E) WT FA1090 and isogenic ΔmlaA were cultured in liquid medium containing either 10 μM (top panel) or 25 μM (bottom panel) desferal for 6 h. At each hour, samples were withdrawn and diluted for CFU/mL enumeration. Both graphs contain growth curves from cultures maintained under standard conditions (blue curves). Statistical significance was assessed by two-way ANOVA using Sidak’s multiple comparisons test. (F) WT FA1090 and isogenic knockout ΔmlaA were cultured in liquid medium under standard conditions until OD600 had at least doubled (~3 h). Cultures were standardized to an OD600 of 0.2 and dilutions were spotted onto GCB. Plates were prepared as normal (SGC); under iron deprivation (-Fe); supplemented with 7.5% normal human serum (+NHS); or supplemented with 1.2 mM NaNO2 and cultured anaerobically (-O2). Strains were maintained at 37 °C in 5% CO2 for approximately 22 h or at 37 °C anaerobically for 48 h and CFU were enumerated. (G) WT bacteria cultured as in (F), with the addition of a desferal titration from 0–25 μM under anaerobiosis, were collected from plates, separated by SDS-PAGE, and probed with polyclonal rabbit anti-MlaA antiserum or anti-BamA antiserum as a loading control. (H) Densitometry analyses of MlaA abundance under each host relevant condition presented in (G). Densitometry was performed twice on each of three independent experiments. (I) WT FA1090 and conditional knockout Δfur/Plac::fur were cultured in liquid medium in the absence (SGC) or presence of 25 μM desferal (-Fe). Fur expression was induced by the addition of 10, 50, or 100 μM IPTG. Samples were collected after 6 h of growth and probed with indicated antisera. (J) Densitometry analyses of MlaA abundance in immunoblots from three independent Fur induction experiments with and without iron starvation. (K) WT FA1090, isogenic knockout ΔmlaA, and ΔmlaA/Plac::mlaA were cultured aerobically in liquid medium until culture density had doubled (~3 h). Cultures were diluted to an OD600 of 0.05 in sterile PBS and diluted 1000-fold in EMEM. Suspensions were combined with an equal volume of EMEM, NHS, or heat-inactivated NHS and incubated for 1 h in 5% CO2 at 37 °C. Bacteria from each well were spotted onto GCB plates. CFUs were scored after 20–22 h incubation in 5% CO2 at 37 °C (p value between WT and ΔmlaA exposed to active NHS, 0.07). (L) Rapidly growing cultures of WT FA1090 and isogenic knockout ΔmlaA were diluted to 105 CFU/mL and cultured for 3 h in the presence of liquid medium (SGC), 0.01% acetic acid (vehicle), or 10 μM human defensin (HBD1). Bacteria were serially diluted and spotted onto GCB plates. CFUs were scored after 20–22 h incubation in 5% CO2 at 37 °C and survival was calculated relative to SGC. n≥3; mean ± SEM presented for all experiments; panels D, H, J, and L present values from each replicate; *p < 0.05; OD600, optical density at 600 nm; SGC, standard growth conditions; IPTG, isopropyl β-D-thiogalactopyranoside; SDS-PAGE, sodium dodecyl sulfate-polyacrylamide gel electrophoresis; CFU, colony forming unit; GCB, gonococcal base medium; GCBL, gonococcal base liquid medium; PBS, phosphate buffered saline; EMEM, Eagle’s minimal essential medium.
Fig 6.
Loss of MlaA results in a reduction in gonococcal colony size that is exacerbated in the presence of the antimicrobial peptide polymyxin B.
(A) Supernatants from mid-logarithmic cultures of WT and ΔmlaA bacteria were separated by low-speed centrifugation and filtration, treated with DNAseI, and precipitated with a pyrogallol red-molybdate-methanol procedure. Precipitated supernatants and whole cell lysates were standardized by the OD600 of the source culture, separated by SDS-PAGE, and probed with indicated antisera. (B) WT FA1090, isogenic knockout ΔmlaA, complementation strain ΔmlaA/Plac::mlaA, and PldA overexpression strain ΔmlaA/Plac::pldA were cultured aerobically in liquid medium for 3 h, back diluted to an OD600 of 0.1, cultured 2 h longer, serially diluted, and spotted onto GCB without (left column) or with (right column) polymyxin B (800 U/mL) and either without (top row) or with (bottom row) 0.5 mM IPTG. Dilution spots from each condition were imaged with a Bio-Rad ImageDoc system. (C) CFUs for permissive and restrictive conditions with or without IPTG were counted and relative viability was calculated. Experiment was performed on three separate occasions (mean ± SEM on graph; *p < 0.05), and typical plate images are presented. (D) Representative micrographs from 10−4 dilution taken with a Zeiss AxioObserver.D1 microscope at 10× magnification 0.25 Phase Contrast 1 of WT (Row 1), ΔmlaA (Row 2), and ΔmlaA/Plac::mlaA (Row 3). MlaA expression was induced by inclusion of 0.1 mM IPTG in the solid medium for the complementation strain. (E) Images of 10−4 dilution were also taken at 2.5× magnification 0.06 Phase Contrast 1 and colony diameters were measured with ImageJ software. Colonies were measured for each of two independent experiments for the − polymyxin B condition (WT, n = 548; ΔmlaA, n = 755; ΔmlaA/Plac::mlaA, n = 664) and for the + polymyxin B condition (WT, n = 836; ΔmlaA, n = 1197; ΔmlaA/Plac::mlaA, n = 1121; mean ± SEM on graphs; *p < 0.05). (F) Rapidly growing liquid cultures of WT, ΔmlaA, ΔmlaA/Plac::mlaA, and ΔmlaA/Plac::pldA incubated in the presence or absence of polymyxin B were lysed and treated with proteinase K to isolate LOS. Subsequently, LOS was separated by SDS-PAGE and visualized by silver staining. IPTG was added to ΔmlaA/Plac::mlaA and ΔmlaA/Plac::pldA cultures to 0.5 mM as indicated. DF, dilution factor; OD600, optical density at 600 nm; IPTG, isopropyl β-D-thiogalactopyranoside; LOS, lipooligosaccharide; SDS-PAGE, sodium dodecyl sulfate–polyacrylamide gel electrophoresis.
