Download Protein targeting, translocation and Escherichia coli Proteomic analysis of substrate-pathway relationships

Protein targeting, translocation and
insertion in Escherichia coli
Proteomic analysis of substrate-pathway relationships
Louise Baars
Stockholm University
Doctoral thesis
© Louise Baars, Stockholm 2007
ISSN 978-91-7155-481-9, pp 1-55
Printed in Sweden by Universitetsservice AB, Stockholm 2007
Distributor: Stockholm University Library
To my family
Cover illustration
A cross-section of a small part of an Escherichia coli cell. The average
distribution of molecules is shown at the proper scale. The outer and the
inner membranes, studded with transmembrane proteins, are shown in green.
A flagellum extends upwards from the cell surface. The motion of the
flagellum is driven by the large flagellar motor that crosses both membranes.
The cytoplasmic area is colored blue and purple. The large purple molecules
are ribosomes and the small, L-shaped maroon molecules are tRNA, and the
white strands are mRNA. Enzymes are shown in blue. The nucleoid region is
shown in yellow and orange, with the long DNA circle shown in yellow.
Reprinted with permission from David S. Goodsell©1999
List of Publications
This thesis is based on the following publications:
I.
Marani P, Wagner S, Baars L, Genevaux P, de Gier JW, Nilsson I,
Casadio R, von Heijne G. New Escherichia coli outer membrane proteins
identified through prediction and experimental verification. (2006) Prot
Sci 15, 884-9.
II.
Linda Fröderberg, Edith Houben, Louise Baars, Joen Luirink and JanWillem L de Gier. Targeting and translocation of two lipoproteins in Escherichia coli via the SRP/Sec/YidC pathway. (2004) J Biol Chem 279,
31026-32.
III.
Baars L, Ytterberg AJ, Drew D, Wagner S, Thilo C, van Wijk KJ, de
Gier JW. Defining the role of the Escherichia coli chaperone SecB using
comparative proteomics. (2006) J Biol Chem 281, 10024-34.
IV.
Louise Baars, Samuel Wagner, David Wickström, Mirjam Klepsch, A.
Jimmy Ytterberg, Klaas J. van Wijk and Jan-Willem de Gier. Global
analysis of an Escherichia coli SecE depletion strain. Submitted to J Biol
Chem
Reprints were made with permission from the publishers.
Additional publications:
Samuel Wagner, Louise Baars, A. Jimmy Ytterberg, Anja Klußmeier,
Claudia S. Wagner, Olof Nord, Per-Åke Nygren, Klaas J. van Wijk, JanWillem de Gier. Consequences of membrane protein overexpression in
Escherichia coli. (2007) Mol Cell Proteomics
Drew D, Fröderberg L, Baars L, de Gier JW. Assembly and over expression of membrane proteins in Escherichia coli. (2003) Biochim Biophys
Acta 1610, 3-10.
Annika Sääf, Louise Baars, Gunnar von Heijne. The two internal repeats
in the E.coli YrbG hypothetical Na+/Ca2+ transporter have opposite orientations in the inner membrane. (2001) J Biol Chem 276, 18905-7.
Contents
Cover illustration ............................................................................................ iv
List of Publications .......................................................................................... v
Abbreviations ............................................................................................... viii
1. Introduction .................................................................................................1
1.1 The biological membrane .......................................................................................... 2
1.2 Protein structure ........................................................................................................ 3
1.2.1 Globular Proteins .............................................................................................. 4
1.2.2 Membrane Proteins........................................................................................... 4
1.3 The model organism - Escherichia coli ..................................................................... 6
1.3.1 The cytoplasmic proteome................................................................................ 8
1.3.2 The inner membrane proteome ........................................................................ 8
1.3.3 The periplasmic proteome ................................................................................ 8
1.3.4 The outer membrane proteome ........................................................................ 9
1.3.5 The extracellular proteome ............................................................................... 9
2. Biogenesis of secretory and integral
membrane proteins in E. coli.......11
2.1 Targeting to the inner membrane............................................................................ 14
2.1.1 Trigger Factor ................................................................................................. 15
2.1.2 The SecB-pathway.......................................................................................... 15
2.1.2 SRP-pathway .................................................................................................. 18
2.2 Translocation across, and insertion into, the inner membrane ............................... 21
2.2.1 The Sec-translocon......................................................................................... 21
2.2.2 The SecDFYajC-complex ............................................................................... 25
2.2.3 YidC ................................................................................................................ 26
2.2.4 The TAT-pathway ........................................................................................... 27
2.3 Maturation and sorting of lipoproteins ..................................................................... 28
2.4 Insertion of β-barrel proteins into the outer membrane........................................... 30
3. Objectives .................................................................................................32
4. Summary of papers...................................................................................33
4.1 Paper I - New Escherichia coli outer membrane proteins identified through
prediction and experimental verification........................................................................ 33
4.2 Paper II - Targeting and translocation of two lipoproteins in Escherichia coli via the
SRP/Sec/YidC pathway................................................................................................. 34
4.3 Paper III - Defining the role of the Escherichia coli chaperone SecB using
comparative proteomics ................................................................................................ 35
4.4 Paper IV - Effects of SecE depletion on the inner and outer membrane proteomes
of E. coli......................................................................................................................... 36
5. Concluding remarks ..................................................................................39
6. Sammanfattning på svenska – Hur hittar proteiner rätt i en cell?.............40
7. Acknowledgements ...................................................................................42
8. References................................................................................................44
Abbreviations
1DE
2DE
ABC
ADP
Asp
ATP
BN
C-terminus
Cys
D
DNA
E. coli
GDP
GTP
IMP
LPS
M. jannaschii
mRNA
MS
N-terminus
OMP
ORF
PAGE
pmf
RNA
RNC
SRP
TAT
TF
TM
tRNA
one-dimensional electrophoresis
two-dimensional electrophoresis
ATP-binding cassette
adenosine diphosphate
aspartic acid
adenosine triphosphate
blue native
carboxy terminus
cysteine
aspartic acid
deoxyribonucleic acid
Escherichia coli
guanosine 5'-diphosphate
auanosine triphosphate
integral inner membrane protein
lipopolysaccharide
Methanococcus jannaschii
messenger ribonucleic acid
mass spectrometry
amino terminus
integral outer membrane protein
open reading frame
polyacrylamide gel electrophoresis
proton motive force
ribonucleic acid
ribosome nascent chain complex
Signal recognition particle
twin-arginine protein transport
Trigger factor
transmembrane
transfer ribonucleic acid
1. Introduction
Proteins are biological macromolecules with diverse cellular functions,
ranging from enzymatic catalysis to the formation of cytoskeletal structures
that gives shape and rigidity to the cell. Once a protein has been synthesized
inside the cell, it has to be sorted to the cellular compartment(s) where it is
designed to function. Miss-localization of proteins can be harmful to the cell
for a number of reasons, one being the formation of toxic protein aggregates.
The focus of this thesis is the cellular infrastructure that ensures the efficient
sorting of proteins to their final destination. Central questions are how
proteins destined for a certain compartment are recognized, and what cellular
components are involved in the translocation of proteins across, or insertion
into, biological membranes. These questions have been addressed using the
Gram-negative bacterium Escherichia coli as a model organism.
