PROTEOMICS IN GRAM-NEGATIVE BACTERIAL OUTER
MEMBRANE VESICLES

Eun-Young Lee,

1

Dong-Sic Choi,

1

Kwang-Pyo Kim,

2

and Yong Song Gho

1

*

1

Department of Life Science and Division of Molecular and Life Sciences,

Pohang University of Science and Technology, Pohang, Republic of Korea

2

Institute of Biomedical Science and Technology, Department of Molecular

Biotechnology, Konkuk University, Seoul, Republic of Korea

Received 12 December 2007; accepted 28 February 2008

Published online 17 April 2008 in Wiley InterScience (www.interscience.wiley.com) DOI 10.1002/mas.20175

Gram-negative bacteria constitutively secrete outer membrane
vesicles (OMVs) into the extracellular milieu. Recent research
in this area has revealed that OMVs may act as intercellular
communicasomes in polyspecies communities by enhancing
bacterial survival and pathogenesis in hosts. However, the
mechanisms of vesicle formation and the pathophysiological
roles of OMVs have not been clearly defined. While it is
obvious that mass spectrometry-based proteomics offers great
opportunities for improving our knowledge of bacterial OMVs,
limited proteomic data are available for OMVs. The present
review aims to give an overview of the previous biochemical,
biological, and proteomic studies in the emerging field of
bacterial OMVs, and to give future directions for high-
throughput and comparative proteomic studies of OMVs that
originate from diverse Gram-negative bacteria under various
environmental conditions. This article will hopefully stimulate
further efforts to construct a comprehensive proteome data-
base of bacterial OMVs that will help us not only to elucidate
the biogenesis and functions of OMVs but also to develop
diagnostic tools, vaccines, and antibiotics effective against
pathogenic bacteria. # 2008 Wiley Periodicals, Inc., Mass
Spec Rev 27:535–555, 2008
Keywords: outer membrane vesicles; gram-negative bacteria;
communicasomes; antibiotics; proteomics; vaccines

I. INTRODUCTION

Communication between cells and the environment is an
essential process in living organisms, and intercellular commu-
nication is believed to be mediated mainly by the secretion of
soluble factors, cell-to-cell contacts, and tunneling machinery

such as nanotubes (Ratajczak et al., 2006). Recently, a
mechanism mediated by membrane vesicles (MVs), which are
spherical, bilayered proteolipids with an average diameter of
0.03–1 mm, has drawn much attention (Beveridge, 1999;
Ratajczak et al., 2006). The secretion of MVs is a universal
cellular process occurring from simple organisms to complex
multicellular organisms, including humans (Thery, Zitvogel,
& Amigorena, 2002; Mashburn-Warren & Whiteley, 2006).
Throughout evolution, both prokaryotic and eukaryotic cells
have adapted to manipulate MVs for intercellular communi-
cation via outer membrane vesicles (OMVs) in the case of
Gram-negative bacteria and microvesicles in eukaryotic cells.
Increasing evidence suggests that MVs act as potent communi-
casomes, that is, nano-sized extracellular organelles that play
diverse roles in intercellular communication (Choi et al., 2007),
and that the biogenesis and functions of MVs may share many
features in different biological systems. Thus, the study of MVs
provides crucial keys to understanding the intercellular commu-
nication network in living organisms and the evolutionary
connections between prokaryotes and eukaryotes (Mashburn &
Whiteley, 2005).

A wide variety of Gram-negative bacteria constitutively

secrete OMVs during growth (Beveridge, 1999), including
Escherichia coli, Neisseria meningitidis, Pseudomonas aerugi-
nosa, Shigella flexneri, and Helicobacter pylori (Devoe &
Gilchrist, 1973; Hoekstra et al., 1976; Fiocca et al., 1999;
Kadurugamuwa & Beveridge, 1999). OMVs are spherical,
bilayered proteolipids with an average diameter of 20–200 nm;
they are composed of outer membrane proteins, lipopolysac-
charide (LPS), outer membrane lipids, periplasmic proteins,
cytoplasmic proteins, DNA, RNA, and other factors associated
with virulence (Horstman & Kuehn, 2000; Wai et al., 2003;
Kuehn & Kesty, 2005; Bauman & Kuehn, 2006; Nevot et al.,
2006; Lee et al., 2007). Studies of OMVs from diverse bacterial
strains suggest their roles in the delivery of toxins to host cells, the
transfer of proteins and genetic material between bacterial cells,
cell-to-cell signals, and the elimination of competing organisms
(Kuehn & Kesty, 2005; Mashburn-Warren & Whiteley, 2006).
Because OMVs are essential to bacterial survival and patho-
genesis in the host, modulation of vesicle formation and their
functions may be a useful objective in relation to the development
of antibiotics (Henry et al., 2004; Lee et al., 2007).

Although recent research in this area has revealed the diverse

functions of OMVs, the mechanisms of vesicle formation and of
protein sorting into OMVs, as well as the pathophysiological

Mass Spectrometry Reviews, 2008,

27, 535– 555

# 2008 by Wiley Periodicals, Inc.

————

Contract grant sponsor: National R&D Program for Cancer Control,
Ministry of Health & Welfare, Republic of Korea; Contract grant
number: 0320380-2; Contract grant sponsor: Korea Basic Science
Institute K-MeP; Contract grant number: T27021; Contract grant
sponsor: Korea Science and Engineering Foundation (KOSEF)
(MOST); Contract grant number: R15-2004-033-05001-0; Contract
grant sponsor: Brain Korea 21 fellowship.
*Correspondence to: Yong Song Gho, Department of Life Science,
Division of Molecular and Life Sciences, Pohang University of Science
and Technology, San31 Hyojadong, Pohang, Kyungbuk 790-784,
Republic of Korea. E-mail: [e-mail]

roles of OMVs, have not been clearly defined. To address these
issues, vesicular proteins should be comprehensively identified.
Proteomics offers a powerful approach to decode the protein
components of OMVs. Mass spectrometry (MS)-based proteo-
mic studies have been used in human microvesicles to identify
thousands of vesicle-associated proteins from diverse cancer cell
lines, immune cells, and human fluids including serum, urine, and
breast milk (Pisitkun, Shen, & Knepper, 2004; Jin et al., 2005;
Yates et al., 2005; Admyre et al., 2007; Choi et al., 2007).
However, only a few proteomic analyses of bacterial OMVs have
been reported, although they are more ubiquitous and easier to
obtain than human samples (Post et al., 2005; Bauman & Kuehn,
2006; Nevot et al., 2006; Lee et al., 2007). These studies did not
achieve high-throughput proteomics and identified only a small
number of well-known proteins, except for E. coli-derived OMVs
(Lee et al., 2007). In contrast to native OMVs, several proteomic
studies have been performed on detergent-extracted OMVs
(DOMVs), which are made from whole bacteria with a detergent
treatment (Nally et al., 2005; Ferrari et al., 2006; Uli et al., 2006;
Vipond et al., 2006). Since outer membrane proteins and LPS of
OMVs can induce a host immune response, DOMVs from
pathogenic strains are promising vaccine candidates. DOMVs
derived from N. meningitidis are now in clinical trials (Girard
et al., 2006). However, although DOMVs are clinically important
and similar in size and morphology to native OMVs (Ferrari et al.,
2006), information on DOMV components does not provide any
clue to the biogenesis and functions of native OMVs in bacterial
communities. Therefore, global proteomic studies of native
OMVs derived from diverse nonpathogenic and pathogenic
bacteria will help us not only elucidate the biogenesis and
functions of OMVs but also develop diagnostics, vaccines, and
antibiotics effective against pathogenic strains.

After an overview of previous biochemical and biological

studies on Gram-negative bacterial OMVs, this review will focus
on current strategies used for proteomic analyses of OMVs,
emphasizing the impact of those studies on this emerging field.
Finally, future directions for high-throughput and comparative
proteomics studies of OMVs from diverse Gram-negative
bacteria under various environmental conditions will be high-
lighted in hopes of advancing both basic and clinical sciences.

II. PROTEIN SECRETION IN
GRAM-NEGATIVE BACTERIA

A. Conventional Protein Secretion Pathways in
Gram-Negative Bacteria

Gram-negative bacteria are enclosed within two lipid bilayers,
consisting of a phospholipid-rich inner membrane and a
phopholipid- and LPS-rich outer membrane. The periplasmic
space separates these membranes and contains peptidoglycans
(Beveridge, 1999). In contrast to nucleated eukaryotic cells,
bacterial cytoplasm is not compartmentalized, no large organ-
elles are present, and no active transport mechanisms such as
molecular motors are known (Howard, Rutenberg, & de Vet,
2001). However, protein secretion is a basic cellular function
found in all living organisms. Gram-negative bacteria have

evolved several secretion pathways, including some mechanisms
common among human and plant pathogens (Lory, 1992).
Approximately 20% of the polypeptides synthesized by bacteria
are located partially or completely outside of the cytoplasm
following secretion (Pugsley, 1993; Kostakioti et al., 2005).

Previously, six major protein secretion pathways in Gram-

negative bacteria were known; they can be classified by the
presence or absence of a required signal sequence, Sec (Table 1).
Sec-dependent pathways include the type II, IV, and V secretion
systems, which utilize cleavable N-terminal signal peptides
for protein transport across the inner membrane (Kostakioti
et al., 2005). The type II secretion system, also known as the
general secretory pathway, is responsible for the secretion of
several toxins and utilizes the Tat signal motif in addition to Sec
(Voulhoux et al., 2001). The type IV secretion system allows the
transfer of DNA and multi-subunit toxins, including pertussis
toxin, by conjugation machinery (Cascales & Christie, 2003).
Depending on the bacterial strain, both Sec-dependent and Sec-
independent secretion have been observed in the type IV system
(Desvaux et al., 2004). Proteins using type V secretion, also
known as the autotransporter pathway, are translocated across
the outer membrane via a transmembrane pore formed by a
self-encoded b-barrel structure (Desvaux, Parham, & Henderson,
2004).

Sec-independent pathways include the type I, III, and

VI secretion systems, which are one-step mechanisms that do not
involve periplasmic intermediates (Kostakioti et al., 2005). In the
type I secretion system, an ATP-binding cassette transporter-like
channel transports various molecules from ions and drugs to
proteins (Binet & Wandersman, 1995). The type III system is
specific for the transport of factors by pathogenic bacteria and
allows the direct injection of a protein into a eukaryotic host cell
(Galan & Collmer, 1999). Recently, the secretion of several
proteins via the type VI pathway was reported in Vibrio cholerae
and P. aeruginosa (Mougous et al., 2006; Pukatzki et al., 2006).

B. OMVs as a Novel Protein Secretion Pathway in
Gram-Negative Bacteria

Protein secretion via OMVs in Gram-negative bacteria has
come of age by defining a distinct type that is independent of
the type I–VI secretory systems (Kuehn & Kesty, 2005). The
dynamic feature of Gram-negative cell wall is that it constantly
discharges OMVs from the cell surface (Fig. 1), which is not
observed in Gram-positive bacteria (Beveridge, 1999). Growing
evidence suggests that hundreds of proteins, lipids, and genetic
material might be secreted via OMVs (Dorward, Garon, & Judd,
1989; Horstman & Kuehn, 2000; Mashburn-Warren & Whiteley,
2006; Lee et al., 2007). For example, export of cytolysin A
(ClyA) into the extracellular milieu, which does not follow the
six known protein secretion mechanisms (Wai et al., 2003), is
mediated by OMVs (Bendtsen et al., 2005). ClyA is a pore-
forming cytotoxin protein expressed by E. coli and some other
enterobacteria. Therefore, shedding of bacterial OMVs is a novel
protein secretion mechanism in Gram-negative bacteria and
this system performs a variety of important ‘‘remote-control’’
functions for bacterial growth and survival, as well as patho-
genesis in hosts.

&

LEE ET AL.

