1

Production of outer membrane vesicles by the human pathogen Burkholderia

cenocepacia: purification and preliminary characterization

Rui Filipe Brás Teles Martins

Abstract

Cystic Fibrosis (CF) is an autossomal recessive disease resulting from mutations in the cystic

fibrosis transmembrane conductance regulator (CFTR) gene, leaving affected patients more

susceptible to infection by bacteria. One group of bacteria, the Burkholderia cepacia complex (Bcc), is

responsible for dangerous infections in these patients, due to their resistance to multiple antibiotics

being therefore necessary to develop alternative treatments. Outer membrane vesicles (OMV) have

been used for the development of vaccines against pathogens such as Neisseria meningitidis and

Burkholderia pseudomallei, with an OMV vaccine against N. meningitidis already available in the

market. This study aimed at the development of a scalable production method of OMVs, in order to

allow their use in vaccine development studies. OMVs from B. cenocepacia K56-2 were isolated and

purified by size exclusion chromatography (SEC), however the final purity was not high, with OMVs

still contaminated with pili and flagella. This prompted the development of an alternative strategy to

obtain purified OMVs in which the strain used for OMV production was changed to B. cenocepacia

AU1054 non piliated strain. With this strain it was possible to obtain contaminant-free OMVs after the

isolation process, which was confirmed by transmission electron microscopy (TEM). OMVs obtained

from B. cenocepacia AU1054 were used to perform cytotoxic assays in 16HBE14o- human bronchial

epithelial cells and in Galleria melonella. 16HBE14o- cells suffered reduced survival upon exposure to

AU1054-derived OMVs, however this result was not reproduced in G. melonella, with all injected

caterpillars surviving the assay.

Keywords: Outer membrane vesicles, Cystic fibrosis, purification, Burkholderia cepacia complex, Size

exclusion chromatography, cytotoxicity

Introduction

Cystic Fibrosis (CF) is an autossomal recessive

disease resulting from mutations in the cystic

fibrosis transmembrane conductance regulator

(CFTR) [53], among other problems CF patients

suffer from thicker mucus in the lungs [3; 62],

which leaves these patients more susceptible to

bacterial infections [62; 66]. Among the bacterial

infections these patients may suffer it is included

infection by species belonging to the

Burkholderia cepacia complex (Bcc), which

despite the small percentage of infections in CF

patients being caused by these bacteria [36],

leads to the development of problematic

infections [21].

Bcc is a group of 20 closely related species of

bacteria, to which Burkholderia cenocepacia,

among other species, belongs [15].

Infection with Bcc bacteria usually causes

reduced lung function, and in a subset of patients

it may cause necrotizing pneumonia and a rapidly

fatal septicaemia termed “Cepacia Syndrome”

[21]. The severity of Bcc infection is explained by

several characteristics of this group of bacteria,

such as the ability of resisting exposure to a wide

range of antibiotics [16; 37], ability to form

biofilms [12], among others [13; 46; 64], which

2

facilitate the infection with Bcc strains and difficult

eradication of the bacteria. Furthermore, these

species are easily spread by contact between

infected CF patients [25; 36], saline solutions

[14], contaminated hospital equipment [32] and

even from the environment [38].

Due to the severity of infections caused in CF

patients and the poor outcome of infected

patients, as well as the resistance to multiple

antibiotics, it becomes interesting to investigate

possible alternative treatments and mechanisms

to avoid the infection.

There have been several attempts for the

development of a vaccine against species of the

Bcc, several strategies have been employed in

order to achieve this objective [9; 13], with the

most promising results being achieved with outer

membrane proteins (OMP) [5; 43]. However

strategies employed were not able to achieve the

desired efficiency.

In the last decade there has been an increasing

interest in the use of outer membrane vesicles

(OMV) for the development of vaccines, since

these vesicles exhibit some characteristics which

make them interesting candidates for vaccine

development, such as the presence of multiple

antigens, being acellular and stability at 4ºC [19;

51]. Studies in vaccine development using OMVs

have been performed in several bacteria species

[2; 27; 50], with one OMV-based vaccine against

N. meningitids already in the market [24] and

studies for a vaccine against B. pseudomallei

well advanced [47; 49]. However there are no

studies regarding development of OMV vaccines

against Bcc species.

OMVs are formed by the blebbing of the outer

membrane of bacterial cell wall, they do not result

from cell death or lysis, and are produced during

all phases of growth [7; 44; 69]. OMVs are

spherical with a bilayered membrane, with sizes

ranging from 50 to 250nm and produced by

organisms from the three branches of the tree of

life. They are produced by Gram-negative and

Gram-positive bacteria [28; 50], Fungi [48],

archaea [18], and even in parasites [55], in which

seems to be an evolutionary conserved

mechanism [35]. These vesicles have several

biological functions which provide an advantage

to OMV-producing bacteria, they provide

resistance to antibiotics [42; 52], provide

protection from environmental stresses [44],

deliver virulence factors [28; 65], among others

[6; 34].

OMVs may be isolated from in vitro cultures.

For the isolation of OMVs the most usual process

is to perform a bacterial culture until late

logarithmic phase, followed by a series of

centrifugations and filtrations in order to remove

bacteria and concentrate the vesicles [28].

However in order to obtain purified OMVs it is

necessary to apply a purification step, usually

OMVs are purified by density gradient

centrifugation [28]. In order to evaluate the purity

of the OMV preparation the method most

commonly used is Transmission Electron

Microscopy (TEM), which allows to visualize the

vesicles and other contaminants, if they are

present [28; 41; 44].