Importantly, several of the underlying principles that govern protein sorting
are evolutionary conserved and discoveries made in a unicellular organism
such as E. coli bacteria are often applicable to the specialized cells of
multicellular organisms and vice versa.
1
1.1 The biological membrane
Biological membranes are thin, fluid structures that form the boundary
between the exterior and interior of the cell, as well as between different
cellular compartments. Biological membranes consist of a mixture of lipids
and proteins. Membrane lipids assemble into bilayers, illustrated in Figure
1a. A bilayer functions as a permeability barrier that prevents leakage and
enables different chemical environments to exist on each side of the
membrane. This is a fundamental aspect of life since many biological
processes depend on the formation of concentration gradients across a
membrane. However, for a cell to survive, it is essential that membranes
permit passage of various molecules such as nutrients and ions. Proteins are
therefore essential components of the biological membrane since they can
function as pores, channels or transporters that allow selective passage
across the lipid bilayer (Figure 1a).
Figure 1. The Biological membrane. (a) Schematic representation of a biological
membrane consisting of lipids (light grey) and membrane proteins (dark grey) [1].
(b) Chemical structure of a membrane lipid, phosphatidyletanolamin.
Membrane lipids are amphipathic molecules consisting of a hydrophilic head
group and a hydrophobic tail (Figure 1b). The lipid tail region consists of
two acyl carbon chains that are unable to form hydrogen bonds. To avoid
disrupting the hydrogen bonding network that exists between water
molecules, membrane lipids therefore self-assemble into a bilayer where the
lipid tails point towards the interior, while the head groups face the aqueous
exterior on each side of the bilayer. In this way, a 20-30 Å thick hydrophobic
core is created in the middle of the membrane, boarded on each side by a 15
Å thick hydrophilic interfacial region [2]. This organization effectively
2
minimizes the energetic cost for lipids to exist in a polar, aqueous
environment. The hydrophobic core region makes the bilayer impermeable
by blocking passage of polar molecules. Membrane lipids have different
properties depending on the chemical composition of their head and tail
regions. The head groups vary in size and polarity; they can be neutral,
zwitterionic or negatively charged. The acyl chains of the lipid tails are
typically 16-18 carbons long and differ in the degree of saturation. Double
bonds reduce the flexibility of the hydrocarbon tail and give the lipid a cone
shaped conformation, whereas fully saturated lipids are cylindrical. The
proportion of different lipids in a bilayer determines the physical properties
of the membrane such as thickness, curvature, strength and fluidity [3].
In 1972, Singer and Nicolson presented the fluid mosaic model in which
proteins were viewed as icebergs floating freely in a sea of lipids [4].
However, the lateral movement of both lipids and proteins can be restricted.
Today, most membranes are believed to contain patches that have a different
lipid and protein composition than the surrounding membrane. These
patches can vary in thickness and fluidity. One function of membrane
patches is to locally increase the concentration of proteins (and lipids)
involved in the same biological process [1]. Furthermore, the fluid mosaic
model and most text book pictures of biological membranes greatly
exaggerate the area occupied by lipids and underestimates the areas occupied
by protein [1]. In addition, lipids do not only function as membrane protein
solvents, but can sometimes interact specifically with membrane proteins
and act as ‘co-factors’. For instance, a number of proteins involved in
bioenergetics, such as NADH dehydrogenase, cytochrome bc1, ATP
synthase, cytochrome oxidase, and the ADP/ATP carrier require cardiolipin
to function properly [5].
1.2 Protein structure
The basic building blocks of all proteins are 20 different amino acids that
contain an amino group, a carboxyl group and a side chain that varies in size
and polarity. Proteins consist of polypetide chains formed by amino acids
joined together by peptide bonds. These bonds are formed in a condensation
reaction when the amino group from one amino acid reacts with the carboxyl
group of another. The specific combination and linear arrangement of amino
acids are unique for each protein and encoded in the DNA. To form a
functional protein, the polypeptide chain has to “fold” into a three
dimensional structure in which all atoms are organized to form energetically
3
favorable interactions, either with each other or with atoms present in its
surrounding. Proteins are therefore designed to fit into a specific
environment, e.g., in an aqueous environment, in a complex with other
proteins or in a membrane where they interact with lipids. When a protein
fails to reach the localization where it is supposed to be, it may form
unproductive interactions leading to e.g., protein aggregation. To avoid this,
the cell contains ‘chaperones’, proteins that prevent unproductive
interactions and facilitate proper folding, but also proteases that degrade
miss-folded proteins.
1.2.1 Globular Proteins
Proteins that function in an aqueous environment typically acquire a
‘globular‘ fold. A major driving force in the folding of globular proteins is
the hydrophobic effect. The polypeptide folds into a conformation where
hydrophobic side chains are buried in the interior, while hydrophilic amino
acids typically are localized to the surface of the protein. The structure is
stabilized by favorable intramolecular interactions as well as by the positive
entropic effect on the surrounding water molecules.
1.2.2 Membrane Proteins
Membrane proteins are defined as ‘integral’ or ‘peripheral’ depending on
their mode of interaction with the membrane. Peripheral membrane proteins
are associated with the membrane via electrostatic and hydrophobic
interactions with membrane lipids and/or other membrane proteins. Addition
of chemicals (e.g., high salt or alkaline conditions) that disrupt electrostatic
interactions leads to removal of most peripheral proteins from the
membrane. Proteins can also be anchored to the membrane by covalently
attached hydrophobic moieties. One example is bacterial lipoproteins that
are modified with fatty acids that insert into the lipid bilayer. In contrast,
integral membrane proteins consist of one or more segments that are fully
embedded in the membrane. The folding environment of these
transmembrane segments is fundamentally different from the aqueous
environment outside the membrane. For a transmembrane segment to exist in
the hydrophobic core of the lipid bilayer, the polypeptide chain has to fold
into a structure that stabilizes the amino acid side chains and the polar amide
and carbonyl groups of the peptide bond. Two different types of integral
membrane proteins are known; β-barrel membrane proteins and α-helical
membrane proteins.
4
Figure 2. Membrane proteins. Ribbon diagram of (a) a β-barrel membrane protein
(PDB 1PRN) and (b) an α-helical membrane protein (PDB 2BRD). The yellow lines
indicate the boarders of the membrane.
So far, β-barrel membrane proteins have only been found in the outer
membrane of Gram-negative bacteria and mitochondria, and in the outer
envelope of chloroplasts. β-barrel membrane proteins are composed of an
even number of anti-parallel β-strands arranged in a barrel like structure. The
amide and carbonyl groups of the polypeptide backbone are stabilized by
hydrogen bonds between neighboring strands. Amino acid side chains of
mixed polarity point towards the center of the barrel, while amino acids with
apolar side chains project into the lipid bilayer on the outside surface. βbarrel membrane proteins often function as aqueous pores that allow passage
of small solutes.