536

Mass Spectrometry Reviews DOI 10.1002/mas

III. GRAM-NEGATIVE BACTERIAL OMVS

A. History of Gram-Negative Bacterial OMVs

The discovery of OMVs as the universal secretory machinery in
Gram-negative bacteria dates back almost 40 years, owing to the
electron microscope (EM). In the 1960s, active research on the
ultrastructure of bacteria revealed the presence of OMVs and
they were named blebs, membranous elements, or channels
(Bladen & Waters, 1963; Bayer & Anderson, 1965). Thin section

EM studies of Gram-negative bacteria in that era suggested
that secretory vesicles constantly emanate from bacteria under
lysine-limited, phosphate-limited, and normal growth conditions
(Knox, Vesk, & Work, 1966; Mergenhagen, Bladen, & Hsu,
1966; Ingram, Cheng, & Costerton, 1973; Lindsay et al., 1973).

Moreover, OMVs have been found in every environment in

which Gram-negative bacteria reside, from laboratory cultures,
including planktonic and surface-attached biofilm, to natural
environments such as domestic water drains, sewage, and
riverbeds (Beveridge, 1999; Schooling & Beveridge, 2006).
The identification of OMVs from a variety of Gram-negative
strains including E. coli, Veillonella, V. cholerae, P. aeruginosa,
Salmonella typhimurium, and N. meningitidis implies that
virtually all Gram-negative bacteria produce OMVs as an active
and essential process (Bladen & Mergenhagen, 1964; Bayer &
Anderson, 1965; Chatterjee & Das, 1967; Devoe & Gilchrist,
1973; Beveridge, 1999).

B. Current Research Trends on Gram-Negative
Bacterial OMVs

Although bacterial OMVs have long been studied, previous
reports have focused primarily on pathogenic strains, and many
questions remain to be answered before attaining an integrated
view of the biogenesis and the pathophysiological functions of
OMVs from both nonpathogenic and pathogenic bacteria (Kuehn
& Kesty, 2005). Recently, biochemical analyses and a few
proteomics applications revealed that bacterial OMVs include
proteins, lipids, and genetic material (Kadurugamuwa &

TABLE 1. Conventional protein secretion pathways in Gram-negative bacteria

FIGURE 1.

Discharge of OMVs by Gram-negative bacteria. (A) Thin-

section EM of E. coli DH5a, showing the formation of OMVs (arrows)
on the cell surface. Bar

¼ 100 nm (B) Magnified EM image of OMVs.

The membrane bilayer (arrow) is easily visible. Bar

¼ 50 nm. [Reprinted

with permission from Proteomics 7: 3143-3153, 2007, Lee EY et al.,
Global proteomic profiling of native outer membrane vesicles derived
from Escherichia coli, with permission from Wiley-VCH Verlag GmbH
& Co. KGaA. Weinheinm, Germany. Copyright 2007.]

PROTEOMICS IN BACTERIAL OUTER MEMBRANE VESICLES

&

Mass Spectrometry Reviews DOI 10.1002/mas

537

Beveridge, 1995; Horstman & Kuehn, 2000, 2002; Post et al.,
2005; Bauman & Kuehn, 2006; Nevot et al., 2006; Lee et al.,
2007). Moreover, genetic studies examining some candidate
genes that modulate the level of vesiculation suggest possible
biogenesis models for bacterial OMVs (Mashburn-Warren &
Whiteley, 2006; McBroom et al., 2006; McBroom & Kuehn,
2007). In addition, the pathophysiological roles of OMVs in
the interspecies world, as well as polymicrobial communities,
are being gradually elucidated. Some groups are studying the
physiological relevance of stress responses, including heat-shock
and antibiotic treatment, which alter the generation of OMVs
(Katsui et al., 1982; Kadurugamuwa & Beveridge, 1995).

In the following subsections, we summarize previous

biochemical and biological research on Gram-negative bacterial
OMVs.

1. Components of Gram-Negative Bacterial OMVs

Although bacterial OMVs were believed to consist of proteins
and lipids from the outer membrane and periplasm but not
from either the inner membrane or cytoplasmic components
(Horstman & Kuehn, 2000), growing evidence suggests that
virulence factors including LPS, cytoplasmic proteins, and
genetic material such as DNA and RNA are components of
OMVs (Dorward, Garon, & Judd, 1989; Kolling & Matthews,
1999; Ferrari et al., 2006; Lee et al., 2007).

The presence of vesicular proteins has been analyzed by

sodium dodecyl sulfate–polyacrylamide gel electrophoresis
(SDS–PAGE) with Coomassie or silver staining, as well as
Western blotting with in-house antibodies (Horstman & Kuehn,
2000; Ferrari et al., 2006). Several outer membrane proteins
including OmpA, OmpC, and OmpF have been identified, which
represent the most abundant proteins; they have been found in all
strains of E. coli studied to date (Kesty et al., 2004). Biochemical
analysis has detected periplasmic proteins such as alkaline
phosphatase and AcrA in OMVs, supporting the hypothesis that
some periplasmic proteins are sorted into OMVs by encapsula-
tion during vesicle formation (Horstman & Kuehn, 2000).

Pathogenic bacteria secrete OMVs that contain several

virulence factors, including toxins, adhesins, invasins, and other
related enzymes (Kesty & Kuehn, 2004; Kesty et al., 2004;
Kuehn & Kesty, 2005). Toxins are the best characterized factors
that might be involved in OMV-mediated bacterial pathogenesis.
For example, LPS, which are critical components of all
Gram-negative bacteria, can activate host immune responses
via production of various cytokines. The large number of
vesicular toxins that contribute to the pathogenesis of infection
was summarized in a recent report (Kuehn & Kesty, 2005). The
ability of OMVs to adhere to and invade host cells via adhesins
and invasins is important in initiating vesicle-mediated patho-
genesis. For example, Ail, IpaB, IpaC, and IpaD play important
roles in interactions with and invasion of host cells (Kaduruga-
muwa & Beveridge, 1998; Kesty & Kuehn, 2004).

Thin layer chromatography revealed the presence of

glycerophospholipids, phosphatidlyethanolamine, phosphatidyl-
glycerol, and cardiolipin in enterotoxigenic E. coli-derived
OMVs (Horstman & Kuehn, 2000). The lipid profiles of OMVs
are similar to those of the outer membrane, but their specific types

and ratios have not been determined. Therefore, systematic
studies on the lipid composition of OMVs that might govern
vesicular fate and function under physiological and pathological
conditions are a challenge for lipidomics.

The presence of DNA within OMVs was identified in N.

gonorrhoeae, Haemophilus influenzae, P. aeruginosa, and E. coli
O157:H7 (Mashburn-Warren & Whiteley, 2006). The fact that
vesicular DNA is resistant to DNase treatment suggests that it is
present in the lumen of OMVs. Therefore, DNA within OMVs is
expected to be protected from nucleases, thereby increasing the
efficiency of vesicle-mediated DNA delivery into a recipient cell.
DNA within OMVs can originate by one of two mechanisms:
DNA exists in the periplasm, and along with other periplasmic
components, becomes encapsulated, or DNA in the extracellular
environment, potentially derived from lysed bacteria, is incorpo-
rated into OMVs by the ‘‘opening and closing’’ phenomenon
(Renelli et al., 2004). RNA is also a vesicular component,
although in lesser amounts than DNA (Dorward, Garon, & Judd,
1989).

While multiple sources of vesicular components obviously

contribute to the biogenesis and functions of OMVs, the exact
vesicular composition of OMVs derived from different bacterial
strains should not be identical. These differences in vesicular
components may define unique physiological and pathological
functions of OMVs derived from specific strains of Gram-
negative bacteria. However, limited data are available on strain-
specific OMV components, a problem that must be solved in the
near future.

2. Biogenesis of Gram-Negative Bacterial OMVs

The mechanisms by which Gram-negative bacteria shed OMVs
and sort vesicle-targeted proteins have not been fully determined.
However, genetic mutant studies, biochemical experiments, and
microscopic observations suggest three plausible models for
OMV formation, as shown in Figure 2 (Mashburn-Warren &
Whiteley, 2006). The first model suggests that vesicles are
generated by the loss of cell envelope integrity that occurs when
the outer membrane expands more quickly than the underlying
peptidoglycan layer (Wensink & Witholt, 1981). The second
model is that the formation of OMVs is linked to the turgor
pressure of the cell envelope, which changes with the
accumulation of peptidoglycan fragments in the periplasm (Zhou
et al., 1998). A recent study in P. aeruginosa proposed the third
model, in which quinolone signal molecules enhance the anionic
repulsion between LPS by destabilizing the Mg

2

þ

and Ca

2

þ

salt

bridges in the outer membrane, thereby causing membrane
blebbing (Mashburn & Whiteley, 2005). These three mechanisms
may not be mutually exclusive and may contribute collectively to
the biogenesis of bacterial OMVs. Further study is needed, since
recent studies suggest that OMV production is independent of
membrane instability (McBroom et al., 2006) and controversial
results have come from the mutation of lipoproteins (Bernadac
et al., 1998; Rolhion et al., 2005).

Currently, little is known about the mechanisms by which

bacteria sort proteins into OMVs. However, analysis of vesicular
proteins by SDS–PAGE have shown different banding patterns in
the OMVs compared to the outer membrane, periplasm, and other

&

LEE ET AL.

538

Mass Spectrometry Reviews DOI 10.1002/mas

cellular fractions, suggesting that specific protein sorting
mechanisms are in effect when OMVs are produced (Fig. 3)
(Horstman & Kuehn, 2000; Lee et al., 2007). Although further
study of this issue is needed, these findings imply the presence of
‘‘hot spots’’ for vesicle budding: the outer membrane becomes
loosely attached to the bacterium at specific sites and forms

OMVs that are shed into the extracellular milieu. Budding of
OMVs from the limiting membrane can be inferred from human
cell-derived microvesicles that are reportedly associated with
specific membrane sites called lipid rafts (Rajendran & Simons,
2005).

3. Physiological and Pathological Functions of
Gram-Negative Bacterial OMVs

Since OMVs bear diverse proteins, LPS, outer membrane lipids,
genetic material, other factors associated with virulence, and
conjugational machinery to target other bacteria or host cells,
they should play diverse roles as intercellular communicasomes
not only in bacterial communities but also in interspecies worlds.
Gram-negative bacterial OMVs might play a role in the transfer
of proteins and genetic material between bacterial cells, in the
elimination of competing organisms, in cell-to-cell signaling and
bacterial survival, and in the delivery of toxins to host cells
(Kuehn & Kesty, 2005; Mashburn-Warren & Whiteley, 2006).

OMVs are involved in the transfer of proteins as well as

genetic material in polymicrobial communities. OMVs derived
from P. aeruginosa showed beneficial effects to their own group
by transferring an antibiotic resistance protein, b-lactamase,
to increase survival (Mashburn-Warren & Whiteley, 2006).
Predatory roles of OMVs have been proposed, in which E. coli-
and P. aeruginosa-derived vesicles can kill competing bacteria by
peptidoglycan degradation or cell lysis via vesicular components
such as murein hydrolases (Kadurugamuwa & Beveridge, 1996;
Li, Clarke, & Beveridge, 1998). Furthermore, OMVs package
chromosomal, plasmid, and phage DNA as well as RNA, which
may increase genetic diversity by transforming neighboring
bacteria (Kuehn & Kesty, 2005).

FIGURE 3.

Proteins present in OMVs derived from E. coli DH5a.

Coomassie blue-stained SDS–PAGE comparison of the proteins from
whole-cell lysates (WC), periplasmic proteins (PP), outer membrane
proteins (OMP), and OMVs showing specific protein sorting into
vesicles. Molecular weight standards are indicated on the left (kDa).
[Reprinted with permission from Proteomics 7: 3143–3153, 2007, Lee
EY et al., Global proteomic profiling of native outer membrane vesicles
derived from Escherichia coli, with permission from Wiley-VCH Verlag
GmbH & Co. KGaA. Copyright 2007.]