This study will evaluate the purity of a B.

cenocepacia OMV preparation. Followed by

optimization of OMVs production and purification

methods. Purified OMVs will be used for

cytotoxicity assays, in order to perform a

preliminary evaluation of OMV cytotoxicity.

Materials and methods

2.1 Culture conditions and bacterial strains- 10

different bacterial strains were used in this work (Table

1), the strains were maintained frozen in glycerol at -

80ºC. When necessary strains were defrosted to Luria-

Bertani (LB) plates.

3

OMV production and purification- Vesicles were

purified by a method adapted from Huang et al. [29]

and Allan et al. [1]. Bacteria were grown in 2 L of LB

medium, they were incubated with an initial OD640nm of

0.1 and the culture was stopped when OD640nm

reached 1.5. Cells were removed by pelleting (10,000

x g, 10 min.). Supernatants were filtered through a

0.45µm cellulose membrane (Milipore) and

concentrated via a 100-kDa tangential filtration

concentration unit (GE Healthcare) to approximately 70

mL. The OMVS present in the retentate were pelleted

by ultracentrifugation (150,000 x g, 3h), resuspended

in 1.5 mL of PBS (pH 7.4). Ressuspended vesicles

were filtered through a 0.2 µm polyethersulfone

membrane (Whatman, GE Healthcare).

Size exclusion chromatography (SEC)- Sephacryl

S-1000 Superfine chromatography (GE Healthcare

Life Sciences ) was performed using an Äkta™ Purifier

10 system, using Tricorn™ 5/100 or Tricorn™ 10/600

columns. 1.5mL of OMV solution were loaded using a

1 mL loop with a flow rate of 0.5 mL/min. The flow-

through and elute were continuously collected in 0.5

mL fractions in a FRAC 950 Fraction collector (GE

Healthcare).

Optimization of growth parameters, conditions

and strains used- In order to optimize the OMVs

production process the OMVs production was

evaluated in different Bcc isolates. The selected

isolates were cultured in 150 mL of LB medium. The

initial OD640nm was 0.1 and the culture was maintained

until OD640nm reached 1.5. Analysis of OMV production

in different points of the growth was performed using

the same protocol, differing only in the volume used

(500 mL) and the length of the culture (4h, 10h, 16h

and OD640nm 1.5).

OMVs production under different growth conditions

and mediums was performed using a volume of 500

mL of LB. Two different mediums were used, regular

LB medium and LB medium with 300 mM NaCl.

Culture was performed until OD640nm reached 1.5,

initial OD640nm was 0.1.

Microaerophilic conditions were achieved using an

anaerobic box (Mitsubishi Gas Chemical Company)

and GENbox microaerophilic generator (BioMérieux),

bacteria were incubated over-night, until a OD640nm 1.5

was reached. Cells were removed by centrifugation

(10,000 x g, 10 min). Supernatants were filtered

through a 0.45µm cellulose membrane (Milipore) and

concentrated using centrifugal filter Amicon® (Merck

Millipore) with 100,000 Da NMWL.

Proteomic analysis- OMVs were analyzed by 2-DE

(CBMA- Centre of Molecular Environmental Biology,

Department of Biology, University of Minho, Braga,

Portugal) using the immobilized pH gradient technique

(IPG). OMV samples were resuspended in 7 M urea, 2

M thiourea, 1% (w/v) dodecylmaltoside prior to loading

in the first dimension of IPG. IPG strips used were 24

cm non-linear, with pH range from 3 to 10. Isoelectric

focusing (IEF) was performed during 3 hours at 7,000

V. The completed first-dimensional strip was subjected

to a second dimension using 12% dodecyl sulfate

polyacrylamide gel (SDS-PAGE), separation was

performed at 110V. Gels were stained with silver

nitrate and scanned on a Typhoon Trio laser scanner

(GE Healthcare).

Proteins spots were excised manually and identified

by mass spectrometry (MS) (Proteomic and Analytical

Biochemistry Laboratory, Department of Biology,

University of York, York, UK ). Proteins in gel spots

were digested with trypsin, tryptic peptides analyzed by

matrix-assisted laser desorption ionization (MALDI)-

MS and MS/MS using a Bruker ultraflex III (Bruker-

Daltonic, USA) tandem time of flight (TOF/TOF) mode.

For identification MS spectra were searched using

MASCOT software (Matrix Science) using data from

the NCBI non-redundant protein database.

Classification into functional categories was performed

using the Burkholderia genome database

(http://www.burkholderia.com), TIGR database

(http://www.tigr.org/), the Role Category Lists, KEGG

database (http://www.genome.jp/kegg/) and NCBI

(http://www.ncbi.nlm.nih.gov/). Subcellular prediction

was performed using the bioinformatic tool PSORTb

version 2.0.4 [23].

OMVs quantification (BCA assay)- In this study

Total protein quantification was used as an indicator of

the OMV present in the analyzed sample. Total protein

quantification was performed by a bicinchoninic acid

(BCA) assay kit (Pierce). Standard solutions of bovine

albumin serum (BSA) ranging from 25 to 450 µg/mL

4

Table 1- Characteristics of the Bcc isolates used in this work.