α-helical membrane proteins represent a conserved structural fold that is
found in all organisms. In this class of proteins, the transmembrane (TM)
domain folds into an α-helix where the amide and carbonyl groups of the
polypeptide backbone are stabilized by internal hydrogen bonds. The TM
segments are enriched in hydrophobic side chains that can interact favorably
with the acyl carbons chains of the lipids. Aromatic residues that can interact
with the lipid head groups are enriched at the end of the TM segment [6].
The length of each TM domain is limited by the thickness of the membrane.
Typically, a membrane spanning segment consists of approximately 20
amino acids arranged in an α-helix that is perpendicular to the membrane.
However, the TM segments can also be tilted, which allows longer α-helices
to fit into the membrane. Based on studies of membrane proteins with known
structures, it has been estimated that the average number of amino acids in
TM α-helices is 26 [7]. Furthermore, an increasing number of high
resolution structures has revealed the existence of atypical α-helices that
vary greatly in length, tilt angle and position in the membrane [8].
5
Single spanning α-helical membrane proteins have of one TM domain, while
multispanning α-helical membrane proteins have of two or more TM
domains connected together by hydrophilic loop regions. The loop regions
can be short linkers consisting of just a few amino acids or they can consist
of a long segment, which folds into globular domain in the aqueous
environment outside the membrane. A topological model of a membrane
protein is a two-dimensional representation of the protein, showing the
number and position of TM domains and loop regions and their orientation
relative to the membrane. The simplest way to predict TM domains is to
search for 20 amino acid long segments with an overall hydrophobic
character. Theoretical and experimental studies have revealed that the
orientation of TM domains is determined by the relative distribution of
positively charged amino acids in the regions flanking them [9, 10]. In all
organisms, the TM flanking regions that are localized on the cytoplasmic
side of the membrane are enriched in positively charged residues [11]. This
observation, referred to as ‘the positive inside rule’, can be used to predict
the orientation of membrane proteins TM domains.
1.3 The model organism - Escherichia coli
The Gram-negative bacterium E. coli is by far the most well studied
organism on Earth, and the E. coli genome was one of the first to be
completely sequenced. The chromosome consists of a single, circular DNA
containing approximately 4.6 million base pairs, harboring around four
thousand open reading frames (ORFs) [12]. The chromosome is stored in the
cytoplasm, an aqueous compartment encapsulated by the inner membrane
(Figure 3b). The E. coli inner membrane is composed of phospholipids (70–
80% phosphatidyl-ethanolamine, 15–20% phosphatidylglycerol and <5%
cardiolipin) and membrane proteins [13]. Gram-negative bacteria such as E.
coli also have an outer membrane that surrounds the inner membrane (Figure
3b). The composition of the outer membrane is highly asymmetric with
phospholipids (enriched in saturated fatty acids and phosphatidylethanolamine) in the inner leaflet, while the main constituent of the outer
leaflet is lipopolysaccharide (LPS) [13]. LPS is a unique, complex molecule
that is important for the barrier function of the outer membrane. The
compartment between the outer membrane and the inner membrane is called
the periplasm (Figure 3b). The periplasm contains the peptidoglycan layer,
which serves as an extracytoplasmic cytoskeleton that contributes to cell
shape and prevents cells from lysing in dilute environments. It is covalently
6
attached to the outer membrane lipoprotein Lpp. The environment
surrounding a bacterium is called the extracellular space.
Each E. coli compartment contains its own ‘proteome’, a dynamic mixture of
proteins regulated to fit the requirements of the cell. Since all E. coli proteins
are synthesized in the cytoplasm, proteins destined for the inner membrane,
the periplasm, the outer membrane, or the extracellular space, have to be
sorted to their correct compartment.
Figure 3. The E. coli cell. (a) An E. coli cell visualized by electron microscopy. (b)
A cross-section of an E. coli cell showing the cytoplasm, the inner membrane, the
periplasm containing the peptidoglycan layer, and the outer membrane with LPS in
the outer leaflet (adapted from [14]).
7
1.3.1 The cytoplasmic proteome
The E. coli cytoplasm is a crowded compartment [15]. Out of the around
four thousand ORFs in the E. coli genome, approximately 70% encodes
proteins that are predicted to remain in the cytoplasm [8]. The composition
of the cytoplasmic proteome reflects the role of the cytoplasm as the keeper
of the genome; it consists of proteins involved in replication and protection
of the DNA, gene transcription, protein translation and protein sorting. Other
major constituents are proteins involved in metabolic processes. Several of
these proteins interact with proteins localized at the inner membrane. The
cytoplasmic proteome also includes proteins involved in maintaining cellular
homeostasis, e.g., molecular chaperones that prevent protein miss-folding
and proteases that degrade proteins that are miss-folded or no longer needed.
1.3.2 The inner membrane proteome
The E. coli inner membrane proteome consists of α-helical integral
membrane proteins (from hereafter referred to as IMPs), lipoproteins and
peripheral proteins. Based on predictions, it is estimated that approximately
20% of the ORFs in the E. coli genome encode IMPs [16]. As mentioned
earlier, one of the main functions of membrane proteins is to allow selective
passage of e.g., ions, nutrients and polypeptides across the membrane. Thus,
the inner membrane proteome contains a large variety of different channels
and transporters. Several IMPs are involved in transduction of information
across the membrane. These proteins are often receptors that transmit the
information by changing their conformation in response to a specific
stimulus, such as the interaction with a messenger molecule or a change in
the pressure profile in the membrane. The inner membrane is an important
site of action when it comes to bioenergetics and it contains a system of
proteins involved in the conservation and utilization of energy stored in the
form of a proton gradient across the membrane.
1.3.3 The periplasmic proteome
Around ten percent of the ORFs in the E. coli genome encodes periplasmic
proteins [8]. A large proportion of the periplasmic proteome consists of
degradative enzymes and binding proteins for amino acids, carbohydrates,
ions and vitamins. The periplasm is an oxidizing environment and
periplasmic proteins often contain intermolecular disulphide bonds. The
oxidation of cysteins into disulphides is catalyzed by chaperons with an
8
oxido-reductase function [17]. The reducing conditions in the cytoplasm, and
the absence of these chaperones, prevent folding of periplasmic proteins
prior to translocation across the inner membrane.
1.3.4 The outer membrane proteome
The outer membrane proteome consists of outer membrane lipoproteins and
integral outer membrane proteins (OMPs) of the β-barrel type. The E. coli
genome is predicted to encode approximately 80 outer membrane
lipoproteins [18] and around 100 integral OMPs [8]. However, only a
fraction of these proteins has been identified in the outer membrane
experimentally. In paper I, eight putative OMPs were studied and the outer
membrane localization of five of them was confirmed. Proteomics studies,
including paper III, have identified approximately 30 additional proteins in
the outer membrane. The majority of these are ‘porins’, OMPs that function
as aqueous pores and allow passive diffusion of small (<700 Da) molecules
through the membrane. One example is the highly abundant outer membrane
protein OmpA, a porin with low permeability that allows slow penetration of
small solutes [19]. The most abundant E. coli outer membrane protein is the
lipoprotein Lpp, used as a model protein in paper II of this thesis. Lpp forms
a homotrimer and one of the molecules in the trimer is covalently attached to
the peptidoglycan layer in the periplasm. A recent structural study has shown
that the outer membrane of E. coli contains at least one integral membrane
protein (Wza) that is not of the β-barrel type [20]. Instead Wza is structured
as a novel transmembrane α-helical barrel.