FIGURE 2.

Proposed models for biogenesis of Gram-negative bacterial

OMVs. Model 1: OMVs are liberated from specific regions on the cell
surface where peptidoglycan-associated lipoproteins are missing due
to the faster expansion of the outer membrane than the underlying
peptidoglycan layer. Model 2: Accumulation of peptidoglycan frag-
ments in the periplasm causes increased turgor pressure, thereby increas-
ing blebbing of the outer membrane. Model 3: In the P. aeruginosa outer
membrane, PQS sequesters the positive charge of Mg

2

þ

, which results in

enhanced anionic repulsion between LPS molecules and membrane
blebbing. OM, outer membrane; PG, peptidoglycan; IM, inner mem-
brane; PQS, Pseudomonas quinolone signal. [Reprinted with permission
from Molecular Microbiology 61: 839–846, 2006, Mashburn-Warren
and Whiteley, Special delivery: vesicle trafficking in prokaryotes, with
permission from Blackwell Publishing Ltd Oxford, UK. Copyright
2006.]

PROTEOMICS IN BACTERIAL OUTER MEMBRANE VESICLES

&

Mass Spectrometry Reviews DOI 10.1002/mas

539

Bacterial OMVs also play protective roles that contribute to

bacterial survival by reducing toxic compounds and antibiotics,
and by facilitating the release of attacking phages (Loeb &
Kilner, 1978; Kobayashi et al., 2000). Recently, McBroom
and Kuehn (2007) performed a genetic mutant study under
stress conditions by treatment with 10% ethanol or an outer
membrane-damaging antimicrobial peptide and found that over-
vesiculating mutant strains had enhanced survival due to the
release of misfolded proteins via increased shedding of OMVs
(McBroom & Kuehn, 2007).

Some bacteria have coevolved in a symbiotic relationship

with their hosts, although infectious pathogens may cause some
problems in hosts. Therefore, examining the roles of OMVs
in the interspecies community is important to understanding
the mechanisms of symbiosis and pathogenesis. Diverse Gram-
negative pathogens have exploited potent virulence strategies by
vesicle-mediated toxin delivery to host cells (Kuehn & Kesty,
2005). In the case of enterotoxigenic E. coli-derived heat-labile
enterotoxins (LT), most (

95%) are secreted via OMVs (Horst-

man & Kuehn, 2002). Some toxins, including LT and leukotoxin,
are more active when they are associated with vesicles rather than
in a free form (Kuehn & Kesty, 2005). Moreover, LPS and the
outer membrane proteins present in vesicles can activate host
immune responses via Toll-like receptors. These LPS and
surface-localized antigens from pathogenic bacteria-derived
vesicles can cause overstimulated inflammatory responses or
septic shock in hosts (Namork & Brandtzaeg, 2002).

IV. CURRENT PROTEOMICS IN GRAM-NEGATIVE
BACTERIAL OMVS

Although previous biochemical, biological, and genetic studies
help us to understand the vesicular components, biogenesis,
and diverse roles of OMVs in polyspecies communities, the
information provided by those studies does not afford a com-
prehensive understanding of the emerging biology of bacterial
OMVs. MS-based proteomic studies offer a powerful way to
clarify the mechanisms of vesicle formation and the pathophy-
siological roles of OMVs by drawing a global map of diverse
bacterial OMVs.

Proteomic studies have been successfully used to study

whole cellular proteins from diverse bacteria and to study
bacterial adaptation to various stress situations (Washburn &
Yates, 2000; Hecker & Volker, 2004; Bandow & Hecker, 2007).
However, complexity at the whole-cell level has a limited ability
to provide systematic insights into cell biology and often fails
to identify low-abundance proteins, including outer membrane
proteins, which are inevitably masked by high-abundance
proteins (Brunet et al., 2003). Because OMVs may represent
nano-sized extracellular organelles of bacteria, the application
of organelle proteomics to bacterial OMVs will define new
biological processes for interactions among bacterial cells,
symbiosis, and pathogenesis in hosts.

Regardless of the importance of native OMVs, which may

act as key mediators for intercellular communications, previous
proteomic research has focused on DOMVs (Nally et al., 2005;
Ferrari et al., 2006; Uli et al., 2006; Vipond et al., 2006). Since

DOMVs are produced under artificial conditions, DOMVs and
native OMVs may have different protein components. As shown
in Table 2, a few studies have been published regarding the
proteomics of native OMVs. These include OMVs isolated from
pathogenic bacteria such as N. meningitidis and P. aeruginosa
(Post et al., 2005; Bauman & Kuehn, 2006), nonpathogenic
bacteria including Pseudoalteromonas antarctica NF

3

, and

E. coli (Nevot et al., 2006; Lee et al., 2007), and peculiar mutant
forms of N. meningitidis and extraintestinal pathogenic E. coli
(Ferrari et al., 2006; Berlanda Scorza et al., 2007).

In the following subsections, we summarize previous

proteomic studies on native OMVs derived from Gram-negative
bacteria and discuss in detail clues provided by proteomics that
elucidate the biogenesis and functions of bacterial OMVs.

A. Preparation of Gram-Negative Bacterial OMVs

Efficient OMV preparation without any contamination by non-
vesicular components is a critical prerequisite for proteomic
analysis. Generally, bacterial OMVs are isolated from the culture
supernatant using a combination of differential centrifugation
to remove cells and cell debris, and ultracentrifugation to pellet
the OMVs (Wai et al., 2003). However, ultracentrifugation
alone does not discriminate between OMVs and other membrane
debris or large protein aggregates. Filtration of the cell-culture
supernatant through 0.22–0.45 mm filters before ultracentrifu-
gation may reduce contamination (Horstman & Kuehn, 2000;
Ferrari et al., 2006; Berlanda Scorza et al., 2007). Recent progress
in the biology of OMVs shows that density gradient centrifuga-
tion is one of the best separation methods to remove OMVs from
contaminating protein aggregates, pili, and flagella (Bauman &
Kuehn, 2006; Lee et al., 2007). Gel filtration chromatography
is an alternative method to isolate OMVs with high purity.
The beads in a gel filtration chromatography column contain
pores that can fractionate vesicles on the basis of differential
diffusion and size exclusion. Post et al. (2005) purified
OMVs from N. meningitidis using a Sephacryl S500 column
(Post et al., 2005). Gel filtration chromatography is an effective
method for purifying OMVs that are relatively homogeneous
in size.

Recently, we reported the proteomic profile of native OMVs

derived from representative strains of nonpathogenic E. coli
DH5a (Lee et al., 2007). With some modifications of previously
described purification methods (Horstman & Kuehn, 2000; Wai
et al., 2003; Rolhion et al., 2005), we obtained highly pure
OMVs secreted by DH5a cells using two sequential steps
(Fig. 4A). In the first step, OMVs were isolated from the culture
supernatant using a combination of differential centrifugation to
remove cells and cell debris; filtration through a 0.45 mm filter;
pre-centrifugation at 20,000g and 40,000g to remove any large
vesicles, vesicle aggregates, and cell debris; and then ultra-
centrifugation at 150,000g. In the second purification step, the
enriched OMVs were further purified using sucrose density
gradients to remove any remaining contaminants. EM of purified
OMVs revealed that almost all were small, closed vesicles
ranging from 20 to 40 nm in diameter, and no membrane whorls,
fragments of lysed vesicles, large vesicles, or pili were detected
(Fig. 4B).

&

LEE ET AL.

540

Mass Spectrometry Reviews DOI 10.1002/mas

B. Proteomic Analysis of Gram-Negative
Bacterial OMVs

In addition to the purity of OMVs, reduction of proteome
complexity by protein separation is the key step to identifying a
large number of vesicular proteins via proteomic analysis. Two-
dimensional gel electrophoresis (2-DE) is a powerful tool for

protein separation. However, the limitations of this approach
for membrane proteins are well-known. The major obstacle
is poor solubility of membrane proteins in the non-detergent
isoelectric focusing buffer that causes the precipitation of
proteins at their isoelectric points (Wu & Yates, 2003).
Moreover, 2-DE cannot properly resolve high molecular
weight, very basic, or hydrophobic proteins (Wu & Yates,

TABLE 2. Summary of proteomic studies on native bacterial OMVs

FIGURE 4.

Methods of preparation for OMVs derived from E. coli DH5a. A: Procedure for preparing

OMVs from DH5a. B: Negative-staining transmission EM of purified OMVs after sucrose density gradient
centrifugation, showing a homogeneous size of 20–40 nm. Bar

¼ 50 nm. [Reprinted with permission from

Proteomics 7: 3143–3153, 2007, Lee EY et al., Global proteomic profiling of native outer membrane
vesicles derived from Escherichia coli, with permission from Wiley-VCH Verlag GmbH & Co. KGaA.
Copyright 2007.]

PROTEOMICS IN BACTERIAL OUTER MEMBRANE VESICLES

&

Mass Spectrometry Reviews DOI 10.1002/mas

541

2003). The fact that outer membrane proteins, the major
components of OMVs, are highly basic implies that major
components of vesicular proteins are incompletely resolved
in 2-DE (Post et al., 2005).

The combination of one-dimensional (1-D) SDS –PAGE

and liquid chromatography (LC)– MS/MS provides a powerful
alternative to 2-DE-based proteomic analysis (Aebersold &
Mann, 2003). Although 1-D SDS– PAGE can efficiently
separate proteins, even membrane proteins, the limitation of
this approach in high-throughput mass analysis is the increased
protein complexity in each gel fraction. This problem can
easily be overcome by using LC to separate the extracted
peptides based on hydrophobicity. Several groups have used this
strategy (Post et al., 2005; Nevot et al., 2006), but they only
examined prominent protein bands of interest from gels, which
might result in the identification of less than 50 vesicular
proteins because of missing the less abundant and unknown
proteins (Table 2). Because the molecular weights of vesicular
proteins are different and OMVs also contain less abundant
proteins, our group separated vesicular proteins by 1-D SDS–
PAGE, cut the gel into five slices of equal size, and subjected it
to trypsin digestion. From two independent nano-LC electro-
spray ionization (ESI)– MS/MS analyses of the extracted
peptides, we identified 2,606 and 2,816 proteins with high-
confidence peptide sequences, with an error rate less than 1% (F
score > 2.17). Since peptides that are shared by multiple
proteins are less informative than unique peptides, we used
the protein hit score (PHS) for reliable protein identification
(Park et al., 2006). Our analysis showed that proteins with
PHS > 1 were identified by multiple peptides that are unique
and shared with only a few proteins. Using a highly stringent
filter allowing only proteins with PHS > 1 that were filtered
again to reduce any repeated or homologous proteins, we finally
identified a total of 141 proteins, including 127 previously
unknown vesicular proteins, with high confidence, and reprodu-
cibility (Lee et al., 2007).

C. Proteins Identified by Proteomic Analyses of
Gram-Negative Bacterial OMVs

The available proteomic studies on bacterial OMVs have defined
more than 200 vesicular proteins from four native bacterial and
two mutant strains (Table 2) (Post et al., 2005; Bauman &
Kuehn, 2006; Ferrari et al., 2006; Nevot et al., 2006; Berlanda
Scorza et al., 2007; Lee et al., 2007). Although the names of
bacterial proteins are different in each species, they can be
classified into protein families based on both their sequence
homology and function. When the identified vesicular proteins
were categorized by protein family, several protein families
were common in OMVs derived from several species of Gram-
negative bacteria (Table 3). Porins (Omps, PorA, PorB, and
OprF), abundant outer membrane proteins, are found in most
OMVs. Murein hydrolases (Mlt and SLT) are responsible for the
hydrolysis of certain cell wall glycopeptides, particularly
peptidoglycans. Multidrug efflux pumps (Mtr, Mex, and TolC)
function in the release of toxic compounds (Kobayashi et al.,
2000). Moreover, most OMVs derived from different strains
contain ABC transporters (LamB and FadL), protease/chaper-

one proteins (DegQ/SurA), and motility proteins related to
fimbriae (FliC) or pilus (PilQ). These conserved vesicular
proteins provide an integrated view of the biogenesis and
function of OMVs in nonpathogenic and pathogenic bacteria,
which will be discussed in the following subsections. For
pathogenic strains, virulence factors including hemolysin, IgA
protease, and macrophage infectivity potentiator were also
identified (Post et al., 2005; Ferrari et al., 2006).

