Molecular Characterization and Functional Analysis of Outer

Jan 19, 2014 - ... Ch. V. B. Swamy, and Medicharla V. Jagannadham*. CSIR - Centre for Cellular and Molecular Biology, Tarnaka, Hyderabad - 500007, Ind...
0 downloads 0 Views 4MB Size
Article pubs.acs.org/jpr

Molecular Characterization and Functional Analysis of Outer Membrane Vesicles from the Antarctic Bacterium Pseudomonas syringae Suggest a Possible Response to Environmental Conditions Heramb M. Kulkarni, Ch. V. B. Swamy, and Medicharla V. Jagannadham* CSIR - Centre for Cellular and Molecular Biology, Tarnaka, Hyderabad - 500007, India

Downloaded via AUCKLAND UNIV OF TECHNOLOGY on January 29, 2019 at 10:31:34 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.

S Supporting Information *

ABSTRACT: Outer membrane vesicles (OMVs) of Gramnegative bacteria form an important aspect of bacterial physiology as they are involved in various functions essential for their survival. The OMVs of the Antarctic bacterium Pseudomonas syringae Lz4W were isolated, and the proteins and lipids they contain were identified. The matrix-assisted laser desorption/ionization time of flight (MALDI-TOF/ TOF) analysis revealed that phosphatidylethanolamines and phosphatidylglycerols are the main lipid components. The proteins of these vesicles were identified by separating them by one-dimensional gel electrophoresis and liquid chromatography coupled to electrospray ionization tandem mass spectrometry (ESI−MS/MS). They are composed of outer membrane and periplasmic proteins according to the subcellular localization predictions by Psortb v.3 and Cello V2.5. The functional annotation and gene ontology of these proteins provided hints for various functions attributed to OMVs and suggested a potential mechanism to respond to the extracellular environmental changes. The OMVs were found to protect the producer organism against the membrane active antibiotics colistin and melittin but not from streptomycin. The 1-Nphenylnapthylamine (NPN)-uptake assay revealed that the OMVs protect the bacterium from membrane active antibiotics by scavenging them and also showed that membrane and protein packing of the OMVs was similar to the parent bacterium. The sequestering depends on the composition and organization of lipids and proteins in the OMVs. KEYWORDS: outer membrane vesicles, proteomics, lipidomics, Antarctic bacterium, MALDI-TOF/TOF, ESI−MS/MS, antibiotic activity, OMV biogenesis



INTRODUCTION

Adaptation of bacteria to stress is associated with the release of various types of products by the cells. Investigations performed during the past decade led to the discovery of important roles of OMVs in secretion, self-defense, biofilm formation, cell-to-cell communication and bacterial pathogenicity.14−16 Hence, it appeared worthwhile to undertake detailed studies on the molecular composition and functions of OMVs produced by a bacterial isolate Pseudomonas syringae Lz4W, which is adapted to the extreme harsh climate of Antarctica. Recently, the protective role of OMVs against various types of physical and chemical stress is beginning to be understood. Microorganisms sense antibiotics as stressors.17 Role of OMVs in resistance to some antibiotics was reported earlier.18 Recent studies recognize that bacteria isolated from fresh water lakes or bacteria isolated from different places of Antarctica showed resistance to some antibiotics.19,20 These regions are not exposed to antibiotics. How these bacteria acquired antibiotic

Outer membrane vesicles (OMVs) are spherical membrane bilayer structures, with diameters ranging from 20 to 250 nm, discharged predominantly by Gram-negative bacteria during growth by a mechanism that is not clearly understood.1,2 They generally composed of proteins, lipids, and lipopolysaccharides, and in some cases they also contain either RNA or DNA.3−5 Several studies have been carried out from time to time to characterize the contents of the OMVs of different types of bacteria.6−8 The number of proteins carried by OMVs differed largely from different bacterial strains. The OMVs of Pseudomonas aeruginosa contained 338 proteins,9 and those of Pseudoaltereomonas antarctica NF3 contain 44 proteins.10 Escherichia coli OMVs contain 141 proteins,6 and those of Francisella novicida contain 416 proteins.7 A recent study revealed that OMVs extruded from an Antarctic bacterium Shewanella vesiculosa M7 produces outer-inner membrane vesicles (O-IMV) that contain 46 proteins and DNA.11 Proteomics studies on the OMVs of extremophilic bacteria are very few.12,13 © 2014 American Chemical Society

Received: September 10, 2013 Published: January 19, 2014 1345

dx.doi.org/10.1021/pr4009223 | J. Proteome Res. 2014, 13, 1345−1358

Journal of Proteome Research

Article

phosphate buffer pH 7.4. The OMVs were further purified by sucrose density gradient centrifugation as described earlier27 with slight modifications. Layers of equal volumes of 70%, 60%, and 20% sucrose were added in dispensable polyallomer tubes from bottom to top, and the OMV preparation was added on the top of the layers. The tubes were ultracentrifuged in Beckman SW 60Ti at 35 000 rpm (164609g) at 4 °C for 6 h. Different fractions were collected and diluted 30 times in 10 mM phosphate buffer pH 7.4, and the presence of OMVs was detected by measuring the size of the nanoparticles using dynamic light scattering. The fractions containing OMVs were pooled together and ultracentrifuged again in Beckman Type 60 Ti for 90 min at 50 000 rpm (250000g) at 4 °C. The pellet thus obtained was reconstituted in phosphate buffer and stored at −20 °C until further use. The yield of OMVs was determined by estimating the amount of protein by using Bio Rad Bradford protein estimation kit. OMVs prepared from 1 L of culture were found to contain 300 μg of protein.

resistance is not clearly known. Hence, it appeared worthwhile to test the efficacy of OMVs produced by an organism in protecting the producer cells from the effect of different types of antibiotics and to correlate the observations in terms of the information obtained from structural studies on the vesicles. The model organism chosen for the present investigation was Pseudomonas syringae Lz4W, a cold-tolerant bacterium obtained from a lake of Antarctica, characterized earlier at our Institute.21 It is a thoroughly studied organism with a known profile of resistance and sensitivity to various antibiotics. The membrane proteome has also been identified.22−24 Recently, its genome sequence was also published.25 Hence, it has been possible to detect the proteins in OMVs obtained from this organism and to predict their function. In the present study, the OMVs of this Antarctic bacterium P. syringae Lz4W were isolated, and their functional significance was studied. The proteins from these vesicles were identified using LC-coupled electrospray ionization tandem mass spectrometry (ESI−MS/MS) analysis. From the bacterial growth curve experiments, it is found that externally added OMVs prepared from the bacterium offer protection from the growth inhibitory effects exerted by the membrane active antibiotics colistin and melittin by sequestering them. They could not protect from streptomycin that inhibits the protein synthesis.



Dynamic Light Scattering

The size distribution of the OMVs was monitored by using DLS, for which the preparation diluted with phosphate buffer to a protein concentration of 0.06 μg/mL was used to measure the size distribution. The scatter was recorded by Photocor instrument (College Park, MD, USA) at 90° angle at 22 °C with a laser of wavelength 632 nm. The data were analyzed by Dynals software to obtain the average hydrodynamic radius of the OMVs.

MATERIALS AND METHODS

Chemicals and Reagents

Quantification of OMVs

Bradford protein estimation kit, glycine, urea, thiourea, iodoacetamide, and DTT were obtained from Bio Rad (Hercules, CA, USA). Protein molecular weight markers (broad range) used in SDS-PAGE analysis was purchased from Takara (Ishiyama, Japan). Ammonium thiocyanates, 2,5dihydroxybenzoic acid, and iodine crystals were purchased from Sigma-Aldrich (St. Louis, MO, US). The thin layer chromatography precoated silica plates were obtained from Merck (NJ, USA). All other chemicals used are of high purity grade.

The yield of OMVs was quantified per colony forming unit (CFU) of the culture by using protein- and lipid-based assays as described earlier.28 The protein content of the preparation was determined by the Bradford method using Biorad protein estimation kit, and the lipid content was measured by a fluorescence assay using a lipophilic dye FM4-64 as described previously.13 The CFUs/mL of the culture at the time of harvest was counted by diluting and spreading onto ABM agar plates. We have harvested 500 mL of bacterial culture; it was estimated that it contained 3.5 × 1016 CFUs, and the yield of OMVs prepared from this culture was estimated to be 170 μg of protein. Thus, the yield of OMVs was 47.8 μg/1016 CFUs. The lipid to protein ratio (w/w) was found to be 0.455 in the OMVs of P. syringae Lz4W.

Bacterial Strain and Growth Condition

The bacterium used in the present study was isolated from the soil samples collected in and around the lake Zub, Schirmacher Oasis, Antarctica. The geographical coordinates of the sampling area are 70°45′12″ S and 11°46′E. The Antarctic bacterial strain P. syringae Lz4W can grow between 0 to 30 °C with an optimal growth at 22 °C.21 The organism has been elaborately studied for its nutrient requirements and the optimal growth temperature, and it was found to be 22 °C.26 The bacterium was grown at 22 °C in Antarctic bacterial medium (ABM) that contains 0.5% peptone and 0.2% yeast extract, with continuous shaking at 150 rpm up to late stationary phase (OD at 600 nm = 1.1−1.2) for the preparation of OMVs.