1.3.5 The extracellular proteome
Most bacteria export proteins to the extracellular environment. These
extracellular proteins can e.g, be involved in digestion of polymers for
nutritional purposes, or act as toxins during infection of host cells. It is
generally assumed that nonpathogenic E. coli strains cultured for laboratory
purposes do not secrete proteins to the extracellular environment. This view
has recently been challenged by a proteomics study. In this study, it was
shown that several proteins can be precipitated from the culture media of a
laboratory E. coli strain [21]. Using two-dimensional electrophoresis, 66
protein spots were visualized and 44 different proteins were identified by
mass spectrometry. Several of these proteins were outer membrane proteins.
Since nucleic acids could not be detected in the growth medium it was
concluded that no significant cell lysis had occurred. Another study found
9
that native vesicles, containing miss-folded outer membrane and periplasmic
proteins, are released from the bacterial surface upon envelope stress [22].
Thus, it is possible proteins are exported to the extracellular environment by
vesicular secretion.
10
2. Biogenesis of secretory and integral
membrane proteins in E. coli
Protein biogenesis refers to the birth of a polypeptide and its maturation into
a functional protein. In this thesis, E. coli has been used as a model organism
to study one aspect of protein biogenesis - the sorting of proteins to the
correct compartment. All E. coli proteins are synthesized by ribosomes in the
cytoplasm. Using the messenger RNA (mRNA) as a template, ribosomes
translate the genetic information stored in the DNA into the amino acid
sequence specific for a particular protein. Proteins destined for translocation
across, or integration into, the inner membrane are recognized by targeting
factors that directs the proteins to the correct translocation/insertion
machinery at the inner membrane [23].
Most secretory proteins (periplasmic proteins, lipoproteins and OMPs) and
IMPs are targeted to the Sec-translocon, a protein-conducting channel that
mediates the translocation of proteins across, or inserted into, the inner
membrane [24]. Targeting of most IMPs is mediated by the signal
recognition particle (SRP), while the targeting of secretory proteins can be
facilitated by the chaperone SecB. The Sec-translocon is composed of the
IMPs SecY, SecE and SecG [25]. The peripheral subunit SecA is an ATPase
that powers translocation of secretory proteins and large periplasmic
domains of IMPs across the inner membrane. Several accessory proteins;
SecD, SecF, YajC and YidC, interact with the Sec-translocon to facilitate
different aspects of protein translocation and insertion [24]. Some IMPs have
been shown to integrate into the membrane via the Sec-translocon
independent YidC pathway [26]. In addition, at least one E. coli IMP appears
to require neither YidC nor the Sec-translocon for insertion, suggesting the
existence of unknown mechanisms for IMP insertion in E. coli [27]. A
subset of proteins are translocated via the Twin-Arginine protein Transport
(TAT)-pathway. In contrast to the Sec-translocon, which only translocates
proteins in an unfolded state, the TAT-translocon allows passage of folded
proteins and even oligomers across the inner membrane [28].
11
The goal of the work presented in this thesis has been to improve our
understanding of substrate-pathway relationships in protein targeting and
translocation/insertion processes. Simply put, to find out what components
are involved in the targeting, translocation and insertion of a specific
substrate and ultimately - to understand the basis behind the selection of a
certain pathway.
12
13
2.1 Targeting to the inner membrane
Secretory proteins are synthesized with a ‘signal peptide’, an N-terminal
extension that is proteolytically removed upon translocation [29]. Signal
peptides share little sequence homology but display several common
characteristics. They are approximately 20-25 amino acids long and typically
consist of a short, positively charged N-terminal region (n-region), a central,
hydrophobic region (h-region), and a polar C-terminal region (c-region)
containing the signal peptidase cleavage site [30, 31]. The signal peptide of
secretory proteins that are translocated via the TAT-pathway contain a
consensus sequence of S/T-R-R-X-F-L-K (where X is any polar amino acid)
[32]. The two arginine residues that have given the pathway its name are
almost invariant and essential for efficient translocation. TAT signal
peptides are longer than Sec signal peptides and consist on average of 38
amino acids [33]. The h-region of a TAT signal peptide is less hydrophobic
than the h-region of Sec signal peptides, and the c-region contains a
positively charged residue, the so-called ‘Sec-avoidance signal’ that prevents
targeting of TAT signal peptides to the Sec-translocon [34].
The vast majority of IMPs in E. coli does not contain a cleavable signal
peptide. Instead, the hydrophobic segment representing the first TM domain
acts as a signal sequence that directs the protein to the membrane.
Depending on the orientation of the TM domain in the membrane it is
referred to as a ‘signal-anchor sequence’ or a ‘reverse signal-anchor
sequence’. If an IMP does contain a cleavable signal sequence, it is followed
by a so-called ‘stop-transfer sequence’, a hydrophobic stretch that halts
translocation and gets inserted into the membrane [35].
Proteins are targeted to the inner membrane either ‘co-translationally’ or
‘post-translationally’. Co-translational targeting occurs while the nascent
chain is still attached to a translating ribosome. Post-translational targeting
takes place after translation is complete, or when a substantial part of the
polypetide has been synthesized. As the N-terminal end of a growing
polypeptide chain emerges from the ribosome, it is ‘scanned’ by the Signal
Recognition Particle (SRP) and Trigger Factor (TF), which are both
positioned at the ribosomal exit tunnel [36]. Signal sequences with a
hydrophobicity above a certain threshold interact preferentially with the SRP
and are selected for co-translational targeting to the Sec-translocon and, in
some cases, to YidC [24, 26]. Less hydrophobic signal sequences bypass
SRP, and are routed into the post-translational targeting pathway. Posttranslational targeting may be facilitated by the ribosome associated TF and
14
the cytoplasmic chaperone SecB [36]. In E. coli, most secretory proteins
follow the post-translational targeting route, while most IMPs and a few
secretory proteins are co-translationally targeted to the inner membrane.
2.1.1 Trigger Factor
TF is a non-essential chaperone that associates with the ribosome and
interacts with low affinity with nascent chains as they emerge from the
ribosomal exit tunnel [36]. The role of TF in protein targeting is not clear,
but some studies suggest that it prevents translating ribosomes from docking
onto the SecY complex [37].
2.1.2 The SecB-pathway
The E. coli protein SecB is a cytoplasmic chaperone involved in the
biogenesis of secretory proteins [38]. Homologues of SecB are found in
Gram-negative bacteria but not in Gram-positive bacteria, Archaea, and
eukaryotes. SecB can maintain newly synthesized preproteins in an
unfolded, i.e., translocation-competent state. Furthermore, SecB can target
preproteins to the Sec-translocon via a high affinity interaction with the
peripheral translocon subunit SecA. Previous studies have shown that SecB
is required for efficient translocation of Galactose binding protein, LamB,
Maltose binding protein, OmpA, OmpF, OppA and PhoE, while
translocation of β-lactamase, PhoA and Ribose binding protein appears to be
independent of SecB [38-42]. In paper I, the SecB dependence of five
putative outer membrane proteins was tested. In paper II, we investigated the
SecB dependence of two lipoproteins, Lpp and BRP. In paper III, we used a
proteomics approach to characterize a secB null mutant, which resulted in
the identification of an additional 12 SecB substrates.