Many researchers believe that OMVs are composed solely of

outer membrane and periplasmic proteins, whereas cytoplasmic
proteins are excluded (Horstman & Kuehn, 2000). However,
although it is still debated, proteomic analyses have shown that
native OMVs and DOMVs contain cytoplasmic proteins as well
(Molloy et al., 2000; Henry et al., 2004; Ferrari et al., 2006; Wei
et al., 2006; Xu et al., 2006; Lee et al., 2007). Among vesicle-
associated cytoplasmic proteins, highly abundant proteins like
EF-Tu, GroEL, DnaK, and two ribosomal proteins (S1 and
L7/12) have also been detected from cell supernatants or outer
membrane fractions (Ferrari et al., 2006). Moreover, the fact that
vesicles carry DNA and RNA, and that translation of outer
membrane proteins might occur simultaneously with their
integration into the membrane, suggest that transcriptional and
ribosomal proteins can be sorted into vesicles during the
informational process (Dorward, Garon, & Judd, 1989; Kadur-
ugamuwa & Beveridge, 1995; Kolling & Matthews, 1999; Yaron
et al., 2000). Determining whether cytoplasmic proteins are
indeed components of native OMVs should be a goal of future
studies.

D. Proteins Involved in Biogenesis of Gram-Negative
Bacterial OMVs

Although the mechanism of OMV formation has not yet been
elucidated, several vesicular proteins identified by proteomic
analyses support the first and second models (Fig. 2). Omps, Tol-
Pal, YbgF, and Lpp lipoproteins found in the OMV proteome
should be involved in outer membrane integrity and might
help liberate OMVs from the bacterial cell surface by initiating
faster expansion of the outer membrane than the underlying
peptidoglycan layer (Bernadac et al., 1998). Related to
the second model, murein hydrolases, including MltA, MipA,
MltE, and SLP, may lead to the accumulation of peptidoglycan
fragments in the periplasmic space, resulting in increased turgor
pressure and causing the discharge of OMVs (Lommatzsch
et al., 1997).

When OMV proteomes are annotated according to their

subcellular distribution, OMVs are highly enriched in outer
membrane and periplasmic proteins, whereas inner membrane
proteins are excluded (Post et al., 2005; Lee et al., 2007).
For example, of 141 proteins identified in E. coli-derived OMVs,
65 (46.1%), 16 (11.3%), 7 (5.0%), 52 (36.9%), and 1 (0.7%)
proteins were derived from the outer membrane, periplasm, inner
membrane, cytoplasm, and extracellular space, respectively
(Fig. 5) (Lee et al., 2007). In contrast, from the EchoBASE
database of all 4,345 E. coli proteins, 149 (3.4%), 350 (8.1%),
974 (22.4%), 2,862 (65.9%), and 10 (0.2%) are distributed in
the outer membrane, periplasm, inner membrane, cytoplasm, and
extracellular space, respectively, suggesting that outer membrane

&

LEE ET AL.

542

Mass Spectrometry Reviews DOI 10.1002/mas

TABLE 3. Protein families identified by proteomic analyses of Gram-negative bacterial OMVs

(Continued )

PROTEOMICS IN BACTERIAL OUTER MEMBRANE VESICLES

&

Mass Spectrometry Reviews DOI 10.1002/mas

543

TABLE 3. (Continued )

(Continued )

&

LEE ET AL.

544

Mass Spectrometry Reviews DOI 10.1002/mas

and periplasmic proteins are more commonly sorted into OMVs
(Misra et al., 2005). Moreover, the inclusion of particular proteins
in OMVs does not appear to be a strict function of their
abundance. As shown in Table 4, several low-abundance outer
membrane proteins, including FimD, FecA, FhuE, and FepA,
and periplasmic proteins, including YddB, SLT, MalM, and
PRC, were identified, whereas the most abundantly expressed
periplasmic proteins, such as OppA, FimA, HdeA, and LivJ, were
not (Corbin et al., 2003). These results further support the

hypothesis that special sorting mechanisms are in effect and/or
the OMVs bud at specific vesiculation sites.

E. Proteins Involved in Biological Functions of
Gram-Negative Bacterial OMVs

In addition to supporting previously known biological roles of
OMVs, vesicular proteins identified in proteomic studies suggest

TABLE 3. (Continued )

a

Accession numbers of individual proteins originate from each reference.

PROTEOMICS IN BACTERIAL OUTER MEMBRANE VESICLES

&

Mass Spectrometry Reviews DOI 10.1002/mas

545

novel functions, and we can specify what proteins are involved
in each physiological and pathological function. As shown in
Figure 6, protein function can be largely classified by the
interacting partner, including bacteria and host cells. First, the
organic solvent tolerance protein (OstA), multidrug-resistant
efflux pumps (Mex, Mtr, and TolC), and phage target receptors
(FepA, LamB, and OmpA) in E. coli, N. meningitidis, and
P. antarctica NF

3

-derived OMVs may contribute to bacterial

survival by reducing levels of toxic compounds such as n-hexane
and antibiotics, and by facilitating the release of attacking phages
(Post et al., 2005; Nevot et al., 2006; Lee et al., 2007). ABC
transporters for specific nutrients (LamB, BtuB, and FadL),

inorganic ions (FepA, FhuA, and Fiu), and nucleosides (Tsx)
exert their roles as delivery systems in bacterial communities.
In particular, TonB-dependent receptors (BtuB, FhuA, and FhuE)
in OMVs have been suggested to be nutrient sensors and
transporters, and their presence has been postulated to represent
an alternative mechanism for survival in nutrient-limited systems
(Nevot et al., 2006). Furthermore, murein hydrolases (e.g., MltA,
SLT) in OMVs are involved not only in OMV biogenesis as
described above, but also in predatory activities by killing
competing bacteria via cell wall degradation (Kadurugamuwa &
Beveridge, 1996; Li, Clarke, & Beveridge, 1998).

In the host environment, effective pathological functions of

OMVs can be achieved by increased resistance to bactericidal
factors. OmpT in vesicles may degrade cationic antimicrobial
peptides produced by epithelial cells or macrophages, and Iss
may increase serum survival of OMVs (Stumpe et al., 1998;
Nolan et al., 2003). Moreover, in addition to pathogenic-specific
toxins, outer membrane porin proteins, including OmpA and
OmpF, which are enriched in OMVs, have immunostimulatory
activity and induce leuckocyte migration (Galdiero et al., 1999).
Pathogenic-specific adherent/invasive proteins and outer mem-
brane porins, including OmpA, OmpW, and OmpX, are involved
in targeting OMVs to host cells. Furthermore, OmpA enhances
the uptake of LPS by macrophages and contributes to the invasion
of brain microvascular endothelial cells (Korn et al., 1995;
Prasadarao et al., 1996).

V. FUTURE DIRECTIONS FOR PROTEOMIC
STUDIES IN GRAM-NEGATIVE BACTERIAL OMVS

Previous proteomic studies have established a proteome database
of OMVs derived from four native and two mutant bacteria.
These studies have provided important information about the bio-
genesis, pathophysiological functions, and protein composition

TABLE 4. Sorting profiles of vesicular proteins in E. coli DH5a

OMP, outer membrane protein; PP, periplasmic protein.

a

Abundance represents mRNA signal intensity on GeneChip.

FIGURE 5.

Subcellular distribution of vesicular and cellular proteins

present in E. coli. When compared to the complete E. coli proteome,
DH5a-derived OMVs are highly enriched in outer membrane proteins,
whereas inner membrane proteins are excluded. [Adapted from Lee
et al. (2007).]

&

LEE ET AL.

546

Mass Spectrometry Reviews DOI 10.1002/mas

of bacterial OMVs (Post et al., 2005; Bauman & Kuehn, 2006;
Ferrari et al., 2006; Nevot et al., 2006; Berlanda Scorza et al.,
2007; Lee et al., 2007). However, compared to human micro-
vesicles, for which significant progress has been made in defining
the proteomes of vesicular components (Yates et al., 2005),
proteomic information on bacterial OMVs still remains scarce
except for E. coli (Berlanda Scorza et al., 2007; Lee et al., 2007).
Limited proteomic data make it difficult to elucidate the exact
mechanism of vesicle formation and new functions of OMVs.
Further studies are therefore necessary to construct a vesicular
proteome for diverse Gram-negative bacteria under a variety of
conditions. Overall, the future directions for proteomic analyses
on Gram-negative bacterial OMVs are summarized in Figure 7.

A. OMVs Derived from Diverse
Gram-Negative Bacteria

Diverse Gram-negative bacteria provide unlimited material for
OMVs, as most have been reported to secrete vesicles. Valuable
samples can be obtained from host fluids and tissues (Table 5).
The facts that Gram-negative bacteria are the main causes of
septic shock, and that LPS and outer membrane proteins of
bacterial OMVs elicit a complex pattern of inflammatory

reactions suggest that bacterial OMVs are involved in the
progress of septic shock. Therefore, proteomic analysis of OMVs
obtained from the serum of septic human patients or septic rats
should increase our understanding about the pathological roles of
OMVs in septic shock (Brandtzaeg et al., 1992; Hellman et al.,
2000). Moreover, vesicles shed by N. meningitidis have been
found in the cerebrospinal fluid and blood of a patient with
meningitis (Stephens et al., 1982; Brandtzaeg et al., 1992;
Namork & Brandtzaeg, 2002). In the case of Borrelia
burgdorferi-infected mice, OMVs were detected in the urine
and blood (Dorward, Schwan, & Garon, 1991). Furthermore,
OMVs derived from Bacteroides (Porphyromonas) gingivalis,
which causes oral cavities, inflammation, and bleeding at
peritonitis sites, can be obtained from dental plaque samples
(Grenier & Mayrand, 1987; Imamura et al., 1995).

B. Preparation of Gram-Negative Bacterial OMVs

As noted above, OMVs are usually prepared by ultracentrifuga-
tion followed by density gradient centrifugation or gel filtration
(Horstman & Kuehn, 2000; Wai et al., 2003; Post et al., 2005; Lee
et al., 2007). Other methods such as free-flow electrophoresis
(FFE) and capillary electrophoresis (CE) can be used to isolate

FIGURE 6.

Proposed physiological and pathological functions of Gram-negative bacterial OMVs.

Functions of Gram-negative bacterial OMVs are predicted based on the available proteomes of OMVs
derived from nonpathogenic and pathogenic bacteria. [Color figure can be viewed in the online issue, which
is available at www.interscience.wiley.com.]

PROTEOMICS IN BACTERIAL OUTER MEMBRANE VESICLES

&

Mass Spectrometry Reviews DOI 10.1002/mas

547

OMVs in intact form and with high purity. FFE is a powerful tool
to separate human cellular organelles such as peroxisomes,
lysosomes, endosomes, melanosomes, and Golgi vesicles, as
well as mitochondria (Morre, Morre, & Heidrich, 1983; Fuchs,
Male, & Mellman, 1989; Marsh, 1989; Kushimoto et al., 2001;
Mohr & Volkl, 2002; Zischka et al., 2006). This system allows
purification of bacterial OMVs in a native state and with high

purity by discriminating similar-density membrane fragments,
which are difficult to remove solely by density gradient
ultracentrifugation.