Transmission Electron Microscopy (TEM)

The OMVs were visualized by using TEM using a Jeol transmission electron microscope (JEM 2100, Tokyo, Japan) at 200 kV. The sample was taken on a carbon coated grid, negatively stained with 0.2% uranyl acetate for observing the vesicles. Images were captured by a Gatan Camera. Purification and Detection of Lipopolysaccharides from the OMVs

Preparation of OMVs from P. syringae Lz4W

OMVs (100 μg) were resuspended in a lysis buffer (20 μL), which is composed of 2% SDS, 4% 2-mercaptoethanol, 10% glycerol, and 0.1 M tris HCl (pH 6.8) and allowed to boil for 15 min. Then lipopolysaccharides (LPS) was pelleted by ultracentrifugation (105000g, 12 h, 15 °C) and washed with phenol/chloroform/petroleum ether (2:5:8 v/v/v) and acetone. The LPS thus purified was dried in a vacuum.29 Proteinase K (10 μg) was added, and the vial was incubated at 60 °C for 1 h. The presence of LPS was detected employing SDS-PAGE analysis at a constant current of 35 mA as described

The OMVs were prepared from P. syringae Lz4W as described earlier with minor modifications.27 Briefly, the bacteria were centrifuged at 10000g for 15 min at 4 °C. The supernatant was collected and filtered through 0.45 μm membrane filter from Millipore (Billerica, US). The filtrate thus obtained was subjected to ultracentrifugation at 35 000 rpm (142032g, RCF (maximum)) for 2 h at 4 °C in Type 45 Ti in Beckman Ultracentrifuge. The outer membrane vesicles which settled as a pellet were then resuspended in a minimum volume of 10 mM 1346

dx.doi.org/10.1021/pr4009223 | J. Proteome Res. 2014, 13, 1345−1358

Journal of Proteome Research

Article

earlier.30,31 LPS were detected calorimetrically by using 2-keto3-deoxyoctonate (KDO) assay as described earlier.32,33

instrument obtained from Thermo Scientific (San Jose, CA). The peptides were separated using Proxeon LC system on a Biobasic C18 (100 mm × 0.18 mm) reverse phase nanocolumn, with a pore size of 300 Å and particle size of 5 μm (New Objective, MA, USA) on a 90 min gradient. The LC system was connected to ESI−MS/MS which recorded the collisioninduced dissociation (CID) MS of the peptides. The mass resolution of the precursor ion scans is 60000. The flow rate was set at 350 nL/min. The mobile phases A and B were 0.2% formic acid in water and 0.2% formic acid in 95% acetonitrile, respectively. The gradient was started at 10 min and increased to 60% B in 40 min and to 100% B in 55 min. The MS and MS/MS spectra were obtained at a heated capillary temperature of 200 °C, and the electrospray ionization (ESI) voltage was set at 1.6 kV. The peptides were fragmented using normalized collision energy of 35%. The MS/MS spectrum of the top 20 peptides with a signal threshold of 500 counts was acquired with 10 ms activation time and a repeat duration of 30 s.

Identification of Phospholipids

The phospholipids from the OMVs were obtained by chloroform/methanol extraction34 and separated by thin layer chromatography (TLC), using a solvent system containing chloroform/methanol/water in the ratio of 65:25:4 respectively. The spots were visualized by staining with iodine. The lipids from the spots were re-extracted in a chloroform− methanol mixture (10:3) and dried under nitrogen. Analysis of the phospholipids by matrix-assisted laser desorption/ionization time of flight (MALDI-TOF/TOF) was carried out as described earlier.35 The samples were analyzed on 4800 MALDI TOF/TOF obtained from Applied Biosystems (Foster City, CA) equipped with Nd: YAG laser (355 nm). The mass spectra were recorded in positive ion reflector mode. The CID mass spectra were recorded for different peaks that were manually selected from MS spectra. The collision gas used was air. Mass spectra were recorded in the mass range m/z 500− 1500. The matrix 2,5-dihydroxybenzoic acid (1% w/v) in acetonitrile containing 0.1% TFA was used. The fatty acid methyl esters (FAMEs) analysis of the OMVs was carried out by using a procedure already described in the literature.36 In brief, the fatty acids present in the lipids of OMVs were converted to FAMEs. The electron impact mass spectra were recorded by using GC-MS (column HP-5, 25 m long × 0.2 mm id GC column) and identified by using Wiley library of mass spectra corresponding to standard FAMEs. The structures of identified PLs were predicted by using the text/ontology based search in Lipid Maps Structure Database (LMSD).37 The lipid structures identified agreed well with the identified headgroup in MALDI-TOF/TOF and the fatty acids analysis by GC-MS.

Identification of Proteins

The draft genome sequence of the Antarctic bacterium P. syringae Lz4W shows that it contains 4450 entries and deposited at DDBJ/EMBL/GenBank under the accession number AOGS00000000.25 The MS/MS spectra of the multiply charged peptides were searched against this database of P. syringae Lz4W. The LC−ESI MS/MS spectra were analyzed using the Proteome Discoverer Version 1.4 supplied by the manufacturer. All the MS/MS spectra were analyzed using SEQUEST (Thermo Fisher Scientific) and MASCOT selecting the enzyme trypsin and applying the search parameters of precursor tolerance of 10 ppm and a fragment tolerance of 0.8 Da. The increase in mass due to the oxidation of methionine (15.99 Da) and carboxyamidomethylation of cysteine (57.02 Da) was set as variable and fixed modification, respectively. Only peptides identified with high confidence were included in the list. All the proteins were identified by at least two unique peptides. Proteins were also identified by Peaks 6 software (Peaks DB), by using the same search parameters. Peaks algorithm determines the de novo sequencing of the peptides from the MS/MS spectra, and these sequences are matched against the database for identification of the proteins. Average local confidence (ALC) score set for acceptance was at 30 for de novo sequences. Since the total local confidence depends upon the length of individual peptide, it was not considered for setting any cutoff. The −10 log P score was set at 50 for accepting the match from the database search. The probability of a wrong hit at this score is ≤0.001%. The proteomics data have been deposited to the ProteomeXchange consor tium with identifier PXD000221 (http:// proteomecentral.proteomexchange.org) via the PRIDE partner repository.39 The proteins identified by different algorithms SEQUEST, MASCOT, PEAKS and were combined; proteins common to all three programs were accepted as confident hits and used for further analysis.

Preparation of Outer Membrane, Inner Membrane, and Total Cell Lysate Proteins from P. syringae Lz4W

A cell lysate (CL) of P. syringae Lz4W was prepared as described earlier.23 Briefly, the cells were resuspended in 20% sucrose in 30 mM tris buffer (pH 8.0). Then lysozyme (60 μg/ mL), 0.7 mg of DNase and RNase were added and the cells were sonicated for 5 min in a Branson sonifier (Thomas Scientific, New Jersey, USA). The cell debris and the remaining intact cells were removed by centrifugation at 8000 rpm for 10 min. The supernatant was loaded on a two-step gradient of sucrose (60 and 70% sucrose layers) as described above for OMVs preparation, and the ultracentrifugation was carried out. The inner membrane (IM) proteins and outer membrane proteins (OM) appeared as thick bands at the boundary of 20− 60% and 60−70% sucrose respectively. They were collected separately and stored at −20 °C. The proteins from CL, IM, and OM were estimated by Bio Rad Bradford kit using BSA as standard. The proteins prepared from subcellular fractionations were analyzed on SDS-PAGE to identify their profiles according to Laemmli’s method.38 The proteins were precipitated with acetone and redissolved in loading buffer before PAGE. The samples (∼100 μg in each well) were separated on 12% SDS PAGE and stained by Coomassie Blue.

Estimation of Protein Abundances

Exponentially Modified Protein Abundance Index (emPAI) was estimated using Protein pilot software 4.0.8085 obtained from Ab Sciex. The Proteinpilot software uses MASCOT algorithm for protein identification. It calculates the abundances of the proteins in a given mixture as emPAI on the basis of the ratio of number of peptides observed for a protein in MASCOT and the number of observable peptides for the proteins in the case of a given proteolytic enzyme and mass range as described

Mass Spectrometry of Trypsin Digested Proteins

The proteins from OMV of P. syringae Lz4W were separated on a one-dimensional (1D) gel, excised into 8−10 pieces, and digested with trypsin using standard protocols as described earlier.23 The tryptic peptides were subjected to mass spectrometry using an LC−ESI−MS/MS Orbitrap velos 1347

dx.doi.org/10.1021/pr4009223 | J. Proteome Res. 2014, 13, 1345−1358

Journal of Proteome Research

Article

earlier.40 The raw spectral data from Thermo Orbitrap was converted to .mgf format and submitted to Proteinpilot, and the results were obtained in the “Protein Families” view. The search parameters for protein identification are identical as described above in this section. The emPAI for all the identified proteins are shown in Supplemental Table S1, Supporting Information.

ditions. These antibiotics were added above their respective growth inhibitory concentrations to ensure that bacteria cannot grow at this concentration. The qualitative assays on agar plates were carried out to examine the effect of OMVs on the growth of P. syringae Lz4W in the presence of antibiotics. In a 96-well plate, the culture medium, inoculums, antibiotics and OMVs were added as described in the growth curves experiment. In brief, the positive control contained only inoculated medium (104 CFU/mL), and negative control contained inoculated medium with growth inhibiting concentration of the respective antibiotic. The wells contained the inoculated medium with growth inhibiting concentration of the antibiotic to which different concentrations of the OMVs from P. syringae Lz4W were added. For qualitative studies, the constituents of the wells were streaked on an 80 mm ABM agar plate segmented into eight sections and labeled. The plates were incubated at 22 °C for 24−30 h.