SecB has no affinity for signal sequences but interacts with multiple sites in
the mature part of unfolded preproteins [43-47]. It has been shown that
binding of SecB to a preprotein requires that at least 150 amino acid residues
are exposed from the ribosome [48]. The SecB protein assembles into a
homotetramer with a molecular mass of 69 kDa [40, 49]. The high-resolution
structures of SecB from Haemophilus influenzae and E. coli demonstrate
that the tetramer is organized as a dimer of dimers [39, 40]. Two long
channels, one on each side of the tetramer, have been identified. Each
15
channel contains two putative peptide binding sites, one is a deep cleft
surrounded by aromatic residues, while the other is a shallow, open groove
with a hydrophobic surface (Figure 6). The SecB tetramer forms a
stoichiometric complex with precursors that are presumably wrapped around
the tetramer to access the binding sites of both channels [40, 50]. Multiple
sites in the unfolded preprotein probably interact simultaneously with SecB,
contributing to the high binding affinity (Kd~5-50 nM) [45].
SecB delivers preproteins to the Sec-translocon by interaction with the
peripheral translocon subunit SecA. In solution, the Kd of this interaction is
~1-2 μM [51]. The binding affinity increases (Kd 10–30 nM) when SecA is
bound to the Sec-translocon [52]. The binding is even tighter (Kd 10 nM) if
SecB is loaded with a preprotein [52, 53]. The SecB binding site on SecA is
localized primarily in the zink containing C-terminal region of SecA,
although additional sites have also been suggested [54-56]. Upon binding to
SecB, SecA recognizes and binds the signal sequence of the preprotein with
high specificity. The subsequent dissociation of the SecA-SecB complex is
coupled to the binding of ATP to SecA. Hydrolysis of ATP by SecA also
provides energy for the initial insertion and subsequent translocation of the
preprotein through the Sec-translocon [57-59]. It has been suggested that
SecB promotes the ATPase activity of SecA directly, by acting as an
intermolecular regulator of ATP hydrolysis [60].
How does SecB recognize its substrates? A peptide library screen was used
to identify a SecB binding motif that is nine amino acids long and enriched
in aromatic and basic residues, whereas acidic residues are disfavored [41].
This motif fits well into the proposed peptide binding site in the SecB
structure [40]. However, the SecB binding motif is common in both SecB
and independent secretory proteins, as well as in cytoplasmic proteins. The
presence or absence of the SecB binding motif can therefore not be used to
predict whether a secretory protein is targeted to the Sec-translocon by SecB.
It has been suggested that formation of a complex between SecB and its
substrates depends on a kinetic partitioning between the rate of substrate
folding and aggregation relative to the rate of its association to SecB [38].
SecB has no affinity for native structures and, as a consequence, only binds
proteins that are sufficiently slow folding. Signal sequences can sometimes
affect SecB binding indirectly; by retarding folding of the precursor, thereby
16
Figure 5. Model for SecB mediated targeting of secretory proteins to the Sectranslocon. The SecB tetramer binds the preprotein post-translationally or at a late
stage of translation (1). The SecB-preprotein complex interacts with SecA (2), which
binds to the signal sequence of the preprotein. The preprotein is transferred from
SecB to SecA (3). SecA delivers the preprotein to the Sec-translocon and drives
translocation through the channel by hydrolysis of ATP (4) (adapted from [61]).
Figure 6. The peptide-binding groove in the SecB tetramer. The left panel shows the
molecular surface of SecB colored according to the underlying atoms: backbone
atoms, white; non-charged polar and charged side-chain atoms, blue; hydrophobic
side-chain atoms, yellow. The right panel shows a ribbon drawing of SecB tetramer
in the same orientation as in the left panel [48].
17
giving SecB time to associate with the unfolded polypeptide [44, 62, 63].
However, SecB binds nonnative proteins promiscuously and the
discrimination between secretory and cytoplasmic proteins is instead
handled by SecA.
There appears to be a substantial functional overlap between SecB and other
cytoplasmic chaperones [38]. In the absence of the preferred chaperone,
nonnative proteins may bind to other components of the cytoplasmic
chaperone pool. The accumulation of secretory proteins in the cytoplasm,
caused by mutations in the secB or secA genes or by the overproduction of
export defective proteins, results in the increased synthesis of σ32-regulated
proteins, such as the chaperones DnaK/DnaJ and GroEL/GroES [64, 65]. It
has been shown that the DnaK/DnaJ chaperone system can facilitate the
export of some SecB dependent proteins in a SecB deficient strain [64]. A
recent study shows that the temperature sensitive and aggregation prone
phenotypes of a strain lacking both TF and DnaK/DnaJ is suppressed by
overproduction of SecB, suggesting that SecB can function as a general
cytoplasmic chaperone [66]. Furthermore, SecB expression is increased in
GroEL and GroES temperature sensitive mutant strains at non-permissive
temperatures [67]. In contrast, the level of SecB expression is unaffected by
mutations in the sec genes [67].
2.1.2 SRP-pathway
Most IMPs are co-translationally targeted to the Sec-translocon via the SRPpathway [36]. The coupling of translation, targeting and insertion of TMs
into the membrane is thought to prevent aggregation of IMPs in the
cytoplasm. Some secretory proteins also appear to be SRP dependent. SRP
docks onto the ribosomal component L23 located near the ribosomal exit
tunnel and binds to hydrophobic signal sequences as they emerge from the
tunnel [68, 69] The E. coli SRP is a ribonucleoprotein complex consisting of
two essential components; the 4.5S RNA and the protein Ffh (Ffh stands for
fiftyfour kDa protein homologue after its eukaryotic homologues). Structural
studies of SRP suggest that signal sequence binding is mediated primarily by
a deep, flexible, hydrophobic groove in Ffh [36]. The flexibility of the
residues lining this groove probably facilitates interaction with a wide range
of unrelated sequences. The 4.5S RNA probably contributes to the groove,
and may help to orient signal sequences by interaction with positively
charged residues present both in the regions flanking TM domains and in the
n-domain of signal sequences [36].
18
The ribosome nascent-chain complex (RNC) is directed to the Sectranslocon via an interaction between SRP and its receptor FtsY [70]. FtsY
exists both free in the cytoplasm and associated with the membrane where it
can interact with the Sec-translocon [71, 72]. Only membrane associated
FtsY promotes the release of SRP from the RNC [73] and the role of
cytoplasmic FtsY is not clear. It has been suggested that FtsY may act as a
receptor for ribosomes, although this proposal is highly controversial [74].