CE is an analytical technique that employs narrow

capillaries for electric field-mediated separation of particles with
a surface charge. Recently, CE was used to separate human
organelles including mitochondria, acidic organelles, nuclei, and

FIGURE 7.

Future directions for high-throughput and comparative proteomics in OMVs originating from

diverse Gram-negative bacteria under various environmental conditions.

&

LEE ET AL.

548

Mass Spectrometry Reviews DOI 10.1002/mas

lipid vesicles (Gunasekera, Musier-Forsyth, & Arriaga, 2002;
Duffy et al., 2002; Fuller & Arriaga, 2003; Owen, Strasters, &
Breyer, 2005). The electrophoretic mobility of organelles is
determined by the electrical charge on the surface of organelles,
their morphology, and size (Owen, Strasters, & Breyer, 2005).
Using this technique, a very small number of organelles can be
separated electrokinetically or hydrodynamically into a capillary
(Fuller & Arriaga, 2004). Therefore, CE should be used to
prepare OMVs with high purity.

C. Strategies for the Proteomic Analysis of
Gram-Negative Bacterial OMVs

1. Protein Identification

For a comprehensive proteomic analysis of OMVs derived from
various Gram-negative bacteria, extensive prefractionation of
samples on the protein and/or peptide levels before mass analysis
should be considered. Although 2-DE is a powerful tool for
protein separation, this method cannot properly resolve high-
molecular-weight, very basic, or hydrophobic proteins (Wu &
Yates, 2003), suggesting that protein separation by 2-DE may
be not suitable for a global proteomic analysis of OMVs.
Combination of 1-D SDS–PAGE and LC–MS/MS provides a
powerful alternative to 2-DE-based proteomic analysis, as noted
above (Aebersold & Mann, 2003). Alternatively, multidimen-
sional protein identification technology (MudPIT) (Washburn,
Wolters, & Yates, 2001) should be useful for proteome analysis of
OMVs. Because MudPIT separates peptides using a combination
of two different kinds of LC prior to MS analysis, this system
greatly reduces the complexity of the proteome at the peptide
level, resulting in the identification a large number of proteins.
Using this process, 1,484 proteins were identified from
Saccharomyces cerevisiae (Washburn, Wolters, & Yates, 2001).
Furthermore, MudPIT is suitable for identifying proteins with

extreme pI, integral membrane proteins, and low-abundance
proteins (Graham, Graham, & McMullan, 2007).

Furthermore, a combination of matrix-assisted laser

desorption/ionization time-of-flight (MALDI-TOF) and ESI
mass spectrometer can also increase coverage and the number
of identified proteins. ESI-MS/MS and MALDI-TOF-MS
analyses of the same sample usually identify different sets of
proteins (Bodnar et al., 2003). Therefore, high proteome
coverage for OMVs can be achieved by the combination of
multidimensional protein and/or peptide fractionation methods,
as well as by the combination of various MS methods.

2. Bioinformatics for Annotation of Vesicular Proteins

To clarify how bacteria shed vesicles and identify the
physiological and pathological functions of OMVs, annotation
of identified vesicular proteins based on subcellular localization
and function is important. Previous proteomic analyses of OMVs
showed that vesicular proteins are derived from various
subcellular locations in bacteria (Post et al., 2005; Ferrari et al.,
2006; Lee et al., 2007). Annotation of vesicular proteins by
subcellular localization in bacteria can help us elucidate the
mechanism of OMV biogenesis, as well as identify possible
contaminants in proteomic analyses. Currently, several bio-
informatics tools are available for bacterial protein localization
including PSORTb (Gardy et al., 2005), EchoBASE (Misra et al.,
2005), Proteome Analyst (Lu et al., 2004), and SubLoc (Hua &
Sun, 2001). They predict the localization of bacterial proteins on
the basis of known motifs or cleavage sites. Furthermore, the
computational prediction of subcellular localization of vesicular
proteins offers numerous insights that, for example, can assist
in functional analyses of OMVs.

For the functional analysis of vesicular proteins, several

bioinformatics tools are available (Ouzounis et al., 2003). One
of the most influential classification schemes comes from a
hierarchy of properties for the gene products of E. coli, which was

TABLE 5. OMVs found in host fluids and tissues

PROTEOMICS IN BACTERIAL OUTER MEMBRANE VESICLES

&

Mass Spectrometry Reviews DOI 10.1002/mas

549

later extended to develop multifunctional classes (Riley, 1993;
Serres & Riley, 2000). Inspired by this classification, the
automatic genome-annotation system GeneQuiz was developed
based on the keyword mapping of protein families to 14
functional classes (Andrade et al., 1999). Gene ontology
classification, which comprises the three categories of molecular
function, biological process, and cellular component, is com-
monly used for protein annotation (Ashburner et al., 2000). The
Kyoto Encyclopedia of Genes and Genomes uses a different
method of classifying genes and proteins by their participation
in or association with metabolic pathways (Kanehisa et al.,
2002). Since each data mining method uses a different strategy to
predict the function of bacterial proteins, a combination of
bioinformatics tools should be used to annotate the identified
vesicular proteins to elucidate the numerous physiological and
pathological functions of Gram-negative bacterial OMVs.

D. Comparative Proteomics in Gram-Negative
Bacterial OMVs

In contrast to the static nature of the genome sequence, which
provides the blueprint for all protein-based cellular building
blocks, the proteome is highly dynamic (Bandow & Hecker,
2007). The protein composition of bacteria is constantly adjusted
to facilitate survival, growth, and reproduction in an ever-
changing environment. Bacteria face highly variable growth
conditions and stress situations with respect to temperature, pH,
osmolarity, nutrient availability, and host infection, among other
factors. Like the whole bacteria proteome, components of OMVs
are influenced by environmental stress and bacterial status
(Katsui et al., 1982; Horstman & Kuehn, 2002; McBroom &
Kuehn, 2007). Some vesicular proteins are temporarily expressed,
whereas other proteins are expressed after signal transduction
from the extracellular milieu, or at certain bacterial growth
phases (Kuehn & Kesty, 2005). Moreover, the release of OMVs
increases when bacteria are exposed to conditions such as
high temperature, exposure to antibiotics or serum, or nutrient
deprivation (Post et al., 2005). Therefore, comparative proteomic
analyses of OMVs derived from diverse conditions are expected
to provide new functional insights into bacterial OMV biology,
facilitate the identification of pathogenic markers, and contribute
to the discovery of proteins as therapeutic targets.

Three approaches have been most commonly used to

generate quantitative profiles of complex protein mixtures
(Zhang, Yan, & Aebersold, 2004). The first is a combination of
2-DE and MS (Aebersold & Mann, 2003). Classical 2-DE is still
the method of choice for quantitative analysis of the proteome,
and difference gel electrophoresis (DIGE) is a commonly used
comparative 2-DE technique (Marouga, David, & Hawkins,
2005). In DIGE, proteins from different samples are labeled with
different fluorescent dyes, mixed equally, and resolved by 2-DE.
The protein samples are visualized using fluorescence imaging to
enable detection of differences in protein abundance between
samples, and the proteins that differ in abundance can be
identified by MS (Marouga, David, & Hawkins, 2005). Using
DIGE, comparative proteomic studies were carried out on
DOMVs and OMVs derived from N. meningitidis (Ferrari
et al., 2006). However, in spite of the progress in 2-DE

technology, it is still not technically feasible to obtain a complete
expression profile from a single two-dimensional gel because
some proteins are not well separated, such as those that
are extremely basic or acidic, small or large, or of low abundance
(Wu & Yates, 2003).

The second method for comparative proteomics is based on

stable isotope tagging of proteins and automated LC–MS/MS
analysis of peptides derived from complex protein mixtures
(Conrads et al., 2001). Isotope-coded affinity tags (ICATs) and
isobaric tags for relative and absolute quantification (iTRAQ) are
commonly used chemical isotopic labeling strategies that
differentially label sulfhydryls or primary amines of proteins or
peptides, respectively (Gygi et al., 2002; Choe et al., 2005). Using
ICATs, several comparative analyses of bacterial proteomes have
been reported, including the P. aeruginosa proteome during
anaerobic growth (Peng et al., 2005) and magnesium-limited
conditions (Guina et al., 2003). The major disadvantage of this
technique is that cysteine is a relatively rare amino acid that is not
present in 10–20% of bacterial proteins (Cordwell, 2006). Since
all proteins and peptides have primary amines on their N-terminal
amino acids or lysine, iTRAQ should be better for comparative
proteomics than using ICATs (Danielsen et al., 2007).

Another widely applicable technique is stable-isotope

labeling of amino acids in cell culture (SILAC) (Mann, 2006;
Ong et al., 2002). In the SILAC procedure, cells are grown in the
presence of an isotopically heavy amino acid for several
generations, thereby replacing essentially all of the naturally
occurring light amino acid in all proteins (Gingras et al., 2007).
The advantage of SILAC over ICATs and iTRAQ is that SILAC
can efficiently label all proteins present in complex samples,
whereas the labeling efficiency of ICATs and ITRAQ may depend
on several factors, including the status of proteins (i.e., native vs.
denatured) (Gruhler et al., 2005). SILAC has been successfully
used in several recent comparative proteomic studies, including
profiling of the dynamic association of chaperonin-dependent
protein folding in E. coli (Kerner et al., 2005).

VI. APPLICATIONS OF PROTEOMIC STUDIES ON
GRAM-NEGATIVE BACTERIAL OMVS

As described above, high-throughput and comparative proteo-
mics on bacterial OMVs will decipher the mechanisms under-
lying intercellular communication by elucidating the biogenesis
and specialized functions of bacterial OMVs. In addition to
improving our understanding of the basic biology of bacterial
OMVs, knowledge about the vesicular proteome can facilitate a
variety of biotechnology applications of OMVs, especially in
human medical research, by identifying specific biomarkers that
can be used to develop diagnostic and therapeutic tools against
pathogenic organisms. Because bacterial OMVs can be dissemi-
nated from infectious sites and circulated in fluids within the host
(Brandtzaeg et al., 1992; Hellman et al., 2000), characterization
of vesicles from patients with bacterial-associated disease
symptoms is an effective way to diagnose pathogenesis.

Moreover, particular interest has emerged in the use of

OMVs as vehicles for stimulation of host immune responses.
These studies have already led to clinical trials with DOMVs,

&

LEE ET AL.

550

Mass Spectrometry Reviews DOI 10.1002/mas

which are made artificially from bacterial cell membranes (de
Moraes et al., 1992; Drabick et al., 1999; Girard et al., 2006).
However, based on previous proteomic reports, the ratio of outer
membrane proteins present in DOMVs was relatively low
compared to native OMVs and was not consistent from batch
to batch, resulting in different and diverse immune responses in
the host depending on the manufacturer (Ferrari et al., 2006).
Therefore, vaccination strategies should also be envisioned
using native OMVs, which carry the enriched and native
topology of strain-specific outer membrane antigens. Creating
super-blebbing bacterial mutants that can produce several times
more vesicles than wild-type strains and genetic mutants that can
control LPS amounts might reduce the side effects of septic
symptoms in patients and will provide a particularly useful
source for vaccine materials (van der Ley et al., 2001; Berlanda
Scorza et al., 2007).

Another important application of OMVs comes in the area of

antibiotics. Adverse reactions in patients undergoing therapy,
resulting from antibiotic-induced liberation of bacterial compo-
nents, including OMVs, have been a long-standing concern
(Kadurugamuwa & Beveridge, 1995; Morand & Muhlemann,
2007). Moreover, treatment with broad-spectrum antibiotics can
disturb natural and beneficial microflora, leaving the patient more
susceptible to infection by opportunistic pathogens. Therefore,
modulating the production of OMVs presents an attractive
approach for treating bacteria-associated diseases.