Prediction of Subcellular Localizations

The subcellular localizations of the proteins identified from OMVs and the total proteins of P. syringae Lz4W were predicted by using the online tool Psortb v.3 (http://www. psort.org) which integrates the information of amino acid composition, similarity to proteins with known localizations, and presence of localization-specific motifs. The sequences of the identified proteins were submitted in fasta format to this tool by choosing Gram-negative bacteria, and results were obtained in normal output format. The predictions having scored more than 7.5 out of 10 were accepted as the final results. However, this algorithm cannot predict the localization of lipoproteins, and several proteins were predicted as unknown localization. Hence, another prediction algorithm Cello was used to predict the subcellular localizations.41 The proteins predicted as OM by both the algorithms was considered as most confident. The proteins predicted different localization sites by Cello V2.5, but clearly predicted as OM proteins by Psortb were also considered. The proteins whose localization could not be predicted by Psortb but predicted by Cello as OM proteins were also considered.

Efficacy of OMVs in Restoring Growth in Presence of Antibiotics

The contents of the wells (described above) were diluted 100 times in ABM and 25 μL of it was spread plated onto an 80 mm ABM agar plate for viable count. The quantitative plate assays were performed to estimate the extent of protection provided by OMVs to the cells in the presence of growth inhibiting concentration of antibiotics. The colonies grown on the plates were counted following incubation at 22 °C for 24−30 h. The experiment was repeated three times, and an average of the colony count was obtained for each OMVs concentration. The corresponding viable counts were calculated. No growth was observed in the antibiotic-containing plate. The viable count observed in the absence of antibiotics was taken as 100%. The percentage of survivors in the presence different concentrations of OMVs and various antibiotics was calculated by counting the CFUs.

Functional Annotation of the Identified Proteins

The Gene Ontology’s and functional annotation for the identified proteins was obtained from the Universal Protein Knowledge Base (Uniprot KB). This database provides all the available current information on the protein of interest.42 Thus, the Gene Ontology, the pathways in which the proteins are involved, the concerned biological processes, their subcellular localizations, and molecular functions were retrieved wherever available.

NPN-Uptake Assay

To understand the mechanism of OMVs in protecting the bacterium 1-N-phenylnapthylamine (NPN) uptake assay was performed as a function of different concentrations of antibiotics. Six micrograms of OMVs was suspended in 1 mL of 10 mM phosphate buffer pH 7.4, containing 5 μM NPN. Increasing concentrations (0.5−5 μg) colistin was added, and the changes in fluorescence intensity of NPN were recorded. The excitation wavelength for NPN was fixed at 356 nm, and the emission spectra were recorded from 370 to 550 nm by setting excitation and emission slit widths at 5 nm. Similar experiments were carried out using melittin and streptomycin. All the experiments were performed at 22 °C on Hitachi fluorescence spectrometer F4500 (Tokyo, Japan).

Effect of OMVs on the Growth of Bacteria in the Presence of Antibiotics

The effect of antibiotics on the growth patterns of P. syringae Lz4W in the presence of OMVs prepared from this bacterium was monitored using Bioscreen C growth monitoring system as described earlier.43 An identical amount of inoculums as described earlier and varying amount of OMVs were added to different wells of a honeycomb plate. The antibiotics with different mechanisms of activity such as the membrane active antibiotics colistin, melittin, and streptomycin which prevent bacterial growth by inhibiting protein synthesis, were selected for the study. The growth inhibiting concentrations of colistin, melittin, and streptomycin were determined by broth dilution method44 and modified for Bioscreen C growth monitoring system (Helsinki, Finland). The medium was inoculated by using 1% inoculum having cell density of 2 × 105 CFU/mL. The antibiotics were diluted to a range of different concentrations in different wells. The readings of growth curves for each well were recorded by using Bioscreen C. The growth temperature was maintained at 22 °C with constant shaking on “low” mode (120 rpm). The concentration of antibiotic at and above which the cells fail to grow was identified as the growth inhibiting concentration for the respective antibiotic−bacterium pair in the described con-

Interaction of Antibiotics with OMVs

The OMVs of P. sryingae are incubated with the antibiotic peptides colistin and melittin. After incubating for 24 h, the MALDI TOF mass spectra were recorded for observing the degradation of these antibiotics if any.



RESULTS

Physical Characteristics of OMVs from P. syringae Lz4W

The main purpose of this study is to identify different components of the OMVs from the Antarctic Gram-negative bacterium P. syringae Lz4W to understand their functional significance. OMVs of this bacterium were prepared, and their size was estimated using transmission electron microscopy 1348

dx.doi.org/10.1021/pr4009223 | J. Proteome Res. 2014, 13, 1345−1358

Journal of Proteome Research

Article

Figure 1. Transmission electron microscopy pictures of OMVs recorded after negative staining with uranyl acetate and size distribution of OMVs as observed by dynamic light scattering. Panel A shows a cell of P. syringae Lz4W releasing OMVs, and panel B shows the isolated OMVs. Panel C shows that in dynamic light scattering of OMVs exhibiting hydrodynamic radii in the range of 60−80 nm.

(TEM) and dynamic light scattering (DLS). Figure 1A,B shows the TEM images of the OMVs obtained from P. syringae Lz4W, wherein their diameter was estimated to be in the range of 60− 100 nm. The hydrodynamic radius of majority of these spherical vesicles shown in Figure 1C lies in the range of 60−80 nm, as determined by DLS. The hydrodynamic radius obtained by this method depends on the concentration of the particles, diffusion coefficient, viscosity of the solvent, surface structure, and type of ions in the medium. Hence, the size measured by this technique can be larger than the value obtained by electron microscopy. Earlier, detailed analysis was carried out on the effect of solvents, temperature, and sample preparation methods on the size of the particles, and comparative analysis of electron microscopy and dynamic light scattering was reported.45,46 Considering these facts, the size distribution of the vesicles is in general good agreement as determined by these two methods. Identifications of Proteins from the OMVs of P. syringae Lz4W

A comparative profile of the whole cell protein extract with inner membrane, outer membrane, and OMV proteins resolved on SDS-PAGE is shown in Figure 2A. From these results, it is clear that some proteins specifically partitioned into the OMVs and hence appeared as thick bands in the gel. These proteins may play an important role in the functions of the OMVs. Figure 2B shows the presence of LPS in these OMVs. Using the MS and MS/MS spectra of these tryptic digests, the proteins from the OMVs were identified (Figure 2A). The number of proteins identified by SEQUEST, MASCOT, and PEAKS were 566, 602, and 470 respectively. The numbers of proteins commonly identified by all these three algorithms are 429. Thus, using stringent criteria, the proteome of OMVs from the Antarctic bacterium was shown to contain at least 429 proteins, which were considered as confident identifications, and all further studies were focused on these proteins. Flagellin domain containing protein was identified with sequence coverage of 99% with 77 peptides. Similarly, some more proteins identified with sequence coverage above 80% (Supplementary Table S1, Supporting Information) are bacterioferittin with 14 peptides, porin with 38 peptides, OprD family outer membrane porin with 47 peptides, leucine ABC transporter with 31 peptides, cheperonin GroEL with 77 peptides (Supplementary Table S1). The top 40 most abundant proteins present in the OMVs of P. syringae Lz4W along with their predicted functions are shown Table 1.

Figure 2. The protein profiles and lipopolysaccharides on 12% SDSPAGE from the OMVs and protein profiles from subcellular components of P. syringae Lz4W. About 100 μg of protein was loaded in each well. Panel (A) shows a representative gel stained with Coomassie Blue comparing the protein profiles of cell lysate (CL), inner membrane fraction (IM), outer membrane fraction (OM), and OMVs. The selective sorting and enrichment of some proteins to the OMVs from different subcellular locations can be visualized. Panel (B) shows the silver stained profile of lipopolysaccharides from the OMVs of P. syringae.