Strain name Other name Source, location Relevant information Source

Burkholderia cenocepacia

K56-2 LMG 18863 CF patient,

Canada

ET-12 lineage BCESM+;

cblA+; recA-A

J. J. LiPuma

J2315 LMG 16656 CF patient, UK ET-12 lineage BCESM+;

cblA+; recA-A

G. Doring

AU1054 LMG 24506 CF patient, USA recA-B PHDC lineage BCCM™/LMG

MCO-3 LMG 24308 Maize

rhizosphere, USA

recA-B BCCM™/LMG

HI2424 LMG 24507 Soil, USA recA-B PHDC lineage BCCM™/LMG

Burkholderia multivorans

ATCC17616 Soil, USA ATCC

D2095 CF patient,

Canada

D. P. Speert

HI2229 LMG 17588 CF patient, USA BCESM-; cblA

- J. J. LiPuma

Burkholderia cepacia

PC783 LMG 1222 Onion, USA J. J. LiPuma

Burkholderia dolosa

AU0654 LMG 18943 CF patient, USA J. J. LiPuma

BCESM: Burkholderia cenocepacia epidemic strain marker – cblA: cable pilus gene – recA: recombinase A gene – PHDC: Philadelphia-District of

Columbia clone – ET: Electrophoretic type

were used to establish a calibration curve. The assays

were performed in 96 wells microplates (Orange

Scientific) with 200 µL of BCA reagent and 10 µL of

sample. Absorbance was read at 562nm in a

Spectrostarnano

(BMG Labtech) microplate reader.

MTT assay- 16HBE14o- human bronchial epithelial

cells were maintained in fibronectin/vitrogen coated

flasks in Minimum Essential Medium with Earle’s salt

(MEM) (Gibco) supplemented with 10% fetal bovine

serum (FBS) (Lonza), 0.292 g/L L-glutamine (Sigma

Aldrich) and Penicilin/Streptomycin 100 U/mL (Gibco),

in a humified atmosphere at 37ºC and 5% CO2. Cells

were seeded in a 96 well microplate (Orange

Scientific) at a density of 7.5x103 cells per well. 0.5 µg,

1 µg and 2 µg of B. cenocepacia AU1054-derived

OMVs were added to 100 µL of MEM medium and

incubated in the same conditions for 24 hours, after

which 20 µL of 5 mg/mL [3-(4,5 dimethylthiazol-2-yl-2,5

tetrazolium bromide)] (MTT) were added in each well

and incubated for 3 hours in the same conditions.

Following the incubation MTT was removed and 150

µL of 40 mM HCl in isopropanol were added and

incubated for 15 min., the crystals were resuspended

and plate’s absorbance was read at 590nm in a

SpectroStarnano

(BMG Labtech) microplate reader.

Galleria melonella killing assays- Galleria

melonella killing assays were based on the methods

previously described [45]. A micrometer was adapted

to control the volume of a mycrosyringe and inject 3.5

µL (0.2 µg) or 10 µL (1µg) of purified OMVs in each

caterpillar, via the hindmost left proleg, previously

sanitized with 70% (v/v) ethanol. Following injection

larvae were placed in petri dishes and incubated in the

dark, at 37ºC. Survival was followed during a 96 hour

period, larvae were considered dead when no

response to touch was displayed. Control larvae were

injected with PBS (pH 7.4).

Transmission Electron Microscopy (TEM)- 10µL

of purified OMV preparation were negatively stained by

freshly prepared 2% (v/v) uranyl acetate and applied to

200-mesh carbon-coated cooper grids (Electron

Microscopy Sciences), being visualized with a JEOL

1200-EX transmission electron microscope.

5

Results

B. cenocepacia K56-2-derived OMVs purity

analysis

OMVs were isolated from cultures of B.

cenocepacia K-56-2 in order to evaluate the

purity of the OMVs by TEM. From a 2 L culture of

B. cenocepacia, it was possible to obtain,

following isolation of OMVs, a yield of 0.1 ±0.03

mg of OMVs/L of initial culture. The OMVs

obtained were analyzed by TEM, allowing to

confirm the presence of OMVs with diameters

ranging from 10 nm to 300 nm, with OMVs with

50 nm being more commonly observed (Fig.1). It

is also possible to observe the presence of

contaminants such as flagella and cable pili

(Fig.1).

Proteomic analysis of B. cenocepacia K56-2-

derived OMVs

Three different samples of purified OMVs were

subjected to a proteomic analysis. 33 different

proteins were identified from 73 spots (Fig.2).

The proteins identified were divided into 8

different functional groups with the most

abundant groups being cell envelope biogenesis

and outer membrane, energy conversion and

amino acid transport and metabolism. It was

possible to observe the presence of a large

number of cytoplasmic proteins which is

expected since OMVs incorporate proteins from

multiple bacterial compartments. It is also

possible to observe the presence of cable pilus

and flagella, which corroborates TEM results.

A set of proteins identified in the proteomic

analysis is of particular interest. This group is

composed by proteins which are immunogenic. In

this study 6 of 15 immunogenic proteins were

identified [54].

Fig. 1- TEM analysis of OMV samples obtained from B.

cenocepacia K56-2 in vitro growth

Optimization of OMVs production

The process of OMVs production was

optimized, by analyzing OMVs production by

different strains of the Bcc, under different

incubation periods, and under different growth

conditions. The strains analyzed were chosen

from the four clinically most relevant Bcc species

[8; 16]. From the strains tested B. cenocepacia

K56-2 achieved the best yields (Fig.3).