VII. CONCLUDING REMARKS

Growing evidence suggests that Gram-negative bacterial OMVs
are essential for bacterial survival and pathogenesis in hosts by
acting as intercellular communicasomes in polyspecies com-
munities. In spite of recent progress in this emerging field,
previous biochemical and biological studies are limited in their
ability to provide comprehensive information for understanding
the mechanisms of vesicle formation and the pathophysiological
roles of OMVs. In addition to previous proteomic studies, further
high-throughput and comparative proteomics studies of OMVs
originating from diverse Gram-negative bacteria under various
environmental conditions will take us one step closer to an
integrated view of bacterial OMVs with regard to their biogenesis
and pathophysiological functions. Furthermore, these studies
will stimulate the development of diagnostic tools, novel
vaccines, and antibiotics effective against clinically important
Gram-negative bacteria. We hope this review encourages further
studies on proteomics in Gram-negative bacterial OMVs.

VIII. ABBREVIATIONS

1-D

one-dimensional

2-DE

two dimensional gel electrophoresis

CE

capillary electrophoresis

ClyA

cytolysin A

DIGE

difference gel electrophoresis

DOMV

detergent-extracted outer membrane vesicle

EM

electron microscope

ESI

electrospray ionization

FFE

free-flow electrophoresis

ICAT

isotope-coded affinity tags

iTRAQ

isobaric tags for relative and absolute
quantification

LC

liquid chromatography

LPS

lipopolysaccharide

LT

heat-labile enterotoxins

MALDI-TOF

matrix-assisted laser desorption/ionization
time-of-flight

MS

mass spectrometry

MudPIT

multidimensional protein identification
technology

MV

membrane vesicle

OMV

outer membrane vesicle

SILAC

stable-isotope labeling by amino acids in cell
culture

SDS–PAGE

sodium dodecyl sulfate–polyacrylamide gel
electrophoresis

ACKNOWLEDGMENTS

This work was supported by a grant of the National R&D

Program for Cancer Control, Ministry of Health & Welfare,
Republic of Korea (0320380-2), supported by the Korea Basic
Science Institute K-MeP (T27021), and supported by the Korea
Science and Engineering Foundation (KOSEF) grant funded by
the Korea government (MOST, No. R15-2004-033-05001-0) to
Yong Song Gho. Eun-Young Lee and Dong-Sic Choi were
recipients of Brain Korea 21 fellowship.

REFERENCES

Admyre C, Johansson SM, Qazi KR, Filen JJ, Lahesmaa R, Norman M, Neve

EP, Scheynius A, Gabrielsson S. 2007. Exosomes with immune
modulatory features are present in human breast milk. J Immunol
179:1969–1978.

Aebersold R, Mann M. 2003. Mass spectrometry-based proteomics. Nature

422:198–207.

Andrade MA, Brown NP, Leroy C, Hoersch S, de Daruvar A, Reich C,

Franchini A, Tamames J, Valencia A, Ouzounis C, Sander C. 1999.
Automated genome sequence analysis and annotation. Bioinformatics
15:391–412.

Ashburner M, Ball CA, Blake JA, Botstein D, Butler H, Cherry JM, Davis AP,

Dolinski K, Dwight SS, Eppig JT, Harris MA, Hill DP, Issel-Tarver L,
Kasarskis A, Lewis S, Matese JC, Richardson JE, Ringwald M, Rubin
GM, Sherlock G. 2000. Gene ontology: Tool for the unification of
biology. The Gene Ontology Consortium. Nat Genet 25:25–29.

Bandow JE, Hecker M. 2007. Proteomic profiling of cellular stresses in

Bacillus subtilis reveals cellular networks and assists in elucidating
antibiotic mechanisms of action. Prog Drug Res 64:81–101.

Bauman SJ, Kuehn MJ. 2006. Purification of outer membrane vesicles from

Pseudomonas aeruginosa and their activation of an IL-8 response.
Microbes Infect 8:2400–2408.

Bayer ME, Anderson TF. 1965. The surface structure of Escherichia coli.

Proc Natl Acad Sci USA 54:1592–1599.

PROTEOMICS IN BACTERIAL OUTER MEMBRANE VESICLES

&

Mass Spectrometry Reviews DOI 10.1002/mas

551

Bendtsen JD, Kiemer L, Fausboll A, Brunak S. 2005. Non-classical

protein secretion in bacteria. BMC Microbiol 5:58.

Berlanda Scorza F, Doro F, Rodrı´guez-Ortega MJ, Stella M, Liberatori S,

Taddei AR, Serino L, Gomes Moriel D, Nesta B, Fontana MR,
Spagnuolo A, Pizza M, Norais N, Grandi G. 2008. Proteomic
characterization of outer membrane vesicles from the extraintestinal
pathogenic Escherichia coli DtolR IHE3034 mutant. Mol Cell
Proteomics 7:473–485.

Bernadac A, Gavioli M, Lazzaroni JC, Raina S, Lloubes R. 1998. Escherichia

coli tol-pal mutants form outer membrane vesicles. J Bacteriol 180:
4872–4878.

Beveridge TJ. 1999. Structures of gram-negative cell walls and their derived

membrane vesicles. J Bacteriol 181:4725–4733.

Binet R, Wandersman C. 1995. Protein secretion by hybrid bacterial ABC-

transporters: Specific functions of the membrane ATPase and the
membrane fusion protein. EMBO J 14:2298–2306.

Bladen HA, Mergenhagen SE. 1964. Ultrastructure of Veillonella and

morphological correlation of an outer membrane with particles
associated with endotoxic activity. J Bacteriol 88:1482–1492.

Bladen HA, Waters JF. 1963. Electron microscopic study of some strains of

bacteroides. J Bacteriol 86:1339–1344.

Bodnar WM, Blackburn RK, Krise JM, Moseley MA. 2003. Exploiting the

complementary nature of LC/MALDI/MS/MS and LC/ESI/MS/MS
for increased proteome coverage. J Am Soc Mass Spectrom 14:971–
979.

Brandtzaeg P, Bryn K, Kierulf P, Ovstebo R, Namork E, Aase B, Jantzen E.

1992. Meningococcal endotoxin in lethal septic shock plasma studied
by gas chromatography, mass-spectrometry, ultracentrifugation, and
electron microscopy. J Clin Invest 89:816–823.

Brunet S, Thibault P, Gagnon E, Kearney P, Bergeron JJ, Desjardins M. 2003.

Organelle proteomics: Looking at less to see more. Trends Cell Biol
13:629–638.

Cascales E, Christie PJ. 2003. The versatile bacterial type IV secretion

systems. Nat Rev Microbiol 1:137–149.

Chatterjee SN, Das J. 1967. Electron microscopic observations on the excretion

of cell-wall material by Vibrio cholerae. J Gen Microbiol 49:1–11.

Choe LH, Aggarwal K, Franck Z, Lee KH. 2005. A comparison of the

consistency of proteome quantitation using two-dimensional electro-
phoresis and shotgun isobaric tagging in Escherichia coli cells.
Electrophoresis 26:2437–2449.

Choi DS, Lee JM, Park GW, Lim HW, Bang JY, Kim YK, Kwon KH, Kwon

HJ, Kim KP, Gho YS. 2007. Proteomic analysis of microvesicles
derived from human colorectal cancer cells. J Proteome Res 6:4646–
4655.

Conrads TP, Alving K, Veenstra TD, Belov ME, Anderson GA, Anderson DJ,

Lipton MS, Pasa-Tolic L, Udseth HR, Chrisler WB, Thrall BD, Smith
RD. 2001. Quantitative analysis of bacterial and mammalian proteomes
using a combination of cysteine affinity tags and 15N-metabolic
labeling. Anal Chem 73:2132–2139.

Corbin RW, Paliy O, Yang F, Shabanowitz J, Platt M, Lyons CE Jr, Root K,

McAuliffe J, Jordan MI, Kustu S, Soupene E, Hunt DF. 2003. Toward a
protein profile of Escherichia coli: Comparison to its transcription
profile. Proc Natl Acad Sci USA 100:9232–9237.

Cordwell SJ. 2006. Technologies for bacterial surface proteomics. Curr Opin

Microbiol 9:320–329.

Danielsen M, Hornshoj H, Siggers RH, Jensen BB, van Kessel AG, Bendixen

E. 2007. Effects of bacterial colonization on the porcine intestinal
proteome. J Proteome Res 6:2596–2604.

de Moraes JC, Perkins BA, Camargo MC, Hidalgo NT, Barbosa HA, Sacchi

CT, Landgraf IM, Gattas VL, Vasconcelos Hde G, et al. 1992. Protective
efficacy of a serogroup B meningococcal vaccine in Sao Paulo, Brazil.
Lancet 340:1074–1078.

Desvaux M, Parham NJ, Henderson IR. 2004. The autotransporter secretion

system. Res Microbiol 155:53–60.

Desvaux M, Parham NJ, Scott-Tucker A, Henderson IR. 2004. The

general secretory pathway: A general misnomer? Trends Microbiol
12:306–309.

Devoe IW, Gilchrist JE. 1973. Release of endotoxin in the form of cell wall

blebs during in vitro growth of Neisseria meningitidis. J Exp Med
138:1156–1167.

Dorward DW, Garon CF, Judd RC. 1989. Export and intercellular transfer of

DNA via membrane blebs of Neisseria gonorrhoeae. J Bacteriol
171:2499–2505.

Dorward DW, Schwan TG, Garon CF. 1991. Immune capture and detection of

Borrelia burgdorferi antigens in urine, blood, or tissues from infected
ticks, mice, dogs, and humans. J Clin Microbiol 29:1162–1170.

Drabick JJ, Brandt BL, Moran EE, Saunders NB, Shoemaker DR, Zollinger

WD. 1999. Safety and immunogenicity testing of an intranasal group B
meningococcal native outer membrane vesicle vaccine in healthy
volunteers. Vaccine 18:160–172.

Duffy CF, Fuller KM, Malvey MW, O’Kennedy R, Arriaga EA. 2002.

Determination of electrophoretic mobility distributions through the
analysis of individual mitochondrial events by capillary electrophoresis
with laser-induced fluorescence detection. Anal Chem 74:171–176.

Ferrari G, Garaguso I, Adu-Bobie J, Doro F, Taddei AR, Biolchi A, Brunelli

B, Giuliani MM, Pizza M, Norais N, Grandi G. 2006. Outer membrane
vesicles from group B Neisseria meningitides Dgna33 mutant:
Proteomic and immunological comparison with detergent-derived
outer membrane vesicles. Proteomics 6:1856–1866.

Fiocca R, Necchi V, Sommi P, Ricci V, Telford J, Cover TL, Solcia E. 1999.

Release of Helicobacter pylori vacuolating cytotoxin by both a
specific secretion pathway and budding of outer membrane vesicles.
Uptake of released toxin and vesicles by gastric epithelium. J Pathol
188:220–226.

Fuchs R, Male P, Mellman I. 1989. Acidification and ion permeabilities of

highly purified rat liver endosomes. J Biol Chem 264:2212–2220.

Fuller KM, Arriaga EA. 2003. Analysis of individual acidic organelles by

capillary electrophoresis with laser-induced fluorescence detection
facilitated by the endocytosis of fluorescently labeled microspheres.
Anal Chem 75:2123–2130.

Fuller KM, Arriaga EA. 2004. Capillary electrophoresis monitors changes in

the electrophoretic behavior of mitochondrial preparations. J Chroma-
togr B Analyt Technol Biomed Life Sci 806:151–159.

Galan JE, Collmer A. 1999. Type III secretion machines: Bacterial devices for

protein delivery into host cells. Science 284:1322–1328.

Galdiero M, Folgore A, Molitierno M, Greco R. 1999. Porins and

lipopolysaccharide (LPS) from Salmonella typhimurium induce
leucocyte transmigration through human endothelial cells in vitro. Clin
Exp Immunol 116:453–461.

Gardy JL, Laird MR, Chen F, Rey S, Walsh CJ, Ester M, Brinkman FS. 2005.