Subcellular Localization of the Proteomes of OMVs and P. syringae

The number of outer membrane- and periplasmic- proteins were 42 and 35 respectively as predicted by Psortb V.3 program. There were 76 proteins whose localization could not be predicted by Psortb v3, which contained many lipoproteins and other proteins, supposedly a subset of proteins from other locations. Due to this, another predictive tool Cello was used as an additional support for subcellular localization of the identified OMV proteins. Cello predicts subcellular localization of all the identified proteins. In some cases, it predicts multiple localization sites. In these cases, they were grouped under unknown localization. Cello predicted 42 outer membrane and 94 periplasmic proteins. The outer membrane proteins 1349

dx.doi.org/10.1021/pr4009223 | J. Proteome Res. 2014, 13, 1345−1358

Journal of Proteome Research

Article

Table 1. The 40 Most Abundant Proteins Identified from OMVs of P. syringae Lz4W s. no.

accession (gi)

Uniprot entry

emPAI

length number of aminoacids

1

459959231

M5QMJ0

outer membrane lipoprotein OprI

18239

324

cell outer membrane; integral to membrane; porin activity

2

459960617

M5QH10

chaperonin GroEL

18065

547

ATP binding; cytoplasm; protein refolding

3

459961460

M5QJG9

flagellin domain-containing protein

9630

576

bacterial-type flagellum filament; ciliary or flagellar motility; flagellum organization; structural molecule activity

4

459961695

M5QKB6

2742

427

integral to membrane; porin activity

5

459961194

M5QW62

OmpA/MotB domain-containing protein 50S ribosomal protein L29

2308

63

ribosome; structural constituent of ribosome; translation

6

459961174

M5QGS9

bacterioferritin

1895

156

cellular iron ion homeostasis; ferric iron binding; iron ion transport; oxidoreductase activity

7

459961206

M5QZR6

30S ribosomal protein S7

1863

156

RNA binding; small ribosomal subunit; structural constituent of ribosome; translation

8 9

459959144 459961154

M5QUI1 M5QW14

1739 1602

93 158

10

459961193

M5QKB8

lipoprotein 6,7-dimethyl-8-ribityllumazine synthase 30S ribosomal protein S17

1478

88

(no information)2 riboflavin biosynthetic process; riboflavin synthase complex ribosome; structural constituent of ribosome; translation

11

459961666

M5R111

50S ribosomal protein L19

754.5

116

ribosome; structural constituent of ribosome; translation

12

459961958

M5QEB3

porin

709.1

119

DNA binding; intracellular; regulation of transcription, DNA-dependent

13

459961799

M5QHJ6

671.8

453

integral to membrane; porin activity

14

459960024

M5QH03

ornithine carbamoyltransferase urease subunit gamma

670.2

100

nickel cation binding; urea catabolic process; urease activity

15

459960720

M5QW37

bacterioferritin

488

154

cellular iron ion homeostasis; ferric iron binding; iron ion transport; oxidoreductase activity

16 17

459960685 459960671

M5QGP2 M5QR84

482.2 431

154 73

outer membrane (no information)

18

459959372

M5QIF7

17 kDa surface antigen leucine ABC transporter subunit substrate-binding protein LivK lipoprotein

420.6

201

19

459961186

M5QZP5

50S ribosomal protein L18

393

116

Gram-negative-bacterium-type cell outer membrane assembly; outer membrane ribosome; structural constituent of ribosome; translation

20

459961192

M5QIP6

50S ribosomal protein L14

366.7

122

large ribosomal subunit; structural constituent of ribosome; translation

21

459962107

M5R294

50S ribosomal protein L25/ general stress protein Ctc

309.4

196

5S rRNA binding; ribosome; structural constituent of ribosome; translation

22

459961191

M5QZP9

50S ribosomal protein L24

306.2

104

ribosome; structural constituent of ribosome; translation

23

459961200

M5QI52

50S ribosomal protein L23

278.3

99

nucleotide binding; ribosome; structural constituent of ribosome; translation

24

459961167

M5QXT9

OmpW family protein

277.8

336

25

459960507

M5QKD1

protein MvaT

269

433

amino acid binding; ornithine carbamoyltransferase activity periplasmic space; protein import

26

459962031

M5R223

DNA-binding protein HU, beta subunit

252.1

protein name

90

1350

gene ontology (GO)

DNA binding

gene ontology IDs GO:0009279; GO:0016021; GO:0015288 GO:0005524; GO:0005737; GO:0042026 GO:0009420; GO:0001539; GO:0043064; GO:0005198 GO:0016021; GO:0015288 GO:0005840; GO:0003735; GO:0006412 GO:0006879; GO:0008199; GO:0006826; GO:0016491 GO:0003723; GO:0015935; GO:0003735; GO:0006412 (no information) GO:0009231; GO:0009349 GO:0005840; GO:0003735; GO:0006412 GO:0005840; GO:0003735; GO:0006412 GO:0003677; GO:0005622; GO:0006355 GO:0016021; GO:0015288 GO:0016151; GO:0043419; GO:0009039 GO:0006879; GO:0008199; GO:0006826; GO:0016491 GO:0019867 (no information)

GO:0043165; GO:0019867 GO:0005840; GO:0003735; GO:0006412 GO:0015934; GO:0003735; GO:0006412 GO:0008097; GO:0005840; GO:0003735; GO:0006412 GO:0005840; GO:0003735; GO:0006412 GO:0000166; GO:0005840; GO:0003735; GO:0006412 GO:0016597; GO:0004585 GO:0042597; GO:0017038 GO:0003677

dx.doi.org/10.1021/pr4009223 | J. Proteome Res. 2014, 13, 1345−1358

Journal of Proteome Research

Article

Table 1. continued length number of aminoacids

s. no.

accession (gi)

Uniprot entry

protein name

27

459961176

M5QZN3

50S ribosomal protein L17

238

128

ribosome; structural constituent of ribosome; translation

28

459959212

M5QJI4

lipoprotein

201.1

259

cell outer membrane; integral to membrane

29

459961197

M5QIQ2

50S ribosomal protein L22

170.4

110

large ribosomal subunit; structural constituent of ribosome; translation

30

459961663

M5QLN6

30S ribosomal protein S16

155.3

83

ribosome; structural constituent of ribosome; translation

31 32

459962420 459961184

M5QQJ1 M5QW50

ketol-acid reductoisomerase 50S ribosomal protein L30

149.5 142.7

86 58

(no information) large ribosomal subunit; structural constituent of ribosome; translation

33

459959315

M5QY71

136.4

377

34 35

459960226 459961107

M5QGQ4 M5QIL9

124.4 114.2

125 232

amino acid transport; outer membrane-bounded periplasmic space (no information) cell outer membrane; integral to membrane

36 37

459959836 459960813

M5QF53 M5QNU9

hypothetical protein B195_19491 outer membrane porin lipoprotein, rare lipoprotein B family lipoprotein putative regulatory protein hypothetical protein B195_11651

100.8 100.2

73 338

(no information) branched-chain amino acid biosynthetic process; coenzyme binding; isomerase activity; ketol-acid reductoisomerase activity

38 39

459961985 459961970

M5QJA9 M5QCY2

92.07 91.52

175 83

outer membrane (no information)

40

459962752

M5QPT0

translocation protein TolB OprD family outer membrane porin glutamine synthetase

88.06

468

cytoplasm; glutamate-ammonia ligase activity; glutamine biosynthetic process; nitrogen fixation

emPAI

gene ontology (GO)

gene ontology IDs GO:0005840; GO:0003735; GO:0006412 GO:0009279; GO:0016021 GO:0015934; GO:0003735; GO:0006412 GO:0005840; GO:0003735; GO:0006412 (no information) GO:0015934; GO:0003735; GO:0006412 GO:0006865; GO:0030288 (no information) GO:0009279; GO:0016021 (no information) GO:0009082; GO:0050662; GO:0016853; GO:0004455 GO:0019867 (no information) GO:0005737; GO:0004356; GO:0006542; GO:0009399

vesicles with varying lengths of acyl chains, which are known to be the common constituents of inner and outer membrane of Gram-negative bacteria. The OMVs seem to contain saturated, monounsaturated, and methylated fatty acids possessing carbon chain lengths ranging from 12 to 20. The PEs and PGs were mainly identified by the formation of their characteristic fragment ions, and the sodiated product of the head groups found in the MS/MS spectra at m/z 163.95 and m/z 194.86, respectively. Earlier studies by MALDI MS also revealed the formation of these fragment ions from the bacterial lipids.35 The signals obtained in MALDI-TOF were further assigned to different phospholipid (PL) species depending upon the characteristic headgroup peak. The structures of these species were predicted by using Lipid Maps Structure Database.37 This database provides text/ ontology based search, in which one can find the fitting structures for a given molecular weight. For each mass entry, many possible structures are provided in general. The information about the type and subtype of the lipid species and the identified fatty acids helped to identify the probable structure for a given signal. Since matching structures could not be found for cardiolipins in Lipid Maps, their structures were manually assigned. The list of thus identified PLs is provided in Supplementary Table 3, Supporting Information. Earlier studies indicated that the presence of unsaturated and branched chain fatty acids in the acyl chains of lipids increases the fluidity of the membrane.51 It may be possible that the bacteria shed the vesicles when there was an increase in the fluidity above admissible limits. Further studies are required to establish a relationship for the formation of OMVs and increased unsaturated acyl chains in the bacterial lipids. After the

identified form the OMVs are shown in Table 2. To understand the biogenesis of OMVs, It is also important to know the amount of proteome partitioning in the OMVs from the parent bacterium. Hence, a comparison of the results of Psortb and Cello predicted for the proteome of P. syringae Lz4W and the OMVs is shown in Table 3. From these predictions, it is clear that OMVs contain at least 9.6% of the total proteome of the bacterium. Thus, by combining different methods, the subcellular localization of the identified proteins was predicted with high confidence. OMVs share structural and compositional similarity with the source bacterial cells. These results suggest that 25−30% of the total outer membrane proteins are segregating into OMVs from the bacterium. In addition to the large number of proteins originating from outer membrane and periplasmic regions of the cell, the OMVs also contain 225 proteins segregating from cytosolic regions. It may be noted that the proteins originating from the outer membrane and periplasmic proteins are high abundance, whereas the cytosolic proteins are low abundance according to the emPAI estimations (Supplementary Table S1, Supporting Information). This indicates that these cytosolic proteins may be some contamination arising from the cell lysis. Earlier reports on the OMV proteins from different bacteria also revealed the presence of some proteins from cytosol in the OMVs preparation.47−50 Lipid Analysis of the OMVs

To characterize the OMVs further, the lipids were extracted using a solvent mixture of chloroform and methanol. MALDI MS spectra shown in Supplemental Figures S1−S3, Supporting Information indicate the presence of phosphatidylethanolamine (PE), phosphatidylglycerol (PG), and cardiolipins (CL) in the 1351

dx.doi.org/10.1021/pr4009223 | J. Proteome Res. 2014, 13, 1345−1358

Journal of Proteome Research

Article

molecular composition of the OMVs were determined, studies were focused on the functions, particularly their significance in cold acclimatization of the bacterium.