Fig. 2- Proteome reference 2-DE gel of B cenocepacia K56-2-

derived OMVs. On the left is represented the scale of the

molecular weight (higher on top to lower on bottom), and on

the top is represented the pH scale (from 3 to 10). The circles

and numbers indicate excised protein spots which were

analyzed by MALDI-TOF

6

For the optimization of culture time both B.

cenocepacia K56-2 and B. multivorans HI2229

were tested. The OMVs produced in cultures

stopped at different points suggested that longer

incubation times lead to increased OMVs

production, however the increase in longer

incubation times may be due to cell lysis, which

was not analyzed in this study.

The final optimization step was the growth

conditions. Two stresses, which are present in

CF lung, were tested, namely osmotic stress [33]

represented by LB medium with 300 mM NaCl,

and reduced oxygen concentration [68]

represented by microaerophilic conditions.

The growth conditions tested yielded different

results, while the use of LB with 300 mM of NaCl

resulted in reduced OMVs production compared

with LB medium, the use of microaerophilic

conditions resulted in an increase of OMV

production.

Despite the results obtained in the optimization

of OMVs production process, the conditions were

maintained. The strains being used was the best

producer. The increase in the total protein

produced in longer incubation periods may be

caused by cell lysis.

Table 2 OMV yields (mg/L of initial culture) obtained following culture of B. cenocepacia K56-2, under different growth conditions.

Growth conditions OMV yield (mg/L)

Broth Stiring Oxygen

LB 250 aerobic 0.9

0.3M NaCl LB

250 aerobic 0.7

LB 60 microaerophilic 1.5

Size exclusion chromatography of B.

cenocepacia K56-2-derived OMVs

The results obtained in TEM and proteomic

analysis of B. cenocepacia K56-2-derived OMVs

suggested that further purification of the samples

was required.

Fig. 3- Graphic representation of the OMVs yield (mg of OMVs/L of initial culture) produced by the different strains used in the strain screening

To further purify the OMVs a SEC step was

applied to the samples, to increase purity of the

preparation.

The OMVs were loaded into a Tricorn™ 5/100

with 30 cm of length. The use of this column did

not allow complete separation of the OMVs and

the contaminants present in the sample, with only

a partial separation being achieved.

In order to increase the separation between the

OMVs and the contaminants in the sample, a

longer column was used. The reasoning behind

the employment of a longer column was that the

increased retention time in the column would

allow for a more complete separation.

The longer column used was a Tricorn™

10/600 with 60 cm of length. OMVs were also

loaded into the column. The separation of the

samples was greater using the longer column,

however complete separation was not achieved,

with an small overlap between the OMVs

fractions and contaminants fraction, thus not

achieving complete separation between the

samples and contaminants.

0

0,1

0,2

0,3

0,4

0,5

B. c

eno

cep

acia

K5

6-2

B. c

eno

cep

acia

J2

31

5

B. c

eno

cep

acia

AU

10

54

B. c

eno

cep

acia

MC

O-3

B. c

eno

cep

acia

HI2

42

4

B. m

ult

ivo

ran

s A

TCC

…

B. m

ult

ivo

ran

s D

20

95

B. m

ult

ivo

ran

s H

I22

29

B. c

epac

ia P

C7

83

B. d

olo

sa A

U0

65

4

OM

V y

ield

(m

g/L)

Strains tested

7

Production of B. cenocepacia AU 1054-

derived OMVs

Since the use of SEC for the purification of B.

cenocepacia K56-2-derived OMVs did not

achieved the desired results, a different strategy

was employed for the achievement of purified

OMVs.

Taking into consideration the observations from

the TEM analysis of B. cenocepacia K56-2-

derived OMVs, in which Cable pili and flagella

were the main contaminants, we reasoned that

the use of a strain not capable of producing cable

pili would reduce the contamination of the OMV

preparation.

To achieve this objective the strain used for

OMVs production was switched to B.

cenocepacia AU 1054, which is CblA-, thus not

being able to produce cable pili. This strain was

used for production of OMVs, the protocol used

for production and isolation of B. cenocepacia

AU1054-derived OMVs was the same as the one

used for B. cenocepacia K56-2. From the 2 L of

initial culture and following the isolation of OMVs

from B. cenocepacia AU 1054-derived OMVs, it

was possible to obtain 0.04 ±0.003 mg of OMV/L

of initial culture.

Transmission Electron Microscopy of B.

cenocepacia AU1054-derived OMVs

B. cenocepacia AU1054-derived OMVs were

analyzed by TEM to confirm the presence of

vesicles and evaluate the purity of the OMVs.

TEM analysis of OMVs isolated from B.

cenocepacia AU1054 revealed the presence of

OMVs with diameters between 100 and 150 nm.

It was also possible to observe vesicles with

smaller diameters (75 to 100nm) and large

diameters (175 to 225 nm) were less commonly

observed (Fig. 3).

It is also important to notice that cable pilus or

flagella were not observed in these preparations

(Fig.4).

Fig. 4- Transmission electron microscopy image from a sample of B. cenocepacia AU1054-derived OMVs.

In vitro cytotoxic effects of B. cenocepacia

AU1054-derived OMVs

Following the isolation of purified OMVs from

B. cenocepacia AU1054 it became possible to

perform a preliminary evaluation of cytotoxic

potential of OMVs. Initial cytotoxic assays were

performed in vitro by performing MTT assays in

16HBE14o- human bronchial epithelial cells.

Exposure to OMVs revealed a reduction in cell

survival, which increased with an increase in the

amount of OMVs to which cells were exposed,

reaching almost 50% in the highest quantity of

OMVs (Fig. 5).