Table 2.
Etest assessments of cell envelope integrity.
Fig 7.
Investigations of MlaA effects on cell structure, membrane vesicles, and biofilm formation.
(A) Strains as indicated were cultured in liquid medium without (top row) or with (bottom row) polymyxin B until approximately mid-logarithmic growth. ΔmlaA/Plac::pldA was cultured in the presence or absence of 0.5 mM IPTG, as indicated. Bacteria were subsequently washed twice with PBS, spotted onto 300 mesh copper grids, negatively stained with phosphotungstic acid, and imaged using scanning electron microscopy. (B) Cytoplasmic/periplasmic (C/P), cell envelope (CE), membrane vesicle (MV), and soluble supernatant (SS) subcellular fractions of WT FA1090 collected under standard aerobic conditions were prepared and normalized based on protein concentration, separated by SDS-PAGE, and probed with indicated antisera. (C) MVs collected by ultracentrifugation of culture supernatants derived from strains indicated were quantified by protein concentration (n = 6 for WT and ΔmlaA for − polymyxin B condition; n = 3 for all others; mean ± SEM). (D) WT FA1090 and ΔmlaA bacteria were suspended to an OD550 of 1.5 in GCBL, added to 96 well microtiter plates, and cultured without shaking in 5% CO2 at 37 °C for 24 h. Planktonic bacteria were removed and biofilms were washed with PBS. Biofilms were allowed to dry at room temperature, then stained in 0.1% crystal violet in 2% ethanol. After staining, wells were washed with PBS and dried. Biofilms were dissolved in 30% acetic acid and quantified by A550 measurement. Biofilm experiments were performed 12 times, each with 3 or 4 technical replicates, for a total of 46 datapoints. Mean ± SEM is presented. IPTG, isopropyl β-D-thiogalactopyranoside; C/P, cytoplasmic/periplasmic; C, cytoplasm; CE, cell envelope; OM, outer membrane; MV, membrane vesicle; SS soluble supernatant; SDS-PAGE, sodium dodecyl sulfate-polyacrylamide gel electrophoresis; OD550, optical density at 550 nm; A550, absorbance at 550 nm, GCBL, gonococcal base liquid medium.
Fig 8.
MlaA influences gonococcal fitness in vivo.
(A) In vitro competition assays were performed by combining WT FA1090 bacteria with approximately equal numbers of ΔmlaA or ΔmlaA/Plac::mlaA (~106 CFU total bacteria). Competitions were carried out in liquid medium, and output CFUs were assessed at 2, 4, and 6 h post-inoculation. Competitions with the complemented strain were performed in liquid medium both with and without IPTG. (B) Female BALB/c mice were inoculated intravaginally with approximately equal numbers of CFUs of WT and ΔmlaA bacteria (~106 CFU total N. gonorrhoeae; 7 mice per group). Vaginal swabs were taken on days 1, 3, and 5 post-infection and were cultured for CFU/mL enumeration on solid media containing streptomycin (total bacteria) or media containing streptomycin and kanamycin (ΔmlaA bacteria). Experiments were repeated three times and results are expressed as the geometric mean of the competitive index (CI): [mutant CFU (output) / WT CFU (output)] / [mutant CFU (input) / WT CFU (input)]. A CI > 1 indicates that the mutant was more fit during the competition. 1 CFU was assigned for any strain not recovered from an infected mouse. IPTG, isopropyl β-D-thiogalactopyranoside; CFU, colony forming unit.
Fig 9.
Proteomic investigations of MlaA influence on cell envelope and membrane vesicles.
(A) CE and MV fractions isolated from WT FA1090 and isogenic knockout ΔmlaA were normalized based on protein concentration, separated by SDS-PAGE, and proteins were visualized by coomassie staining. The migration of a molecular weight marker is shown on the left in kDa. Proteins that appeared differentially abundant in the ΔmlaA mutant by visual inspection are labeled with an asterisk. (B) Trypsinized CE and MV proteins from WT and ΔmlaA were labeled with TMT6plex isobaric mass tags, fractionated by strong cation exchange and reverse phase chromatography, and subjected to peptide identification by tandem mass spectrometry. The number of differentially abundant proteins in the ΔmlaA CE or MV protein profiles is noted in the Venn diagrams. (C, D) Lists of differentially abundant proteins in the CE (C) or MVs (D) of the ΔmlaA mutant. Proteins in blue arrows are downregulated in the mutant, while those listed in red arrows are upregulated in the mutant. Proteins are arranged by the magnitude of the mutant:WT ratio. (E) Validation of quantitative proteomics results. CE and MV fractions from WT FA1090 and ΔmlaA were normalized by protein concentration, separated by SDS-PAGE, and probed with antisera against indicated proteins. SDS-PAGE, sodium dodecyl sulfate-polyacrylamide gel electrophoresis; TMT, tandem mass tag.
Table 3.
Differentially expressed proteins in ΔmlaA cell envelopes and membrane vesicles identified in two biological replicate experiments.
Table 4.
Phospholipid composition of N. gonorrhoeae and E. coli membranes.