Table 2. Outer Membrane Proteins Identified in the OMVs of P. syringae Lz4W s. no.

accession number

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

459959231 459961695 459961958 459960226 459961985 459961970 459961986 459961752 459961818 459960942 459960534 459962850 459962650 459959155 459962278 459961805

17 18

459961529 459960655

19

459962557

20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41

459960061 459962117 459958742 459958997 459958996 459960794 459960761 459960957 459959925 459960271 459960157 459959126 459961731 459960321 459961059 459959787 459962648 459962852 459961396 459961148 459960877 459962848

protein name

emPAI

outer membrane lipoprotein OprI OmpA/MotB domain-containing protein porin outer membrane porin translocation protein TolB OprD family outer membrane porin peptidoglycan-associated lipoprotein putative lipoprotein carbohydrate-selective porin OprB OprD family outer membrane porin hypothetical protein B195_10256 outer membrane porin TonB-dependent copper receptor lipoprotein OmpA family lipoprotein TonB-dependent hemoglobin/transferrin/ lactoferrin receptor hypothetical protein B195_08802 efflux transporter, outer membrane factor lipoprotein EmhC TolC family type I secretion outer membrane protein TonB-dependent siderophore receptor hypothetical protein B195_04796 OprD family outer membrane porin choloylglycine hydrolase hypothetical protein B195_21660 hypothetical protein B195_11556 porin B TonB-dependent siderophore receptor outer membrane porin organic solvent tolerance protein hypothetical protein B195_15207 VacJ family lipoprotein hypothetical protein B195_06020 TonB-dependent siderophore receptor outer membrane porin TonB-dependent siderophore receptor TonB-dependent siderophore receptor TonB-dependent siderophore receptor hypothetical protein B195_08137 B12 family TonB-dependent receptor outer membrane porin, OprD family protein TonB-dependent siderophore receptor

18239 2742 709.1 124.4 92.07 91.52 61.6 49.97 45.01 44.25 25.95 24.73 16.85 8.15 7.4 6.32

Functional Prediction of the Identified Proteins from the OMVs

The functional annotation of the most abundant proteins using the Universal Protein Knowledge Base (Uniprot KB) yielded information that could be helpful in confirming the various roles attributed to the OMVs. The detailed information about the functions of the identified proteins has been presented in Supplementary Table S2, Supporting Information and summarized in Figure 3. For example, bacterioferritin is

5.79 3.51 3.51 3.09 2.05 1.98 1.54 1.39 1.2 1.04 1.01 0.95 0.88 0.88 0.82 0.68 0.56 0.56 0.53 0.45 0.44 0.27 0.17 0.16 0.1

Figure 3. Distribution of molecular functions of 349 out of 429 proteins as retrieved from UniprotKB. Information about rest of the proteins is not available.

essential for enriching the rare ions of iron, while the 60 kDa chaperonin is involved in protein folding, etc. The identified proteins are multifunctional, involving proteins for nutrient acquisition, transport, stress responses, ATP metabolism, metabolic enzymes, and receptors to various ligands. These proteins may play an important role in cellular iron homeostasis, integral membrane proteins, and several other functions. Translocation protein Tol B, Tol/Pal system protein YbgF that plays an important role in maintaining membrane integrity was identified in the OMVs. Several ribosomal proteins were identified, which were also shown to be associated with the OMVs of E. coli.6 Universal stress protein, tellurium resistance proteins ter D, ter E which plays a role in the stress response were released to OMVs, suggesting a

Table 3. Prediction of Subcellular Localization of the Proteins of P. syringae Lz4W and Its OMVs Cello

Psortb V3

SCL

OMVs proteins

P. syringae Lz4W proteins

OMVs proteins

P. syringae Lz4W proteins

cytosolic inner membrane periplasmic outer membrane extracellular unknowna total

225 (52.44) 2 (0.47) 94 (21.91) 42 (9.79) 6 (1.40) 60 (13.99) 429

2383 (53.55) 760 (17.07) 442 (9.93) 172 (3.87) 50 (1.12) 643 (14.45) 4450

231 (53.85) 38 (8.86) 35 (8.16) 42 (9.79) 7 (1.63) 76 (17.71) 429

2088 (46.92) 1114 (25.03) 163 (3.66) 119 (2.67) 44 (0.99) 1022 (22.96) 4550

a

Whenever more localization sites were predicted by Cello, the prediction was considered as Unknown (see Table S1 for details). The numbers in the brackets were percentage of the total proteins. 1352

dx.doi.org/10.1021/pr4009223 | J. Proteome Res. 2014, 13, 1345−1358

Journal of Proteome Research

Article

Figure 4. The effect of OMVs on growth of P. syringae Lz4W at 22 °C in the presence of growth-inhibiting concentrations of colistin 2 μg/mL (A), melittin 3 μg/mL (C), and streptomycin 0.7 μg/mL (E). Panels A, C, and E; a, Antarctic bacterial medium (ABM) inoculated 1% from an overnight grown culture; b, ABM with growth-inhibiting concentrations of antibiotics; c−h, medium inoculated with bacteria and growth-inhibiting concentration of antibiotics followed by the addition of 12, 10, 8, 6, 4, 2 μg/mL of OMVs, respectively. The panels B, D, and F show the respective qualitative assays on agar plates plate. The nomenclature of a−h for growth curves and the plate assays is the same.

presence of OMVs prepared from P. syringae Lz4W. These results show that OMVs offer protection to this Antarctic bacterium in a concentration-dependent manner (Figure 4A,B, growth curves, c−h). Similarly, the growth profiles are shown in the presence of melittin in Figure 4C,D. However, when the bacterium was grown in the presence of streptomycin, the OMVs (even at higher concentrations) could not protect the bacterium from the growth inhibitory effects of the antibiotic (Figure 4E,F). From these results, it appears that the OMVs can offer protection to the bacteria from membrane-active antibiotics such as colistin and melittin, whereas they fail to do so in the case of streptomycin, which acts by entering the bacterial cell and inhibiting the protein synthesis. Earlier reports also suggest that OMVs protect bacteria from antibiotics such as colistin and polymixin B that act on the outer membrane of the bacteria resulting in cell lyses.18

significant role of vesicles in cold adaptation. The release of OMVs was shown to be a response in relieving different stresses of the bacteria.14 Out of 429 proteins, functions could be predicted for only 349 proteins. The functions of 80 proteins could not be predicted. It may be possible that these proteins may be involved in cold adaptation. The genome of this bacterium was recently sequenced, and further studies are in progress to identify the genes involved in cold adaptation.25 OMVs Protect Producer Bacterium against Antibiotics

It was observed that OMVs protect the producer bacterium against growth inhibitory effects of antibiotics.18 In the present study, the extent of protection provided by OMV was studied in the Antarctic bacterium, which usually does not get exposure to antibiotics but shows sensitivity and resistance to some antibiotics during taxonomic studies. The release of OMVs from the bacterium is recognized as a stress response,14 and therefore the protection from the chemical stress from the growth inhibitory effects of the antibiotics was studied. Initially, growth-inhibitory concentration for the antibacterial activity for each antibiotic was determined (Supplementary Figure S4, Supporting Information). In order to study the effect of OMVs on the growth of bacteria, each antibiotic was added above its growth-inhibitory concentration. Figure 4A shows the growth curves of the bacterium upon addition of the antibiotic colistin above its growth-inhibitory inhibitory concentration in the