Fig. 5- Cell survival after exposure to purified B. cenocepacia AU1054-derived OMVs during a 24 hour period.

0

20

40

60

80

100

0,5 1 2

Surv

ival

(%

)

OMVs (ug)

8

In vivo cytotoxic effects of B. cenocepacia

AU1054-derived OMVs

Following the in vitro cytotoxic assays

performed with OMVs, the cytotoxic effects of

OMVs were also tested in Galleria melonella. The

results obtained from the in vivo testing of

cytotoxic effects were different from the MTT

assays. Exposure to purified OMVs in G.

melonella did not result in reduced survival, in

fact all larvae injected with OMVs survived the

testing period (Table 3).

Table 3- Number of Galleria mellonella caterpillars alive after a 96 hours period after being injected with a OMV sample isolated from the strain B. cenocepacia AU1054

Sample Caterpillars

inoculated

Caterpillars

alive after 96

hours

Control 10 10

AU1054

(0.2µg)

10

10

AU1054

(1µg)

10

10

Discussion The development of a vaccine against Bcc

species has been subject to various studies,

however they were not able to achieve enough

protection [13].

Recently OMVs have been employed in the

development of vaccines due to their

characteristics [51].

Using B. cenocepacia K56-2 we were able to

produce and isolate OMVs. Assessment of purity

of isolated OMVs by TEM and proteomic analysis

revealed the presence of contaminants, requiring

further purification. Several authors use density

gradients for OMV purification [4; 63], however

this purification process is not easily scalable.

Taking those observations into consideration

we employed a size exclusion chromatography

step for the purification of isolated OMVs based

on a previously described protocol for OMVs

purification in N. meningitidis [59]. Testing of SEC

efficiency in the purification of OMVs revealed

that the process was not capable of achieving the

desired results, therefore a different strategy was

employed.

As said before the proteomic analysis of

B.cenocepacia K56-2-derived OMVs revealed the

presence of 6 of 15 immunoproteins across B.

cenocepacia and B. multivorans species, which

are good candidates for vaccine development,

namely chaperonin GroEL, elongation factor Tu,

OMPA/MotB domain-containing protein, porin,

alkyl hidroperoxide reductase/Thiol specific

antioxidant and phospopyruvate hydratase [54].

Besides, the 33 proteins identified in the

proteomic analysis may also be encountered in

OMVs derived from other organisms, such as P.

aeruginosa [11], N. lactamica [60] and N.

meningitidis [67].

In the optimization of the OMVs production

process, when optimizing culture length a B.

multivorans strain was used, this strain is a

clinical isolate, and was used in order to have a

representative of the two clinically relevant

species. The results suggested that using longer

incubation periods lead to increased OMVs

yields, however the increase in OMVs production

may be caused by cell lysis. Since the cell lysis

was not evaluated, the incubation length was

maintained, this incubation period is similar to the

period used by other authors, which stop the

cultures at lat log phase/early stationary [4; 28].

Growth conditions have an important effect in

vesicle production [40; 41]. Taking that

observation into consideration it was necessary

to evaluate the effect of growth conditions in

OMV production. Two different conditions were

tested, osmotic stress and microaerophilic

9

conditions, which, as previously stated, represent

the conditions encountered by bacteria in host’s

lungs. Literature reports regarding OMV

production under osmotic stress and

microaerophilic conditions are scarce, however

some reports may be found. Regarding OMV

production under osmotic stress Fulsundar et al.

have studied the effect of different stresses in

vesicle production in Acineobacter bayly,

concluding that LB medium supplemented with

500 mM of NaCl for the culture of A. bayly,

resulted in a small increase in vesiculation [22].

The results obtained in the present study are not

in agreement with the results obtained by

Fulsundar et al. In respect to reduced oxygen

availability, Tran et al., have evaluated the

release of Shiga toxin by a strain of E. coli. They

evaluated the release of Shiga toxin in

microaerobic conditions, and concluded that the

release of Shiga toxin was reduced under these

conditions [56]. Since Shiga toxin is released by

OMVs [17], it may be possible that in E. coli

strain used also releases Shiga toxin via OMVs.

The results obtained in the present study, once

again, are not in agreement with the literature,

having achieved higher OMV yields in

microaerophilic conditions.

For the development of an alternative

purification strategy, and taking into account the

results obtained in TEM and proteomic analysis

of B. cenocepacia K56-2, the strain used for

OMVs production was replaced with B.

cenocepacia AU 1054, which does not produce

cable pilus or flagella, facilitating the purification

of OMVs.

The yields obtained by B. cenocepacia AU

1054 are lower than the yields obtained by B.

cenocepacia K56-2, however TEM analysis

revealed the presence of OMVs in B.

cenocepacia AU 1054 OMVs preparations,

without the presence of contaminants, thus

achieving the objective of producing purified

OMV preparations using an alternative strategy.

The evaluation of cytotoxic effects both in vitro

and in vivo revealed some differences, while the

in vitro model exhibited reduced cell survival, the

in vivo model did not exhibited reduced survival.

This discrepancy between the models used may

be explained by the lipopolyssacharide (LPS)

toxicity, which may cause some toxicity [51].

Work performed by Elmi et al., revealed that

Campylobacter jejuni 11168H-derived OMVs

revealed dose-dependent cytotoxic effects in

Caco-2 intestinal epithelial cells and in G.

mellonella [20], while in this work only in

16HBE14o- human bronchial epithelial cells was

possible to observe cytotoxic effects derived from

OMVs exposure, which may be explained by the

higher OMVs dose used in the work performed

by Elmi et al [20].