The mechanism behind the transfer of the nascent chain from the SRP-FtsY
complex to the translocation channel is poorly understood. SRP and FtsY are
both GTPases, and GTP binding and hydrolysis play an essential role in the
regulation and directionality of SRP-mediated targeting [75]. SRP, L23 and
the translocon components SecY and SecE can all be cross-linked to nascent
chains at an early stage of translation [24]. It has been shown that SRP and
the Sec-translocon use the same contact areas for their interaction with the
ribosome, suggesting that their binding is mutually exclusive [69, 76]. SecA
does not seem to be required for SRP mediated targeting to the translocon
[77].
Figure 7. Model for SRP mediated targeting to the Sec-translocon. SRP recognizes
the signal sequence or a hydrophobic transmembrane segment emerging from the
ribosome (1). The SRP/ ribosome-nascent-chain complex interact with FtsY (2). The
signal peptide is transferred from the SRP to the Sec-translocon (3). Hydrolysis of
GTP dissociates SRP from FtsY, allowing SRP to recycle into the cytosol (4). The
ribosome nascent chain is now docked to the Sec-translocon and the polypeptide
chain continues to elongate until translation is complete (adapted from [61]).
19
Secretory proteins that normally follow the post-translational targeting
pathway can be co-translationally targeted by SRP if the hydrophobicity of
their signal sequences is artificially increased [78-80]. However, some native
signal sequences of periplasmic proteins, like DsbA, also promote cotranslational targeting [81]. When the mature part of DsbA is fused to the
signal sequence of proteins that are post-translationally targeted, secretion is
severely hampered [82]. It was suggested that co-translational targeting of
DsbA might have evolved because this protein is essential for proper folding
of proteins in the periplasm. In addition, DsbA may be extra sensitive
towards the reducing environment in the cytoplasm [82]. SecM and the
autotransporter Hbp are other examples of secretory proteins that are cotranslationally targeted by SRP in E. coli [83, 84]. These proteins both have
unusually long, but not very hydrophobic, signal sequences. In paper II, we
showed that two outer membrane lipoproteins, Lpp and BRP, are also cotranslationally targeted to the Sec-translocon in an SRP dependent manner.
What distinguishes the signal sequence of co- and post-translationally
targeted secretory proteins? Hydrophobicity appears to be an important
factor. The signal sequences of DsbA and BRP are very hydrophobic.
However, it is clear that other, so far unidentified, features of the signal
peptide also play a role in SRP recognition [82-84].
An average cell is estimated to contain 20-30.000 ribosomes, 300-600 Sectranslocon channels, 10.000 copies of FtsY but only 40 SRPs [85]. In
addition, although both Ffh and the 4.5S RNA are essential, a substantial
fraction of the IMPs appears to insert correctly when cells are strongly
depleted of these components [86] (and Wickström, personal
communication). Taken together, this suggests that the re-cycling of SRP is
extremely rapid and/or that alternative (SRP-independent) targeting
pathways for IMPs exist. For instance, ribosomes may support cotranslational targeting independently of SRP through their affinity for the
Sec-translocon. Another possibility is targeting of mRNA to ribosomes
associated with the Sec-translocon at the inner membrane.
20
2.2 Translocation across, and insertion into, the inner
membrane
The E. coli inner membrane contains two main protein-translocation
machineries; the Sec-translocon and the TAT-translocon. In addition, YidC
functions as an ‘insertase’ for some IMPs.
2.2.1 The Sec-translocon
The SecB and the SRP targeting pathways converge at the Sec-translocon,
which functions as a protein conducting channel in the inner membrane [87].
The E. coli Sec-translocon is formed by the IMPs SecY, SecE and SecG
[24]. Although only a few model proteins have been studied, it is generally
assumed that the Sec-translocon is required for translocation of most
secretory proteins across, and insertion of most IMPs into, the inner
membrane of E. coli. In paper IV, a proteomics approach was used to study
the inner and outer proteomes of cells depleted of the essential Sectranslocon component SecE (see section 4.4).
Translocation of secretory proteins and large periplasmic domains (>60
amino acids) of IMPs through the SecYEG-channel is powered by the
ATPase SecA and the proton motive force (pmf) [24, 57, 88]. One round of
ATP binding and hydrolysis by SecA is estimated to drive 20-30 amino
acids through the membrane [57]. SecA exists both free in the cytoplasm and
at the membrane where it binds with high affinity to the SecYEG translocon
[89]. It has recently been shown that SecA is involved in insertion of single
spanning, Sec-dependent IMPs that lack large periplasmic domains [90].
While SecG is dispensable, SecY, SecE and SecA are all essential for
viability [91]. It has been shown that purified SecY, SecE and SecA
reconstituted into liposomes are sufficient to promote protein translocation in
vitro [92, 93]. The translation of secA is regulated by the periplasmic
secretion monitor SecM in response to the translocation status of the cell
[94]. The secM gene is located upstream of secA in the secM-secA operon.
secM translation is subjected to a transient elongation arrest that is prolonged
when the export efficiency of nacent SecM is reduced. The translation arrest
disrupts the hairpin structure of the ribosome bound secM-secA mRNA. This
leads to exposure of the Shine-Delgarno sequence for secA and
consequently, translation of secA is initiated [94]. Little is known about the
regulation of SecY, SecE and SecG expression.
21
Structure of the SecYEG heterotrimer
The 3.2 Å resolution X-ray structure of the Sec-translocon from the Archaea
Methanococcus jannaschii suggests that the protein conducting channel is
formed by a single SecYEβ heterotrimer (equivalent to SecYEG in E. coli)
[25]. The structure demonstrates that the ten transmembrane segments of
SecY are divided in two halves, one half consists of TM1-5 and the other
half of TM6-10, arranged in a clamshell like conformation. A channel is
located between the two halves and a lateral gate towards the lipid bilayer is
formed at the front of the molecule. The signal sequence is proposed to
interact with TM2 and TM7 at the gate [88]. It has been suggested that the
signal sequence stays in this position during post-translational translocation,
thereby sealing the channel towards the bilayer. In contrast, during the cotranslational insertion of IMPs, continuous opening and closure (i.e.,
‘breathing’) of the lateral gate would expose hydrophobic signals/TM
domains to the bilayer and allow them to equilibrate into the lipid phase [25,
95]. The tendency of TM domains to partition into the membrane can be
predicted from a recently developed biological hydrophobicity scale [96].
The two halves of SecY are clamped together at the back by SecE [25]. It
has been shown that SecY is rapidly degraded by the integral membrane
protease FtsH in E. coli cells depleted of SecE [97]. The SecE ‘clamp’ is
formed by a tilted TM domain and an amphipathic α-helix that lies along the
membrane interface. In most species, including M. jannaschii, SecE is a
single spanning membrane protein with the N-terminus located in the
cytosol. In E. coli, SecE has two additional N-terminal TM domains that are
not essential for its function [98]. In the M. jannaschii structure, the αsubunit is located at the periphery of the complex [25]. This subunit shares
no obvious sequence similarities with E. coli SecG.