Percentage Survivors of P. syringae Lz4W

The number of survivors was estimated by counting the CFUs on the agar plates. It was observed that about 19% of bacteria could be protected by OMVs in the presence of colistin, whereas about 23% survived in the presence of mellitin (Supplementary Figure S5, Supporting Information). Even after increasing the amount of OMVs, there is no further increase in percentage of survivals. 1353

dx.doi.org/10.1021/pr4009223 | J. Proteome Res. 2014, 13, 1345−1358

Journal of Proteome Research

Article

respectively. It is observed that the fluorescent intensity increases with increasing concentrations of colistin and melittin (Figure 5A,B). This suggests an increment in partition of NPN into the hydrophobic environment of the OMVs with increasing concentrations of colistin and melilitin. This is possible only if the antibiotic molecules bind to the OMV surface and perturb its structure. Hence, it is suggested that the antibiotic is scavenged in the medium by the OMVs, which is why the bacterium was able to grow even in the presence of antibiotic colistin above its growth-inhibiting concentration. A similar study was carried out in the presence of antibiotic streptomycin that binds to ribosome and inhibits the synthesis of proteins, and the results are shown in the Figure 5C, where there was no change in the fluorescence spectrum of NPN. From these results, it is clear that streptomycin is not able to bind to the OMVs, and hence OMVs could not protect the bacterium against lyses. Earlier studies demonstrated that the OMVs are found to transport antibiotic induced enzyme β-lactamase55−57 and inactivate the antibiotic. The OMVs of the P. syringae Lz4W contain a peptidase (Supplemental Table S1, Supporting Information). In the present situation also it is possible that the peptidases or enzymes present in the OMVs may degrade the peptide antibiotics and protect the bacterium by degrading them to short peptides, and they in turn can bind to OMVs and facilitate NPN uptake. To confirm the exact reason, the OMVs of P. syringae Lz4W were incubated with the antimicrobial peptides and analyzed by MALDI TOF. It was observed that the peptides could not be degraded even after incubation for 24 h (Figure 6), indicating that these antimicrobial molecules are sequestered to the vesicles and not degraded by the enzymes present in them. It may be possible that the peptidases may be buried inside the vesicles.

Mechanism of Protection by OMVs against Antibiotics

Our studies were further focused on determining the mechanism of protection offered by OMVs. NPN fluorescence studies were used to identify the mechanism of protection to the bacterium. The fluorescent probe NPN is generally used to detect the presence of hydrophobic environments, and hence its uptake studies have been extensively carried out to understand the action of antibiotics such as polymixin B on the cell surfaces.52 The fluorescent intensity increases with a blue shift in the emission maximum when the probe partitions into the hydrophobic environment.53,54 The results of the NPN uptake by the OMVs in the presence of the antibiotics colistin, melittin, and streptomycin are shown in Figure 5A−C,



DISCUSSION Membrane vesicle production is a well-known process in Gramnegative bacteria. The composition of the secreted vesicles in different organisms may reveal conserved secretary mechanism employed by different bacteria. Recently, our laboratory carried out some studies on the lipid composition of OMVs from P. syringae pv tomato T1.36 The present study shows the presence of unsaturated and branched chain fatty acids in the lipids of the OMVs from P. syringae Lz4W. It is known that an increase in the unsaturated and branched chain fatty acids in the lipids increases the fluidity of the bacterial membrane.50 Hence, it is possible that the above-mentioned increase in the fluidity of the bacterial membrane may also be a facilitator for the formation of OMVs from the bacterium. Further studies are required to establish this prediction. Proteins with different functions were identified from the OMVs of P. syringae Lz4W. A comparison of the proteins from the OMVs of the mesophilic P. aeruginosa PA019 revealed that it contains 338 proteins, whereas the Antarctic P. syringae contains 429 proteins. A total of 85 proteins from the P. syringae Lz4W are identical with the proteins of P. aeruginosa and 16 matched P. syringae pv tomato T1.36 This suggests the protein sorting mechanism of OMVs from the parent bacterium. The protein profiles of the OMVs of these three bacteria contained outer membrane proteins, ABC transporters, lipoproteins, transcription factors, proteins from ATP synthase complex, and many cytosolic enzymes. The OMVs of P. aeruginosa and P. syringae Lz4W (present work) also showed a number of ribosomal proteins, chaperones, and TonB-depend-

Figure 5. The NPN (5 μM) uptake profiles of OMVs (6 μg/mL) in the presence of varying concentrations of colistin (A), melittin (B), and streptomycin (C). Increasing concentration of the antibiotics was added and emission spectra were recorded respectively. A plot of fluorescence emission intensity of NPN at 410 nm as function of concentration of antibiotic is shown from three experiments. A positive correlation of fluorescence intensity with concentrations of colistin and melittin was observed with Pearson’s correlation analysis (p < 0.01). 1354

dx.doi.org/10.1021/pr4009223 | J. Proteome Res. 2014, 13, 1345−1358

Journal of Proteome Research

Article

Figure 6. MALDI TOF mass spectrum of the OMVs from P. syringae Lz4W (A), OMVs incubated with colistin (B) and melittin (C) at 22 °C for 24 h, recorded in reflector mode using HCCA as matrix. The sodium and potassium adducts of colistin were also observed in panel B.

both mesophilic and psychrotrophic bacteria. Some genes may be responsible for cold adaptation which has not been characterized so far. The functions of the genes involved in cold adaptation are not available in the algorithms used for the prediction of function of the proteins. Hence, the functions of 80 proteins could not be predicted. These proteins may be involved in cold-adapted functions. It is also possible that they may be involved in proper packing of the OMVs. The vesicles also protected the bacterium from membrane active antibiotics by sequestering them, while they could not

ent receptors which were not reported in the case of OMVs of P. syringae pv tomato T1. Chaperonin GroEL was found to be a high abundant protein. Chaperon GroEL was found to be present in the OMVs of other bacteria F .novocida also.7 Other proteins such as heat shock protein 90, molecular chaperone Dnak, and heat shock protein GrpE were also identified. These proteins help in refolding of proteins that are misfolded and in solubilization of proteins that are aggregated. Several proteins were predicted to involve in catalytic functions (Figure 3). The biochemical and physiological processes remain the same in 1355

dx.doi.org/10.1021/pr4009223 | J. Proteome Res. 2014, 13, 1345−1358

Journal of Proteome Research

Article

the mechanism of action of OMVs appears to be situationdependent. This study shows that the antibiotic protection by the membrane active antibiotics is a naturally evolved process in bacteria; antibiotic resistance and sensitivity have nothing to do with the environment in which the bacteria have adapted. It is a natural phenomenon of bacterial physiology. OMVs from the cold adapted bacterium were also capable of protecting the bacterium from the chemical stress induced by the growth inhibitory effects of membrane active antibiotics.

protect from other antibiotics. The percentage survivors in the presence of colistin are about 19%. Earlier studies using the OMVs from a mesophilic bacterial strain revealed that these vesicles could protect about 50% of the bacteria.18 During the taxonomic studies of P. syringae Lz4W, a detailed analysis on growth profiles at different temperatures was studied, and it was noticed that this Antarctic psychrotroph exhibited an optimum growth at 22 °C. Growth temperature plays an important role in the biosynthesis of lipids particularly, on the composition of fatty acyl chains in their phospholipids.58,59 It was also shown that vesiculation also increases with increased growth temperature due to temperature-induced stress.14 This psychrotrophic bacterium also exhibited some unique features different from their corresponding mesophilic strains particularly: sensitivity to some antibiotics such as streptomycin, kanamycin, gentamycin, and tetracycline . The mesophilic Pseudomonas sp. are not sensitive to these antibiotics, and this was attributed as an adaptation response.21 The difference in the mesophilic and psychrotrophic strains could be the composition and packing of the lipids and proteins they contain. Earlier studies have shown that the cold adapted bacteria use different strategies for adaptation such as protein content of the membrane, modulation of the fatty acid composition to regulate the fluidity of the membranes, and other methods.60 These studies suggest that packing parameters play an important role in sequestering the antibiotics. Earlier studies have shown that native OMVs were found to be highly enriched in LPS with a longer, highly charged form of Oantigen (B-band LPS) compared to the P. aeruginosa outer membrane which contains both short, uncharged (A-band) and B-band of LPS.61 In the present study, the B-band of LPS and the A-band were observed in the OMVs of P. syringae Lz4W (Figure 2B). The NPN uptake studies on the Antarctic bacterium P. syringae Lz4W revealed that polymyxin B, a membrane active antibiotic, increases the permeability of the membranes of the bacterium.54 The packing of lipids and membrane proteins in the OMVs appears to be similar to that of the originating bacterial strain as the NPN uptake behavior is similar in the presence of membrane active antibiotics. A recent study also demonstrated that the membrane vesicles of a hypervesiculating strain of E. coli protected the bacterium against antibiotics such as polymyxin B and colistin that act on the outer membrane. The estimated the percentage protection was about 50% in the presence of OMVs at a concentration of 4 μg/mL of OMVs.18 In the present study, more OMVs were required for protecting the bacterium, and still the percentage of protection of the bacterial cells is much less (Figure S5, Supporting Information). Thus, subtle differences in the packing and composition of the components in the mesophilic and psychrotrophic bacterial OMVs, particularly due to the differences in the fatty acyl chains in the phospholipids, could be the reason for the difference in the amount of antibiotic sequestered and the extent of protection offered by the OMVs. On the basis of these studies, it appears that different mechanisms operate in different situations for the biological activity of the vesicles. In one type of action, the OMVs appear to release the contents and the molecules thus released appear to exert their activity on the host cells. In another mechanism, the OMVs carry the enzymes such as β-lactamase and inactivate the functioning of the antibiotics. Yet, in another method, the OMVs protect the bacterium against the antibiotics by binding to them, as being shown in the present study. Hence, it appears that the OMVs act in different situations differently. Therefore,



CONCLUSIONS The analysis of the lipids and proteins revealed that their structures play a crucial role in the packing of these components and their function. The vesicles protect the bacterium from the growth inhibitory effects of the membrane active antibiotics by sequestering them. The OMVs of the cold adapted bacterium could protect the producer bacterium from the membrane active antibiotics to a lesser extent as compared to the mesophilic strains perhaps due to the composition of the lipids and organization of lipids and proteins they contain. Since this strain P. syringae Lz4W has never been exposed to antibiotics naturally, it shows that this phenomenon of defense is innate and not adaptive.