Conclusions and future work Some aspects of this work should be further

investigated. The effect of osmotic stress and

microaerophilic conditions should be further

studied, repeating the assays performed, in order

to achieve a more solid conclusion, since

literature reports are not in agreement with the

results obtained.

The cytotoxic effects of OMVs should also be

subjected to a more in-depth analysis, evaluating

the effect of exposure to higher amounts of

OMVs.

One interesting strategy which could be used in

the future to overcome the low OMVs yields

obtained is the use of genetic engineering

techniques to cause a mutation in the rmpM

gene, which would lead to a mutant with

increased vesiculation [57], thus increasing the

OMVs yields.

10

This work lead to the development of a strategy

to obtain purified OMVs from B. cenocepacia.

This strategy allows the production of OMVs

suitable for the development of future studies in

the development of OMVs-based vaccines.

Acknowledgements We would like to acknowledge Dr. Pedro M.

Santos at CBMA- Centre of Molecular and

Environmental Biology, Department of Biology,

University of Minho, Braga, Portugal, for

performing the proteomic analysis.

We would also like to acknowledge Dr. António

Pedro Alves de Matos, at Centro de Investigação

Interdisciplinar Egas Moniz (CiiEM), Egas Moniz–

Cooperativa de Ensino Superior, Caparica,

Portugal, for performing the Transmission

Electron Microscopy.

References 1. Allan, N. D., Kooi, C., Sokol, P. a & Beveridge, T.

J. Putative virulence factors are released in association with membrane vesicles from Burkholderia cepacia. Can. J. Microbiol. 49, 613–624 (2003).

2. Alaniz, R. C., Deatherage, B. L., Lara, J. C. & Cookson, B. T. Membrane vesicles are immunogenic facsimiles of Salmonella typhimurium that potently activate dendritic cells, prime B and T cell responses, and stimulate protective immunity in vivo. J. Immunol. 179, 7692–7701 (2007).

3. Bashkar, K. R. et al. Profound increase in viscosity and aggregation of pig gastric mucin at low pH. Am. J. Physiol. 261, 827–832 (1991).

4. Bauman, S. J. & Kuehn, M. J. Purification of outer membrane vesicles from Pseudomonas aeruginosa and their activation of an IL-8 response. Microbes Infect. 8, 2400–2408 (2006).

5. Bertot, G. M. et al. Nasal immunization with Burkholderia multivorans outer membrane proteins and the mucosal adjuvant adamantylamide dipeptide confers efficient protection against experimental lung infections with B. multivorans and B. cenocepacia. Infect. Immun. 75, 2740–2752 (2007).

6. Beveridge, T. J. Bacteriolytic effect of membrane vesicles from Pseudomonas aeruginosa on other bacteria including pathogens : conceptually new Bacteriolytic Effect of Membrane Vesicles from Pseudomonas aeruginosa on Other Bacteria Including Pathogens : Conceptually New A. 178, 2767–2774 (1996).

7. Beveridge, T. J. Sructure of fram-negative cell walls and their derived mebrane vesicles. J. Bacteriol. 181, 4725–4733 (1999).

8. Biddick, R., Spilker, T., Martin, A. & LiPuma, J. J. Evidence of transmission of Burkholderia cepacia, Burkholderia multivorans and

Burkholderia dolosa among persons with cystic fibrosis. FEMS Microbiol. Lett. 228, 57–62 (2003).

9. Choh, L.-C. et al. Burkholderia vaccines: are we moving forward? Front. Cell. Infect. Microbiol. 3, 5 (2013).

10. Choi, D.-S. et al. Proteomic analysis of outer membrane vesicles derived from Pseudomonas aeruginosa. Proteomics 11, 3424–3429 (2011).

11. Conway, B. D., Chu, K. K., Bylund, J., Altman, E. & Speert, D. P. Production of Exopolysaccharide by Burkholderia cenocepacia Results in Altered Cell-Surface Interactions and Altered Bacterial Clearance in Mice. 4, 957–966 (2004).

12. Conway, B. D., Venu, V. & Speert, D. P. Biofilm Formation and Acyl Homoserine Lactone Production in the Burkholderia cepacia Complex Biofilm Formation and Acyl Homoserine Lactone Production in the Burkholderia cepacia Complex. J. Bacteriol. 184, 5678–5685 (2002).

13. Corbett, C. R., Burtnick, M. N., Kooi, C., Woods, D. E. & Sokol, P. A. An extracellular zinc metalloprotease gene of Burkholderia cepacia. Microbiology 149, 2263–2271 (2003).

14. Cunha, V. et al. Exceptionally High Representation of Burkholderia cepacia among B . cepacia Complex Isolates Recovered from the Major Portuguese Cystic Fibrosis Center ᰔ. 45, 1628–1633 (2007).

15. Depoorter, E. et al. Burkholderia: an update on taxonomy and biotechnological potential as antibiotic producers. Appl. Microbiol. Biotechnol. 100, 5215–5229 (2016).

16. Drevinek, P. & Mahenthiralingam, E. Burkholderia cenocepacia in cystic fibrosis: epidemiology and molecular mechanisms of virulence. Clin. Microbiol. Infect. 16, 821–830 (2010).

17. Dutta, S. et al. Release of Shiga toxin by membrane vesicles in Shigella dysenteriae serotype 1 strains and in vitro effects of antimicrobials on toxin production and release. Microbiol. Immunol. 48, 965–9 (2004).