PSORTb v.2.0: Expanded prediction of bacterial protein subcellular
localization and insights gained from comparative proteome analysis.
Bioinformatics 21:617–623.

Gingras AC, Gstaiger M, Raught B, Aebersold R. 2007. Analysis of protein

complexes using mass spectrometry. Nat Rev Mol Cell Biol 8:645–654.

Girard MP, Preziosi MP, Aguado MT, Kieny MP. 2006. A review of vaccine

research and development: Meningococcal disease. Vaccine 24:4692–
4700.

Graham R, Graham C, McMullan G. 2007. Microbial proteomics: A mass

spectrometry primer for biologists. Microb Cell Fact 6:26.

Grenier D, Mayrand D. 1987. Functional characterization of extracellular

vesicles produced by Bacteroides gingivalis. Infect Immun 55:111–
117.

Gruhler A, Schulze WX, Matthiesen R, Mann M, Jensen ON. 2005. Stable

isotope labeling of Arabidopsis thaliana cells and quantitative
proteomics by mass spectrometry. Mol Cell Proteomics 4:1697–1709.

Guina T, Wu M, Miller SI, Purvine SO, Yi EC, Eng J, Goodlett DR, Aebersold

R, Ernst RK, Lee KA. 2003. Proteomic analysis of Pseudomonas

&

LEE ET AL.

552

Mass Spectrometry Reviews DOI 10.1002/mas

aeruginosa grown under magnesium limitation. J Am Soc Mass
Spectrom 14:742–751.

Gunasekera N, Musier-Forsyth K, Arriaga E. 2002. Electrophoretic behavior

of individual nuclear species as determined by capillary electrophoresis
with laser-induced fluorescence detection. Electrophoresis 23:2110–
2116.

Gygi SP, Rist B, Griffin TJ, Eng J, Aebersold R. 2002. Proteome analysis of

low-abundance proteins using multidimensional chromatography and
isotope-coded affinity tags. J Proteome Res 1:47–54.

Hecker M, Volker U. 2004. Towards a comprehensive understanding of

Bacillus subtilis cell physiology by physiological proteomics. Proteo-
mics 4:3727–3750.

Hellman J, Loiselle PM, Zanzot EM, Allaire JE, Tehan MM, Boyle LA,

Kurnick JT, Warren HS. 2000. Release of gram-negative outer-
membrane proteins into human serum and septic rat blood and their
interactions with immunoglobulin in antiserum to Escherichia coli J5. J
Infect Dis 181:1034–1043.

Henry T, Pommier S, Journet L, Bernadac A, Gorvel JP, Lloubes R. 2004.

Improved methods for producing outer membrane vesicles in Gram-
negative bacteria. Res Microbiol 155:437–446.

Hoekstra D, van der Laan JW, de Leij L, Witholt B. 1976. Release of outer

membrane fragments from normally growing Escherichia coli.
Biochim Biophys Acta 455:889–899.

Horstman AL, Kuehn MJ. 2000. Enterotoxigenic Escherichia coli secretes

active heat-labile enterotoxin via outer membrane vesicles. J Biol Chem
275:12489–12496.

Horstman AL, Kuehn MJ. 2002. Bacterial surface association of heat-labile

enterotoxin through lipopolysaccharide after secretion via the
general secretory pathway. J Biol Chem 277:32538–32545.

Howard M, Rutenberg AD, de Vet S. 2001. Dynamic compartmentalization of

bacteria: Accurate division in E. coli. Phys Rev Lett 87:278102.

Hua S, Sun Z. 2001. Support vector machine approach for protein subcellular

localization prediction. Bioinformatics 17:721–728.

Imamura T, Potempa J, Pike RN, Moore JN, Barton MH, Travis J. 1995. Effect

of free and vesicle-bound cysteine proteinases of Porphyromonas
gingivalis on plasma clot formation: Implications for bleeding tendency
at periodontitis sites. Infect Immun 63:4877–4882.

Ingram JM, Cheng KJ, Costerton JW. 1973. Alkaline phosphatase of

Pseudomonas aeruginosa: The mechanism of secretion and release of
the enzyme from whole cells. Can J Microbiol 19:1407–1415.

Jin M, Drwal G, Bourgeois T, Saltz J, Wu HM. 2005. Distinct proteome

features of plasma microparticles. Proteomics 5:1940–1952.

Kadurugamuwa JL, Beveridge TJ. 1995. 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.

Kadurugamuwa JL, Beveridge TJ. 1996. Bacteriolytic effect of membrane

vesicles from Pseudomonas aeruginosa on other bacteria including
pathogens: Conceptually new antibiotics. J Bacteriol 178:2767–2774.

Kadurugamuwa JL, Beveridge TJ. 1998. Delivery of the non-membrane-

permeative antibiotic gentamicin into mammalian cells by using
Shigella flexneri membrane vesicles. Antimicrob Agents Chemother
42:1476–1483.

Kadurugamuwa JL, Beveridge TJ. 1999. Membrane vesicles derived from

Pseudomonas aeruginosa and Shigella flexneri can be integrated into
the surfaces of other gram-negative bacteria. Microbiology (Reading,
England) 145(Pt 8):2051–2060.

Kanehisa M, Goto S, Kawashima S, Nakaya A. 2002. The KEGG databases at

GenomeNet. Nucleic Acids Res 30:42–46.

Katsui N, Tsuchido T, Hiramatsu R, Fujikawa S, Takano M, Shibasaki I. 1982.

Heat-induced blebbing and vesiculation of the outer membrane of
Escherichia coli. J Bacteriol 151:1523–1531.

Kerner MJ, Naylor DJ, Ishihama Y, Maier T, Chang HC, Stines AP,

Georgopoulos C, Frishman D, Hayer-Hartl M, Mann M, Hartl FU. 2005.

Proteome-wide analysis of chaperonin-dependent protein folding in
Escherichia coli. Cell 122:209–220.

Kesty NC, Kuehn MJ. 2004. Incorporation of heterologous outer membrane

and periplasmic proteins into Escherichia coli outer membrane vesicles.
J Biol Chem 279:2069–2076.

Kesty NC, Mason KM, Reedy M, Miller SE, Kuehn MJ. 2004. Enter-

otoxigenic Escherichia coli vesicles target toxin delivery into
mammalian cells. EMBO J 23:4538–4549.

Knox KW, Vesk M, Work E. 1966. Relation between excreted lip-

opolysaccharide complexes and surface structures of a lysine-limited
culture of Escherichia coli. J Bacteriol 92:1206–1217.

Kobayashi H, Uematsu K, Hirayama H, Horikoshi K. 2000. Novel toluene

elimination system in a toluene-tolerant microorganism. J Bacteriol
182:6451–6455.

Kolling GL, Matthews KR. 1999. Export of virulence genes and Shiga toxin

by membrane vesicles of Escherichia coli O157:H7. Appl Environ
Microbiol 65:1843–1848.

Korn A, Rajabi Z, Wassum B, Ruiner W, Nixdorff K. 1995. Enhancement of

uptake of lipopolysaccharide in macrophages by the major outer
membrane protein OmpA of gram-negative bacteria. Infect Immun
63:2697–2705.

Kostakioti M, Newman CL, Thanassi DG, Stathopoulos C. 2005. Mecha-

nisms of protein export across the bacterial outer membrane. J Bacteriol
187:4306–4314.

Kuehn MJ, Kesty NC. 2005. Bacterial outer membrane vesicles and the host-

pathogen interaction. Genes Dev 19:2645–2655.

Kushimoto T, Basrur V, Valencia J, Matsunaga J, Vieira WD, Ferrans VJ,

Muller J, Appella E, Hearing VJ. 2001. A model for melanosome
biogenesis based on the purification and analysis of early melanosomes.
Proc Natl Acad Sci USA 98:10698–10703.

Lee EY, Bang JY, Park GW, Choi DS, Kang JS, Kim HJ, Park KS, Lee JO, Kim

YK, Kwon KH, Kim KP, Gho YS. 2007. Global proteomic profiling of
native outer membrane vesicles derived from Escherichia coli.
Proteomics 7:3143–3153.

Li Z, Clarke AJ, Beveridge TJ. 1998. Gram-negative bacteria produce

membrane vesicles which are capable of killing other bacteria. J
Bacteriol 180:5478–5483.

Lindsay SS, Wheeler B, Sanderson KE, Costerton JW, Cheng KJ. 1973. The

release of alkaline phosphatase and of lipopolysaccharide during the
growth of rough and smooth strains of Salmonella typhimurium. Can J
Microbiol 19:335–343.

Loeb MR, Kilner J. 1978. Release of a special fraction of the outer membrane

from both growing and phage T4-infected Escherichia coli B. Biochim
Biophys Acta 514:117–127.

Lommatzsch J, Templin MF, Kraft AR, Vollmer W, Holtje JV. 1997. Outer

membrane localization of murein hydrolases: MltA, a third lipoprotein
lytic transglycosylase in Escherichia coli. J Bacteriol 179:5465–5470.

Lory S. 1992. Determinants of extracellular protein secretion in gram-

negative bacteria. J Bacteriol 174:3423–3428.

Lu Z, Szafron D, Greiner R, Lu P, Wishart, DS, Poulin B, Anvik J, Macdonell

C, Eisner R. 2004. Predicting subcellular localization of proteins using
machine-learned classifiers. Bioinformatics 20:547–556.

Marouga R, David S, Hawkins E. 2005. The development of the DIGE

system: 2D fluorescence difference gel analysis technology. Anal
Bioanal Chem 382:669–678.

Marsh M. 1989. Endosome and lysosome purification by free-flow electro-

phoresis. Methods Cell Biol 31:319–334.

Mashburn-Warren LM, Whiteley M. 2006. Special delivery: Vesicle

trafficking in prokaryotes. Mol Microbiol 61:839–846.

Mashburn LM, Whiteley M. 2005. Membrane vesicles traffic signals and

facilitate group activities in a prokaryote. Nature 437:422–425.

McBroom AJ, Johnson AP, Vemulapalli S, Kuehn MJ. 2006. Outer membrane

vesicle production by Escherichia coli is independent of membrane
instability. J Bacteriol 188:5385–5392.

PROTEOMICS IN BACTERIAL OUTER MEMBRANE VESICLES

&

Mass Spectrometry Reviews DOI 10.1002/mas

553

McBroom AJ, Kuehn MJ. 2007. Release of outer membrane vesicles by

Gram-negative bacteria is a novel envelope stress response. Mol
Microbiol 63:545–558.

Mergenhagen SE, Bladen HA, Hsu KC. 1966. Electron microscopic

localization of endotoxic lipopolysaccharide in gram-negative organ-
isms. Ann NY Acad Sci 133:279–291.

Misra RV, Horler RS, Reindl W, Goryanin II, Thomas GH. 2005. EchoBASE:

An integrated post-genomic database for Escherichia coli. Nucleic
Acids Res 33:D329–D333.

Mohr H, Volkl A. 2002. Isolation of peroxisomal subpopulations from mouse

liver by immune free-flow electrophoresis. Electrophoresis 23:2130–
2137.

Molloy MP, Herbert BR, Slade MB, Rabilloud T, Nouwens AS, Williams KL,

Gooley AA. 2000. Proteomic analysis of the Escherichia coli outer
membrane. Eur J Biochem 267:2871–2881.

Morand B, Muhlemann K. 2007. Heteroresistance to penicillin in

Streptococcus pneumoniae. Proc Natl Acad Sci USA 104:14098–
14103.

Morre DJ, Morre DM, Heidrich HG. 1983. Subfractionation of rat liver

Golgi apparatus by free-flow electrophoresis. Eur J Cell Biol 31:263–
274.