Viewed from the side, SecY creates an hourglass shaped funnel with a
central constriction formed by six hydrophobic amino acid residues pointing
towards the interior. This constriction in the middle of the pore is thought to
act as a seal that prevents leakage of ions and small molecules [25, 99]. In
addition, a short α-helix (TM2A) blocks the channel on the extracytoplasmic
and side functions as a ‘plug’ that stabilizes the channel in the closed
conformation [25, 100]. It has been shown that this plug-domain moves
towards the back of the complex during translocation [25, 100]. Plug
displacement may be triggered by binding of a signal sequence in the pocket
formed by TM2b and TM7 [25]. A gating mechanism based on movement of
the plug is supported by cross-linking experiments where cysteines
22
introduced into the plug domain and the TM domain of SecE form
disulphide bonds in vivo [101]. Permanently fixing the channel in this ‘open’
conformation leads to a massive flux of ions and water through the channel
and is lethal to the cell [99, 101]. Furthermore, combinations of different
mutations in the plug-domain and SecE yield synthetic lethal phenotypes
[102].
Figure 8. The structure of the Sec-translocon from M. jannaschii solved by van den
Berg et. al., 2004. Coloring: the α subunit (SecY) TM1–5 in green and TM6–10 in
red; the γ subunit (SecE) in blue; the β subunit in yellow. The ‘plug’ (TM2A) and
the gate helices (TM2/7) are indicated. (a) Front view of the translocon in the plane
of the membrane. (b) Top view of the translocon. The clam-shaped arrangement of
TM1-5 and TM6-10 of the γ subunit is indicated by a red line in the structure [103]
The hydrophobic pore ring in the middle of the channel is too narrow to
allow passage of even an unfolded polypeptide, suggesting that the channel
has to undergo a conformational change to allow translocation [88]. Based
on constraints derived from the X-ray structure, the maximum estimated
diameter of the SecY pore in an open conformation is 20x15 Å. This is
probably too narrow to fit multiple TM domains. It was therefore suggested
that TM domains can not be stored in the channel but rather partition into the
membrane one by one [88]. A larger translocation channel with a diameter of
40-60 Å has been suggested based on permeability experiments with
fluorescent probes [104]. However, it has been argued that this is an
overestimate caused by “snorkeling” of the fluorescent probe and its linker
region [76]. A systematic cross-linking scan suggests that the six TMs
segments of Aquaporin-4 contact Sec61α (the eukaryotic homologue of
SecY) in a strict N- to C-terminal order and that each TM segment leaves the
23
translocon as the next TM segment enters. However, some of the TM
domains exhibited distinct secondary cross-linking patterns suggesting that
they reside within the channel and/or adjacent to the translocon(s) [105].
Oligomeric state of the SecYEG complex and SecA
Several studies using a variety of different experimental techniques have
shown that SecYEG heterotrimers can form higher oligomeric structures
both in the membrane and in detergent solution (reviewed in [89]). Dimeric,
trimeric and tetrameric assemblies of the SecYEG-complex have been
reported and some studies suggest dynamic oligomeric arrangements [89].
One possible function for oligomerization of the SecYEG-complex could be
to provide binding sites for other factors required for translocation and
insertion. It was recently demonstrated that SecA powers the translocation of
a polypeptide chain through a single SecYEG hetrotrimer. However, the
translocation involved oligomers of the heterotrimer, and it was suggested
that SecA interacts through one of its domains with a non-translocating
SecYEG complex while it moves the polypeptide chain through a
neighboring SecYEG complex [95]. A structure of the E. coli translocon in
complex with a translating ribosome has been solved by single-particle cryoEM [76]. When the SecYEβ-complex from M. jannaschii was modeled into
this cryo-EM structure, a dimer in a ‘front-to-front’ conformation gave the
best fit. It is conceivable that two SecYEG heterotrimers in a front-to-front
arrangement could form a larger channel and cooperate in
translocation/insertion [76]. However, the 15 Å resolution of this cryo-EM
structure is too low to unambiguously assign individual TM domains in the
electron density map. A crystal structure of the E. coli SecYEG-complex
determined by EM to a resolution of 8 Å indicates that heterotrimers form
dimers in a ‘back-to-back’ arrangement [106]. Modeling of the SecYEβcomplex from M. jannaschii into this E. coli SecYEG structure shows that
the TM segments in these two structures are arranged in nearly identical
ways [25, 107].
SecA is in a dimer-monomer equilibrium influenced by e.g., ionic strength,
temperature and by the presence of translocating ligands and phospholipids
[108-110]. Most evidence suggests that SecA is predominantly dimeric in
the cytoplasm [89]. Interestingly, comparison of the different SecA crystal
structures indicates that several dimeric associations are possible [111-113].
It is a matter of intense debate whether SecA functions as a monomer or a
dimer at the membrane. Several studies suggest that SecA dissociates into
monomers in the presence of negatively charged lipids alone [109, 110].
24
Furthermore, it has been shown that cross-linking of the SecA dimer can not
be obtained in the presence of liposomes containing the SecYEG-complex
[109]. It was recently demonstrated that nanodiscs containing one SecYEG
heterotrimer embedded in phospholipids promote monomerization of SecA
dimers [114]. It is not clear if nanodiscs containing oligomeric SecYEG
would have the same effect. In contrast, fluorescence resonance energy
transfer experiments indicate that SecA retains its dimeric structure during
translocation [115]. This is supported by experiments showing that artificial
SecA dimers constructed by cross-linking or by fusing two copies of the
SecA gene are functional [116-118]. However, another study demonstrated
that a dimer formed by covalently linked SecA is inactive [119].
Protease accessibility experiments suggested that parts of SecA insert deeply
into the SecYEG-complex during translocation, reaching the other side of
the membrane [58]. This model is at odds with the recent structure of the
Sec-translocon since the channel formed by a single hetrotrimer would be
too small to accommodate these parts of SecA [88]. One controversial
suggestion is that SecA oligomers can insert deeply into the membrane and
form an alternative translocation channel independently of the SecYEGcomplex. This hypothesis is based on the observation that SecA oligomers in
lipid layers form ring-like structures that undergo a structural rearrangement
upon exposure to SecB [120].
2.2.2 The SecDFYajC-complex
The SecDFYajC complex can be co-purified with the SecYEG-complex
[121] and YidC [122, 123]. SecD and SecF both consist of six
transmembrane domains and a large periplasmic domain while YajC is a
single spanning IMP with a C-terminal soluble domain located in the
cytoplasm. It has been estimated that a cell contains ~30 copies of SecD and
SecF [124]. Depletion of SecD and SecF affects protein secretion [125] and
the biogenesis of some IMPs (e.g., [126]). However, the exact role of the
SecDFYajC-complex in protein translocation/insertion is still enigmatic. It
has been proposed that the SecDFYajC-complex is the physical link between
the SecYEG-complex and YidC [123]. It has also been suggested that SecD
is involved in the release of secretory proteins into the periplasm [127] and
that the SecDFYajC complex is involved in the control of the ATPase
activity of SecA [128]. However, this is most likely not the main function of
the complex since Archaea have homologs of SecD/F but not of SecA [129,
25
130]. YajC is not essential for growth and the only clue to its function is the
association with the SecDF complex and the fact that genes encoding these
three proteins are in the same operon.