18. Ellen, A. F. et al. Proteomic analysis of secreted membrane vesicles of archaeal Sulfolobus species reveals the presence of endosome sorting complex components. Extremophiles 13, 67–79 (2009).

19. Ellis, T. N. & Kuehn, M. J. Virulence and Immunomodulatory Roles of Bacterial Outer Membrane Vesicles. Microbiol. Mol. Biol. Rev. 74, 81–94 (2010).

20. Elmi, A. et al. Campylobacter jejuni outer membrane vesicles play an important role in bacterial interactions with human intestinal epithelial cells. Infect. Immun. 80, 4089–98 (2012).

21. Folkesson, A. et al. Adaptation of Pseudomonas aeruginosa to the cystic fibrosis airway: an evolutionary perspective. Nat. Rev Microbiol 10, 841–51 (2012).

22. Fulsundar, S. et al. Gene transfer potential of outer membrane vesicles of Acinetobacter baylyi and effects of stress on vesiculation. Appl. Environ. Microbiol. 80, 3469–3483 (2014).

23. Gardy, J. L. et al. PSORTb v.2.0: Expanded prediction of bacterial protein subcellular localization and insights gained from comparative proteome analysis. Bioinformatics 21, 617–623 (2005).

24. Gorringe, A. R. & Pajon, R. Bexsero: A multicomponent vaccine for prevention of meningococcal disease. Hum. Vaccines Immunother. 8, 174–183 (2012).

25. Govan, J. R. W. et al. Evidence for transmission of Pseudomonas cepacia by social contact in cystic fibrosis . The Lancet. 342, 15–19 (1993).

11

26. Govan, J. R. W., Hughes, J. E. & Vandamme, P. Burkholderia cepacia: Medical, taxonomic and ecological issues. J. Med. Microbiol. 45, 395–407 (1996).

27. Holst, J. et al. Properties and clinical performance of vaccines containing outer membrane vesicles from Neisseria meningitidis. Vaccine 27, 3–12 (2009).

28. Horstman, A. L. & Kuehn, M. J. Enterotoxigenic Escherichia coli Secretes Active Heat-labile Enterotoxin via Outer Membrane Vesicles *. 275, 12489–12496 (2000).

29. Huang, W. et al. Immunization against multidrug-resistant Acinetobacter baumannii effectively protects mice in both pneumonia and sepsis models. PLoS One 9, 16–19 (2014).

30. Isles, A. et al. Pseudomonas cepacia infection in cystic fibrosis: an emerging problem. J Pediatr 104, 206–210 (1984).

31. Kadurugamuwa, 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 Novel Mechanism of Enzyme-Secretion. J. Bacteriol. 177, 3998–4008 (1995).

32. Kamiya, A. BRIEF Microbial contamination and disinfectants. Am. J. Infect. Control 24, 389–395 (1996).

33. Kilbourn, J. P. Bacterial content and ionic composition of sputum in cystic fibrosis. The Lancet. 311, 334 (1978).

34. Lappann, M., Otto, A., Becher, D. & Vogel, U. Comparative proteome analysis of spontaneous outer membrane vesicles and purified outer membranes of neisseria meningitidis. J. Bacteriol. 195, 4425–4435 (2013).

35. Lee, E. Y. et al. Gram-positive bacteria produce membrane vesicles: Proteomics-based characterization of Staphylococcus aureus-derived membrane vesicles. Proteomics 9, 5425–5436 (2009).

36. LiPuma, J. J. et al. Person-to-person transmision of Pseudomonas cepacia between patients with cystic fibrosis . The Lancet.336, 1094–1096 (1990).

37. LiPuma, J. J. Burkholderia cepacia: Management issues and new insights. Clin. Chest Med. 19, 473–486 (1998).

38. LiPuma, J. J., Spilker, T., Coenye, T. & Gonzalez, C. F. An epidemic Burkholderia cepacia complex strain identified in soil. Lancet 359, 2002–2003 (2002).

39. LiPuma, J. J. The changing microbial epidemiology in cystic fibrosis. Clin. Microbiol. Rev. 23, 299–323 (2010).

40. MacDonald, I. A. & Kuehna, M. J. Stress-induced outer membrane vesicle production by Pseudomonas aeruginosa. J. Bacteriol. 195, 2971–2981 (2013).

41. McMahon, K. J., Castelli, M. E., Vescovi, E. G. & Feldman, M. F. Biogenesis of outer membrane vesicles in Serratia marcescens is thermoregulated and can be induced by activation of the Rcs phosphorelay system. J. Bacteriol. 194, 3241–3249 (2012).

42. Manning, A. J. & Kuehn, M. J. Contribution of bacterial outer membrane vesicles to innate bacterial defense. BMC Microbiol 11, 258 (2011).

43. Makidon, P. E. et al. Induction of immune response to the 17 kDa OMPA Burkholderia cenocepacia polypeptide and protection against pulmonary infection in mice after nasal vaccination with an OMP nanoemulsion-based vaccine. Med. Microbiol. Immunol. 199, 81–92 (2010).

44. McBroom, A. J. & Kuehn, M. J. Release of outer membrane vesicles by Gram-negative bacteria is a novel envelope stress response. Mol. Microbiol. 63, 545–558 (2007).