Mougous JD, Cuff ME, Raunser S, Shen A, Zhou M, Gifford CA, Goodman

AL, Joachimiak G, Ordonez CL, Lory S, Walz T, Joachimiak A,
Mekalanos JJ. 2006. A virulence locus of Pseudomonas aeruginosa
encodes a protein secretion apparatus. Science 312:1526–1530.

Nally JE, Whitelegge JP, Aguilera R, Pereira MM, Blanco DR, Lovett MA.

2005. Purification and proteomic analysis of outer membrane vesicles
from a clinical isolate of Leptospira interrogans serovar Copenhageni.
Proteomics 5:144–152.

Namork E, Brandtzaeg P. 2002. Fatal meningococcal septicaemia with

‘‘blebbing’’ meningococcus. Lancet 360:1741.

Nevot M, Deroncele V, Messner P, Guinea J, Mercade E. 2006. Character-

ization of outer membrane vesicles released by the psychrotolerant
bacterium Pseudoalteromonas antarctica NF3. Environ Microbiol
8:1523–1533.

Nolan LK, Horne SM, Giddings CW, Foley SL, Johnson TJ, Lynne AM,

Skyberg J. 2003. Resistance to serum complement, iss, and virulence of
avian Escherichia coli. Vet Res Commun 27:101–110.

Ong SE, Blagoev B, Kratchmarova I, Kristensen DB, Steen H, Pandey A,

Mann M. 2002. Stable isotope labeling by amino acids in cell culture,
SILAC, as a simple and accurate approach to expression proteomics.
Mol Cell Proteomics 1:376–386.

Ouzounis CA, Coulson RM, Enright AJ, Kunin V, Pereira-Leal JB. 2003.

Classification schemes for protein structure and function. Nat Rev
Genet 4:508–519.

Owen RL, Strasters JK, Breyer ED. 2005. Lipid vesicles in capillary

electrophoretic techniques: Characterization of structural properties
and associated membrane-molecule interactions. Electrophoresis
26:735–751.

Park GW, Kwon KH, Kim JY, Lee JH, Yun SH, Kim SI, Park YM, Cho SY,

Paik YK, Yoo JS. 2006. Human plasma proteome analysis by reversed
sequence database search and molecular weight correlation based on a
bacterial proteome analysis. Proteomics 6:1121–1132.

Peng X, Xu C, Ren H, Lin X, Wu L, Wang S. 2005. Proteomic analysis of the

sarcosine-insoluble outer membrane fraction of Pseudomonas aerugi-
nosa responding to ampicilin, kanamycin, and tetracycline resistance. J
Proteome Res 4:2257–2265.

Pisitkun T, Shen RF, Knepper MA. 2004. Identification and proteomic

profiling of exosomes in human urine. Proc Natl Acad Sci USA
101:13368–13373.

Post DM, Zhang D, Eastvold JS, Teghanemt A, Gibson BW, Weiss JP. 2005.

Biochemical and functional characterization of membrane blebs
purified from Neisseria meningitidis serogroup B. J Biol Chem
280:38383–38394.

Prasadarao NV, Wass CA, Weiser JN, Stins MF, Huang SH, Kim KS. 1996.

Outer membrane protein A of Escherichia coli contributes to invasion of
brain microvascular endothelial cells. Infect Immun 64:146–153.

Pugsley AP. 1993. The complete general secretory pathway in gram-negative

bacteria. Microbiol Rev 57:50–108.

Pukatzki S, Ma AT, Sturtevant D, Krastins B, Sarracino D, Nelson WC,

Heidelberg JF, Mekalanos JJ. 2006. Identification of a conserved
bacterial protein secretion system in Vibrio cholerae using the
Dictyostelium host model system. Proc Natl Acad Sci USA
103:1528–1533.

Rajendran L, Simons K. 2005. Lipid rafts and membrane dynamics. J Cell Sci

118:1099–1102.

Ratajczak J, Wysoczynski M, Hayek F, Janowska-Wieczorek A, Ratajczak

MZ. 2006. Membrane-derived microvesicles: Important and under-
appreciated mediators of cell-to-cell communication. Leukemia
20:1487–1495.

Renelli M, Matias V, Lo RY, Beveridge TJ. 2004. DNA-containing membrane

vesicles of Pseudomonas aeruginosa PAO1 and their genetic trans-
formation potential. Microbiology (Reading, England) 150:2161–
2169.

Riley M. 1993. Functions of the gene products of Escherichia coli. Microbiol

Rev 57:862–952.

Rolhion N, Barnich N, Claret L, Darfeuille-Michaud A. 2005. Strong

decrease in invasive ability and outer membrane vesicle release in
Crohn’s disease-associated adherent-invasive Escherichia coli strain
LF82 with the yfgL gene deleted. J Bacteriol 187:2286–2296.

Schooling SR, Beveridge TJ. 2006. Membrane vesicles: An overlooked

component of the matrices of biofilms. J Bacteriol 188:5945–5957.

Serres MH, Riley M. 2000. MultiFun, a multifunctional classification scheme

for Escherichia coli K-12 gene products. Microb Comp Genomics
5:205–222.

Stephens DS, Edwards KM, Morris F, McGee ZA. 1982. Pili and outer

membrane appendages on Neisseria meningitidis in the cerebrospinal
fluid of an infant. J Infect Dis 146:568.

Stumpe S, Schmid R, Stephens DL, Georgiou G, Bakker EP. 1998.

Identification of OmpTas the protease that hydrolyzes the antimicrobial
peptide protamine before it enters growing cells of Escherichia coli. J
Bacteriol 180:4002–4006.

Thery C, Zitvogel L, Amigorena S. 2002. Exosomes: Composition,

biogenesis and function. Nat Rev Immunol 2:569–579.

Uli L, Castellanos-Serra L, Betancourt L, Dominguez F, Barbera R,

Sotolongo F, Guillen G, Pajon Feyt R. 2006. Outer membrane vesicles
of the VA-MENGOC-BC vaccine against serogroup B of Neisseria
meningitidis: Analysis of protein components by two-dimensional gel
electrophoresis and mass spectrometry. Proteomics 6:3389–3399.

van der Ley P, Steeghs L, Hamstra HJ, ten Hove J, Zomer B, van Alphen L.

2001. Modification of lipid A biosynthesis in Neisseria meningitidis
lpxL mutants: Influence on lipopolysaccharide structure, toxicity, and
adjuvant activity. Infect Immun 69:5981–5990.

Vipond C, Suker J, Jones C, Tang C, Feavers IM, Wheeler JX. 2006.

Proteomic analysis of a meningococcal outer membrane vesicle vaccine
prepared from the group B strain NZ98/254. Proteomics 6:3400–3413.

Voulhoux R, Ball G, Ize B, Vasil ML, Lazdunski A, Wu LF, Filloux A. 2001.

Involvement

of

the

twin-arginine

translocation

system

in

protein secretion via the type II pathway. EMBO J 20:6735–6741.

Wai SN, Lindmark B, Soderblom T, Takade A, Westermark M, Oscarsson J,

Jass J, Richter-Dahlfors A, Mizunoe Y, Uhlin BE. 2003. Vesicle-
mediated export and assembly of pore-forming oligomers of the
enterobacterial ClyA cytotoxin. Cell 115:25–35.

Washburn MP, Wolters D, Yates JR III. 2001. Large-scale analysis of the yeast

proteome by multidimensional protein identification technology. Nat
Biotechnol 19:242–247.

Washburn MP, Yates JR III. 2000. Analysis of the microbial proteome. Curr

Opin Microbiol 3:292–297.

&

LEE ET AL.

554

Mass Spectrometry Reviews DOI 10.1002/mas

Wei C, Yang J, Zhu J, Zhang X, Leng W, Wang J, Xue Y, Sun L, Li W, Wang J,

Jin Q. 2006. Comprehensive proteomic analysis of Shigella flexneri 2a
membrane proteins. J Proteome Res 5:1860–1865.

Wensink J, Witholt B. 1981. Outer-membrane vesicles released by normally

growing Escherichia coli contain very little lipoprotein. Eur J Biochem
116:331–335.

Wu CC, Yates JR III. 2003. The application of mass spectrometry to

membrane proteomics. Nat Biotechnol 21:262–267.

Xu C, Lin X, Ren H, Zhang Y, Wang S, Peng X. 2006. Analysis of outer

membrane proteome of Escherichia coli related to resistance to
ampicillin and tetracycline. Proteomics 6:462–473.

Yaron S, Kolling GL, Simon L, Matthews KR. 2000. Vesicle-mediated

transfer of virulence genes from Escherichia coli O157:H7 to other
enteric bacteria. Appl Environ Microbiol 66:4414–4420.

Yates JR III, Gilchrist A, Howell KE, Bergeron JJ. 2005. Proteomics of

organelles and large cellular structures. Nat Rev Mol Cell Biol 6:702–
714.

Zhang H, Yan W, Aebersold R. 2004. Chemical probes and tandem mass

spectrometry: A strategy for the quantitative analysis of proteomes and
subproteomes. Curr Opin Chem Biol 8:66–75.

Zhou L, Srisatjaluk R, Justus DE, Doyle RJ. 1998. On the origin of membrane

vesicles in gram-negative bacteria. FEMS Microbiol Lett 163:223–
228.

Zischka H, Braun RJ, Marantidis EP, Buringer D, Bornhovd C, Hauck

SM, Demmer O, Gloeckner CJ, Reichert AS, Madeo F, Ueffing M.
2006. Differential analysis of Saccharomyces cerevisiae mito-
chondria by free flow electrophoresis. Mol Cell Proteomics 5:2185–
2200.

Eun-Young Lee

received a B.Sc. degree in the Division of Life Science from Korea

University, Republic of Korea (2005). Presently she is a Ph.D. candidate at Pohang
University of Science and Technology, Republic of Korea, with a focus on the diverse roles
of bacterial outer membrane vesicles and mammalian cell-derived microvesicles under the
supervision of Prof. Yong Song Gho.

Dong-Sic Choi

received a B.Sc. degree in the Department of Life Science and Division of

Molecular and Life Sciences from Pohang University of Science and Technology, Republic
of Korea (2006). He is now a Ph.D. candidate at Pohang University of Science and
Technology and his research involves proteomic analysis of microvesicles and plasma
membrane proteins derived from eukaryotic cells under the supervision of Prof. Yong Song
Gho.

Kwang-Pyo Kim

received his B.Sc. and M.Sc. in Chemistry from Seoul National

University, Republic of Korea in 1990 and 1992, respectively. After working for a
pharmaceutical company, CJ corp., he started a Ph.D. program in Chemistry from the
University of Illinois at Chicago. In 2002 he thereafter worked in Harvard Medical School
as a postdoctoral fellow. Since 2004, he joined the Department of Molecular Biotechnology
at the Konkuk University in the Republic of Korea as an Assistant Professor. His current
research areas focus on: development and application of proteomics technologies to
investigate post-translational modifications, discovery of biomarkers with comparative
proteomics technologies, and MALDI Mass Tissue imaging.

Yong Song Gho

received his B.Sc. and M.Sc. degrees in Chemistry from Seoul National

University, Republic of Korea in 1987 and 1989, respectively. He obtained his Ph.D. degree
in Biochemistry and Biophysics from University of North Carolina at Chapel Hill, USA in
1997. During 1998–2000, he was a visiting fellow at NIDCR of National Institutes of
Health, USA. From 2000 to 2004, he was an assistant professor at Kyung Hee University,
Republic of Korea. Since 2004, he became an assistant professor in the Department of Life
Science and Division of Molecular and Life Sciences at Pohang University of Science and
Technology, Republic of Korea. His current research interests aim at elucidating the
biogenesis and pathophysiological functions of extracellular membrane vesicles derived
from bacteria and mammalian cells, as well as determining profiles of membrane vesicular
genes and proteins using microarray and mass spectrometry.

PROTEOMICS IN BACTERIAL OUTER MEMBRANE VESICLES

&

Mass Spectrometry Reviews DOI 10.1002/mas

555