ASSOCIATED CONTENT

* Supporting Information S

Figure S1: The MALDI TOF mass spectrum of one TLC spot of the lipid extracts of the OMVs prepared from the Antarctic bacterium Pseudomonas syringae Lz4W. Figure S2: MALDI TOF mass spectrum of a TLC spot obtained by separating the lipid extracts of the OMVs prepared from the P. syringae Lz4W. Figure S3: The MALDI TOF mass spectrum of the cardiolipins from the OMVs of the Antarctic bacterium P. syringae Lz4W. Figure S4: Determination of growth inhibitory concentrations of the antibiotics. Figure S5: The percentage protection of bacteria in the presence of antibiotics calculated using colony counting method with increasing concentrations of OMVs. Table S1: The list of proteins identified from the OMVs of P. syringae Lz4W (Identified using SEQUEST, MASCOT and PEAKS). Table S2: Gene ontology’s of the proteins identified from the OMVs of P. syringae Lz4W. Table S3: Phospholipids identified from the OMVs of P. syringae Lz4W. This material is available free of charge via the Internet at http://pubs.acs.org.



AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Phone: 040-27192572. Fax: +9140-27160591. Notes

The authors declare no competing financial interest.



ACKNOWLEDGMENTS We thank the Department of Science and Technology (DST), New Delhi, for financial support to carry out this research work. H.M.K. is a recipient of CSIR Senior research fellowship. We acknowledge the help from the proteomics facility of CCMB, Hyderabad, for the help in mass spectrometry. We thank the PRIDE team in their help to deposit our data. We thank Dr. Shashi Singh for help in EM studies, and Dr. C Sivakama 1356

dx.doi.org/10.1021/pr4009223 | J. Proteome Res. 2014, 13, 1345−1358

Journal of Proteome Research

Article

(18) Manning, A. J.; Kuehn, M. J. Contribution of bacterial outer membrane vesicles to innate bacterial defense. BMC Microbiol. 2011, 11, 258. (19) Chattopadhyay, M. K.; Grossart, H.-P. Antibiotic resistance: Intractable and here’s why. BMJ 2010, 341, c6848. (20) Sengupta, S.; Chattopadhyay, M. K.; Grossart, H.-P. The multifaceted roles of antibiotics and antibiotic resistance in nature. Front. Microbiol. 2013, 10.3389/fmicb.2013.00047. (21) Shivaji, S.; Rao, N. S.; Saisree, L.; Sheth, V.; Reddy, G. S.; Bhargava, P. M. Isolation and identification of Pseudomonas spp. from Schirmacher Oasis, Antarctica. Appl. Environ. Microbiol. 1989, 55, 767−770. (22) Jagannadham, M. V. Identification of proteins from membrane preparations by a combination of MALDI TOF-TOF and LC-coupled linear ion trap MS analysis of an Antarctic bacterium Pseudomonas syringae Lz4W, a strain with unsequenced genome. Electrophoresis 2008, 29, 4341−4350. (23) Jagannadham, M. V.; Abou-Eladab, E. F.; Kulkarni, H. M. Identification of Outer Membrane Proteins from an Antarctic Bacterium Pseudomonas syringae Lz4W. Mol. Cell. Proteomics 2011, 10, M110.004549. (24) Jagannadham, M. V.; Saranya, S. Analysis of the Membrane Proteins of an Antarctic Bacterium Pseudomonas syringae. Proteomics Insights 2011, 4, 13−19. (25) Pandiyan, A.; Ray, M. K. Draft Genome Sequence of the Antarctic Psychrophilic Bacterium Pseudomonas syringae Strain Lz4W. Genome Announc. 2013, 1 (3), No. e00377-13. (26) Sahu, B.; Ray, M. K. Auxotrophy in natural isolate: minimal requirements for growth of the Antarctic psychrotrophic bacterium Pseudomonas syringae Lz4W. J. Basic Microbiol. 2008, 48, 38−47. (27) Wai, S. N.; Takade, A.; Amako, K. The release of outer membrane vesicles from the strains of enterotoxigenic Escherichia coli. Microbiol. Immunol. 1995, 39, 451−456. (28) Chutkan, H.; MacDonald, I.; Manning, A.; Kuehn, M. J. Quantitative and Qualitative Preparations of Bacterial Outer Membrane Vesicles. In Bacterial Cell Surfaces; Humana Press: New York, 2013; pp 259−272. (29) Kumada, H.; Haishima, Y.; Umemoto, T.; Tanamoto, K. Structural study on the free lipid A isolated from lipopolysaccharide of Porphyromonas gingivalis. J. Bacteriol. 1995, 177 (8), 2098−2106. (30) Tsai, C. M.; Frasch, C. E. A sensitive silver stain for detecting lipopolysaccharides in polyacrylamide gels. Anal. Biochem. 1982, 119, 115−119. (31) Fomsgaard, A.; Freudenberg, M. A.; Galanos, C. Modification of the silver staining technique to detect lipopolysaccharide in polyacrylamide gels. J. Clin. Microbiol. 1990, 28, 2627−2631. (32) Karkhanis, Y. D.; Zeltner, J. Y.; Jackson, J. J.; Carlo, D. J. A new and improved microassay to determine 2-keto-3-deoxyoctonate in lipopolysaccharide of Gram-negative bacteria. Anal. Biochem. 1978, 85, 595−601. (33) Osborn, M. J. Studies on the Gram-negative cell wall I. evidence for the role of 2-keto-3-deoxyoctonate in the lipopolysaccharide of Salmonella typhimurium. Proc. Natl. Acad. Sci. U. S. A. 1963, 50, 499− 506. (34) Bligh, E. G.; Dyer, W. J. A rapid method of total lipid extraction and purification. Can. J. Biochem. Physiol. 1959, 37, 911−917. (35) Gidden, J.; Denson, J.; Liyanage, R.; Ivey, D. M.; Lay, J. O. Lipid Compositions in Escherichia coli and Bacillus subtilis During Growth as Determined by MALDI-TOF and TOF/TOF Mass Spectrometry. Int. J. Mass Spectrom. 2009, 283, 178−184. (36) Chowdhury, C.; Jagannadham, M. V. Virulence factors are released in association with outer membrane vesicles of Pseudomonas syringae pv. tomato T1 during normal growth. Biochim. Biophys. Acta 2013, 1834, 231−239. (37) Sud, M.; Fahy, E.; Cotter, D.; Brown, A.; Dennis, E. A.; Glass, C. K.; Merrill, A. H., Jr.; Murphy, R. C.; Raetz, C. R.; Russell, D. W.; Subramaniam, S. LMSD: LIPID MAPS structure database. Nucleic Acids Res. 2007, 35, D527−532.

Sundari and Dr. M. K. Chattopadhyay for comments and criticism on the manuscript.