45. Mil-Homens, D., Bernardes, N. & Fialho, A. M. The antibacterial properties of docosahexaenoic omega-3 fatty acid against the cystic fibrosis multiresistant pathogen Burkholderia cenocepacia. FEMS Microbiol. Lett. 328, 61–69 (2012).

46. Mil-Homens, D. & Fialho, A. M. A BCAM0223 mutant of Burkholderia cenocepacia is deficient in hemagglutination, serum resistance, adhesion to epithelial cells and virulence. PLoS One 7, (2012b).

47. Nieves, W. et al. A naturally derived outer-membrane vesicle vaccine protects against lethal pulmonary Burkholderia pseudomallei infection. Vaccine 29, 8381–8389 (2011).

48. Oliveira, D. L. et al. Characterization of yeast extracellular vesicles: Evidence for the participation of different pathways of cellular traffic in vesicle biogenesis. PLoS One 5, (2010).

49. Petersen, H., Nieves, W., Russell-Lodrigue, K., Roy, C. J. & Morici, L. a. Evaluation of a Burkholderia pseudomallei Outer Membrane Vesicle Vaccine in Nonhuman Primates. Procedia Vaccinol. 8, 38–42 (2014).

50. Rivera, J. et al. Bacillus anthracis produces membrane-derived vesicles containing biologically active toxins. Proc. Natl. Acad. Sci. U. S. A. 107, 19002–7 (2010).

51. Rosenthal, J. A., Chen, L., Baker, J. L., Putnam, D. & DeLisa, M. P. Pathogen-like particles: Biomimetic vaccine carriers engineered at the nanoscale. Curr. Opin. Biotechnol. 28, 51–58 (2014).

52. Schaar, V., Nordström, T., Mörgelin, M. & Riesbeck, K. Moraxella catarrhalis outer membrane vesicles carry β-lactamase and promote survival of Streptococcus pneumoniae and Haemophilus influenzae by inactivating amoxicillin. Antimicrob. Agents Chemother. 55, 3845–3853 (2011).

53. Sheppard, D. N. & Welsh, M. J. Structure and function of the CFTR chloride channel. Physiol. Rev. 79, S23–S45 (1999).

54. Shinoy, M. et al. Immunoproteomic analysis of proteins expressed by two related pathogens, Burkholderia multivorans and burkholderia cenocepacia, during human infection. PLoS One 8, (2013).

55. Silverman, J. M. et al. Proteomic analysis of the secretome of Leishmania donovani. Genome Biol. 9, R35 (2008).

56. Tran, S. L., Billoud, L., Lewis, S. B., Phillips, A. D. & Schüller, S. Shiga toxin production and translocation during microaerobic human colonic infection with Shiga toxin-producing E.coliO157: H7 and O104: H4. Cell. Microbiol. 16, 1255–1266 (2014).

57. Van Der Ley, P. et al. Modification of lipid a biosynthesis in Neisseria meningitidis lpxL mutants: Influence on lipopolysaccharide structure, toxicity, and adjuvant activity. Infect. Immun. 69, 5981–5990 (2001).

58. van de Waterbeemd, B. et al. Improved OMV vaccine against Neisseria meningitidis using genetically engineered strains and a detergent-free purification process. Vaccine 28, 4810–4816 (2010).

59. van de Waterbeemd, B. et al. Improved Production Process for Native Outer Membrane Vesicle Vaccine against Neisseria meningitidis. PLoS One 8, 1–11 (2013).

12

60. Vaughan, T. E. et al. Proteomic analysis of Neisseria lactamica and N eisseria meningitidis outer membrane vesicle vaccine antigens. Vaccine 24, 5277–5293 (2006).

61. Veerman, E. C. et al. Viscosity of salivary mucins: effect of pH and ionic strength and role of sialic acid. J. Biol. Bucc.17, 297–306 (1989)

62. Veerman, E. C. I. et al. Sulfated glycans on oral mucin as receptors for Helicobacter pylori. Glycobiology 7, 737–743 (1997).

63. Vida, A. et al. Neutralization of Neisseria meningitidis outer membrane vesicles. Inflamm. Res. 60, 801–5 (2011).

64. Visser, M. B., Majumdar, S., Hani, E. & Sokol, P. A. Importance of the Ornibactin and Pyochelin Siderophore Transport Systems in Burkholderia cenocepacia Lung Infections Importance of the Ornibactin and Pyochelin Siderophore Transport Systems in Burkholderia cenocepacia Lung Infections. Infect. Immun. 72, 2850–2857 (2004).

65. Wai, S. N. et al. Vesicle-mediated export and assembly of pore-forming oligomers of the enterobacterial ClyA cytotoxin. Cell 115, 25–35 (2003).

66. Wanke, C. A., Cronan, S., Goss, C., Chadee, K. & Guerrant, R. L. Characterization of binding of Escherichia coli strains which are enteropathogens to small-bowel mucin. Infect. Immun. 58, 794–800 (1990).

67. Williams, J. N. et al. Proteomic analysis of outer membranes and vesicles from wild-type serogroup B Neisseria meningitidis and a lipopolysaccharide-deficient mutant. Infect. Immun. 75, 1364–1372 (2007).

68. Worlitzsch, D. et al. Effects of reduced mucus oxygen concentration in airway Pseudomonas infections of cystic fibrosis patients. J. Clin. Invest. 109, 317–325 (2002).

69. Zhou, L., Srisatjaluk, R., Justus, D. E. & Doyle, R. J. On the origin of membrane vesicles in Gram-negative bacteria. FEMS Microbiol. Lett. 163, 223–228 (1998).