REFERENCES

(1) Sabra, W.; Lundsdrof, H.; Zeng, A. P. Alterations in the formation of lipopolysacchiride and membrane vesicles on the surface of Pseudomonas aeruginosa PAO1 under oxygen stress conditions. Microbiology 2003, 149 (Pt10), 2789−2795. (2) Mashburn-warren, L.; Howe, J.; Garidel, P.; Richter, W.; Steiniger, F.; Roessle, M.; Brnadenburg, K.; Whitely, M. Interaction of quorum signals with outer membrane lipids: insights in to prokaryotic membrane vesicle formation. Mol. Microbiol. 2008, 69, 491−502. (3) Dorward, D. W.; Garon, C. F. DNA is Packaged within Membrane-Derived Vesicles of Gram-Negative but Not Gram-Positive Bacteria. Appl. Environ. Microbiol. 1990, 56, 1960−1962. (4) Renelli, M.; Matias, V.; Lo, R. Y.; Beveridge, T. J. DNAcontaining membrane vesicles of Pseudomonas aeruginosa PAO1 and their genetic transformation potential. Microbiology 2004, 150, 2161− 2169. (5) Kulp, A.; Kuehn, M. J. Biological functions and biogenesis of secreted bacterial outer membrane vesicles. Annu. Rev. Microbiol. 2010, 64, 163−184. (6) Lee, E. Y.; Bang, J. Y.; Park, G. W.; Choi, D. S.; Kang, J. S.; Kim, H. J.; Park, K. S.; Lee, J. O.; Kim, Y. K.; Kwon, K. H.; Kim, K. P.; Gho, Y. S. Global proteomic profiling of native outer membrane vesicles derived from Escherichia coli. Proteomics 2007, 7, 3143−3153. (7) Pierson, T.; Matrakas, D.; Taylor, Y. U.; Manyam, G.; Morozov, V. N.; Zhou, W.; van Hoek, M. L. Proteomic characterization and functional analysis of outer membrane vesicles of Francisella novicida suggests possible role in virulence and use as a vaccine. J. Proteome Res. 2011, 10, 954−967. (8) Parker, H.; Keenan, J. I. Composition and function of Helicobacter pylori outer membrane vesicles. Microb. Infect. 2012, 14, 9−16. (9) Choi, D. S.; Kim, D. K.; Choi, S. J.; Lee, J.; Choi, J. P.; Rho, S.; Park, S. H.; Kim, Y. K.; Hwang, D.; Gho, Y. S. Proteomic analysis of outer membrane vesicles derived from Pseudomonas aeruginosa. Proteomics 2011, 11 (16), 3424−3429. (10) Nevot, M.; Deroncele, V.; Messner, P.; Guinea, J.; Mercade, E. Characterization of outer membrane vesicles released by the psychrotolerant bacterium Pseudoalteromonas antarctica NF3. Environ. Microbiol. 2006, 8, 1523−1533. (11) Perez-Cruz, C.; Carrion, O.; Delgado, L.; Martinez, G.; LopezIglesias, C.; Elena Mercade, E. New Type of Outer Membrane Vesicle Produced by the Gram-Negative Bacterium Shewanella vesiculosa M7T: Implications for DNA content. Appl. Environ. Microbiol. 2013, 79, 1874−1881. (12) Kamekura, M.; Seno, Y.; Tomioka, H. Detection and expression of a gene encoding a new bacteriorhodopsin from an extreme halophile strain HT (JCM 9743) which does not possess bacteriorhodopsin activity. Extremophiles 1998, 2, 33−39. (13) Frias, A.; Manresa, A.; de Oliveira, E.; Lopez-Iglesias, C.; Mercade, E. Membrane Vesicles: A Common Feature in the Extracellular Matter of Cold-Adapted Antarctic Bacteria. Microb. Ecol. 2010, 59, 476−486. (14) McBroom, A. J.; Kuehn, M. J. Release of outer membrane vesicles by Gram-negative bacteria is a novel envelope stress response. Mol. Microbiol. 2007, 63, 545−558. (15) Deatherage, B. L.; Cookson, B. T. Membrane vesicle release in bacteria, eukaryotes, and archaea: a conserved yet underappreciated aspect of microbial life. Infect. Immun. 2012, 80, 1948−1957. (16) Schooling, S. R.; Beveridge, T. J. Membrane vesicles: an overlooked component of the matrices of biofilms. J. Bacteriol. 2006, 188, 5945−5957. (17) McMahon, M. A.; Xu, J.; Moore, J. E.; Blair, I. S.; McDowell, D. A. Environmental stress and antibiotic resistance in food-related pathogens. Appl. Environ. Microbiol. 2007, 73, 211−217. 1357

dx.doi.org/10.1021/pr4009223 | J. Proteome Res. 2014, 13, 1345−1358

Journal of Proteome Research

Article

(38) Laemmli, U. K. Cleavage of Structural Proteins during the Assembly of the Head of Bacteriophage T4. Nature 1970, 227, 680− 685. (39) Vizcaino, J. A.; Cote, R. G.; Csordas, A.; Dianes, J. A.; et al. The PRoteomics IDEntifications (PRIDE) database and associated tools: status in 2013. Nucleic Acids Res. 2013, 41 (D1), D1063−D10639. (40) Ishihama, Y.; Oda, Y.; Tsuyoshi Tabata, T.; Sato, T.; Nagasu, T.; Rappsilber, J.; Mann, M. Exponentially Modified Protein Abundance Index (emPAI) for Estimation of Absolute Protein Amount in Proteomics by the Number of Sequenced Peptides per Protein. Mol. Cell. Proteomics 2005, 4, 1265−1272. (41) Yu, C. S.; Chen, Y. C.; Lu, C. H.; Hwang, J. K. Prediction of Protein Subcellular Localization. Proteins: Struct., Funct., Bioinformatics 2006, 64, 643−651. (42) Magrane, M.; Consortium, U. UniProt Knowledgebase: a hub of integrated protein data. Database (Oxford), 2011, 2011, DOI: 10.1093/database/bar009. (43) Shatalin, K.; Shatalina, E.; Mironov, A.; Nudler, E. H2S: A Universal Defense Against Antibiotics in Bacteria. Science 2011, 234, 986−990. (44) Andrews J. M.; Determination of minimum inhibitory concentrations J. Antimicrob. Chemother. 2001, 48 (Suppl 1), 5−16. (Erratum in J. Antimicrob. Chemother. 2002 49(6), 1049). (45) Hallet, F. R.; Watton, J.; Krygsman, P. Vesicle sizing: Number distributions by dynamic light scattering. Biophys. J. 1991, 59, 357− 362. (46) Booz, A.; Vogel, V.; Schubert, D.; Kreuter, J. Comparison of scanning electron microscopy, dynamic light scattering and analytical ultracentrifugation for the sizing of poly(butyl cyanoacrylate) nanoparticles. Eur. J. Pharm. Biopharm. 2004, 57, 369−375. (47) Beveridge, T. J. Structures of gram-negative cell walls and their derived membrane vesicles. J. Bacteriol. 1999, 181, 4725−4733. (48) Horstman, A. L.; Kuehn, M. J. Enterotoxigenic Escherichia coli secretes active heat-labile enterotoxin via outer membrane vesicles. J. Biol. Chem. 2000, 275, 12489−12496. (49) Kwon, S. O.; Gho, Y. S.; Lee, J. C.; Kim, S. I. Proteome analysis of outer membrane vesicles from a clinical Acinetobacter baumannii isolate. FEMS Microbiol. Lett. 2009, 297, 150−156. (50) Lee, E. Y.; Choi, D. S.; Kim, K. P.; Gho, Y. S. Proteomics in Gram-negative bacteria outer membrane vesicles. Mass Spectrom. Rev. 2008, 27, 535−555. (51) Nichols, D. S.; Russel, N. J. Fatty acid adaptation in an Antarctic bacterium changes in primer utilization. Microbiology 1996, 55, 3065− 3071. (52) Hancock, R. E.; Wong, P. G. Compounds which increase the permeability of the Pseudomonas aeruginosa outer membrane. Antimicrob. Agents Chemother. 1984, 26, 48−52. (53) Loh, B.; Grant, C.; Hancock, R. E. Use of the Fluorescent Probe 1-N-Phenylnaphthylamine to Study the Interactions of Aminoglycoside Antibiotics with the Outer Membrane of Pseudomonas aeruginosa. Antimicrob. Agents Chemother. 1984, 26, 546−559. (54) Kumar, G. S.; Jagannadham, M. V.; Ray, M. K. Lowtemperature-induced changes in composition and fluidity of lipopolysacchirides in the Antarctic Psychrotrophic Bacterium Pseudomonas syringae. J. Bacteriol. 2002, 184 (23), 6746−6749. (55) Schaar, V.; Paulsson, M.; Morgelin, M.; Riesbeck, K. Outer membrane vesicles shield Moraxella catarrhalis β-lactamase from neutralization by serum IgG. J. Antimicrobial Agents. Chemother. 2013, 68 (3), 593−600. (56) Schaar, V.; Nordstrom, T.; Morgelin, M.; Riesbeck, K. Moraxella catarrhalis outer membrane vesicles carry beta-lactamase and promote survival of Streptococcus pneumoniae and Haemophilus inf luenzae by inactivating amoxicillin. Antimicrob. Agents Chemother. 2011, 55, 3845−3853. (57) Ciofu, O.; Beveridge, T. J.; Kadurugamuwa, J.; WaltherRasmussen, J.; Hoiby, N. Chromosomal beta-lactamase is packaged into membrane vesicles and secreted from Pseudomonas aeruginosa. J. Antimicrob. Chemother. 2000, 45, 9−13.

(58) Nichols, D. S.; Olley, J.; Garda, H.; Brenner, R. R.; McMeekin, T. A. Effect of Temperature and Salinity Stress on Growth and Lipid Composition of Shewanella gelidimarina. Appl. Environ. Microbiol. 2000, 66, 2422−2429. (59) Jagannadham, M. V.; Chattopadhyay, M. K.; Subbalakahmi, C. S.; Narayanan, K.; Vairamani, M.; Mohan Rao, Ch .; Shivaji, S. Carotenoids of an Antarctic psychrotolarent bacterium Spingobacterium antarcticus and a mesophilic bacterium S. multivorum. Arch. Microbiol. 2000, 173, 418−424. (60) Chintalapati, S.; Kiran, M. D.; Shivaji, S. Role of membrane lipid fatty acids in cold adaptation. Cell. Mol. Biol. 2004, 50 (5), 631−642. (61) Kadurgamuwa, J. L.; Beveridge, T. J. Virulence factors are released from Pseudomonas aeruginosa in association with membrane vesicles during normal growth and exposure to gentamicin: a noval mechanism of enzyme secretion. J. Bacteriol. 1995, 63, 545−558.

1358

dx.doi.org/10.1021/pr4009223 | J. Proteome Res. 2014, 13, 1345−1358