2.2.3 YidC
The essential IMP YidC is involved in the biogenesis of IMPs, either
independently or in conjunction with the Sec-translocon [24]. YidC is
relatively abundant (~3000 copies per cell) compared to other components
involved in IMP biogenesis [131]. Homologs of YidC are found in the
cytoplasmic membrane of bacteria, in the mitochondrial inner membrane
(Oxa1) and in the thylakoid membrane of chloroplasts (ALB3) [132].
YidC depletion has a strong effect on the insertion of a subset of Secindependent IMPs. Established substrates of this "YidC-only" pathway
include the M13 procoat protein, Pf3 coat protein, subunit c of the F(1)F(o)
ATPase, and MscL [126, 133-139]. These proteins are all relatively small
and consist of only one or two TM domains. Notably, a similar pathway - the
Oxa1 pathway - is operational in yeast mitochondria, which lack equivalents
to the E. coli SRP, the SRP receptor, or any of the Sec-components [24].
A role for YidC in the biogenesis of Sec-dependent IMPs was first suggested
based on crosslinking experiments. These experiments showed that the
hydrophobic signal anchor sequence of the single spanning protein FtsQ
interacts not only with the Sec-translocon, but also with YidC and lipids
during co-translational insertion [140]. Furthermore, a part of YidC was
found to co-purify with the Sec-translocon [140]. YidC depletion only
marginally affects the insertion of Sec-dependent IMPs in vivo [24, 141]. In
spite of this, several in vitro crosslinking studies suggest a general role for
YidC in IMP biogenesis [24]. Collectively, these studies demonstrate that
YidC interacts with TM domains of IMPs during co-translational insertion.
However, the timing of the YidC interaction is different depending on which
model protein is studied. The TM domain of FtsQ was shown to first interact
with SecY and then move to a combined YidC/lipid environment upon
elongation, suggesting that YidC is involved in the lateral transfer of TM
domains from the translocon into the lipid bilayer [142, 143]. In contrast, the
first TM of Leader peptidase (Lep) was crosslinked to YidC at a very early
stage of translation, before crosslinks to SecY were observed, indicating that
YidC may also be involved in the reception of IMPs at the membrane [144].
26
In support of this, subunit II of the cytochrome o oxidase (CyoA) requires
SRP and YidC for insertion of its N-terminal domain consisting of a
cleavable signal sequence, a short periplasmic domain and a transmembrane
segment, while translocation of the large C-terminal periplasmic domain
requires the Sec-translocon and SecA [145].
Recently, it was shown in vitro that the Sec-dependent IMP Lactose
Permease (LacY) does not require YidC for insertion into the inner
membrane. However, YidC is absolutely required for proper folding of LacY
[146]. This suggests that YidC may function both as an ‘insertase’ and a
‘foldase’ for IMPs. In paper II, we showed that YidC is also involved in the
biogenesis of two lipoproteins, Lpp and BRP [147]. Thus, the role of YidC
may not be restricted to the biogenesis of IMPs.
2.2.4 The TAT-pathway
Proteins are exported via the Sec-pathway in an unfolded state. In contrast,
the Twin-Arginine protein Transport (TAT)-pathway is used for posttranslational translocation of folded proteins across the inner membrane of E.
coli [28]. The main components of the TAT-translocon in E. coli are the
IMPs TatA, TatB, and TatC. Homologs of these proteins exist in Archaea,
prokaryotes, chloroplasts and plant mitochondria [148]. Biochemical studies
suggest that TatA, TatB and TatC participate in dynamic, multimeric
membrane complexes that interact transiently with each other and the
substrate during the translocation cycle [149]. The TatBC complex appears
to be involved in the initial recognition and binding of the TAT signal
peptide. Oligomeric TatA, which is thought to form the protein-conducting
channel, is subsequently recruited in a process that is pmf dependent [149].
There is some evidence suggesting that the protein translocation step is also
dependent on pmf [149].
In E. coli, substrates of the TAT-pathway include redox enzymes requiring
cofactor insertion in the cytoplasm, oligomeric proteins that have to be
assemble into a complex prior to export, and proteins whose folding is
incompatible with Sec-export. Approximately 6% of all secretory proteins in
E. coli are predicted to be TAT-dependent [28]. Substrates are directed to the
TAT-translocon by a distinct twin-arginine motif present in the signal
sequence (see section 2.1). A subset of TAT signal peptides exhibits a high
degree of pathway selectivity, while others are promiscuous and can direct
27
proteins either to the Sec- or the TAT-translocon [28]. A number of TATdependent proteins that do not have a signal sequence of their own have been
identified. These proteins form complexes with proteins that contain a TATsignal and get secreted via the TAT-pathway by a “hitch-hiker mechanism”
[150]. In addition, certain IMPs appear to depend on the TAT-translocon for
insertion [149]. These IMPs have a single TM domain at their C-terminal
end [151].
2.3 Maturation and sorting of lipoproteins
Lipoproteins are anchored to the periplasmic surface of the inner or outer
membrane through a lipid moiety attached to a Cysteine (Cys) residue
located at the N-terminus of the mature protein [18]. The E. coli genome is
predicted to encode more then 90 different lipoproteins [18]. Most of these
are located to the outer membrane, while only a few (<10) are localized to
the inner membrane [18]. Lipoproteins are synthesized with an N-terminal
signal sequence that targets them for translocation across the inner
membrane by the Sec-translocon. It is generally assumed that lipoproteins
are targeted to the Sec-translocon in a post-translational fashion. However,
this has not been shown experimentally. In paper II, we studied the targeting
and translocation of two lipoproteins, Lpp and BRP, and found that both
proteins are co-translationally targeted to the Sec-translocon via the SRPpathway [147]. Surprisingly, YidC was shown to play an important role in
the biogenesis of these two lipoproteins [147]. These findings will be
discussed further in section 4.2.
The signal sequence of lipoproteins ends with a so-called ‘lipobox’, a well
conserved motif (-L-(A/S)-(G/A)+C-, where ‘+’ indicates the signal peptide
cleavage site) required for the modification of lipoproteins in the periplasm
[18]. Upon translocation across the inner membrane, the lipobox is
recognized and a diacylglycerol moiety is attached to the Cys residue next to
the cleavage site [18]. The signal peptide is removed by signal peptidase II
(prolipoprotein signal peptidase) and the lipidated Cys residue thereby
becomes the N-terminal residue. Subsequently, the Cys residue is further
modified by acylation of the free α-amino group [18].
28
The sorting of lipoproteins to either the inner or the outer membrane is
determined by the residue immediately after the N-terminal Cys. An
Aspartic acid (Asp) in this position (position 2) retains the lipoprotein at the
inner membrane while any other amino acid targets the protein for
translocation to the outer membrane [18]. This transport is mediated by the
ATP-binding cassette (ABC) transporter LolCDE along with the periplasmic
carrier protein LolA [18]. Lipoproteins that do not have an Asp at position 2
are recognized by LolCDA, which catalyzes the release of the protein from
the inner membrane [18]. The lipoprotein is subsequently delivered to LolA,
which guides it across the periplasm to the lipoprotein specific receptor LolB
at the outer membrane. The lipoprotein is transferred to LolB and
subsequently to the outer membrane [18]. The LolCD catalyzed release of
lipoproteins from the inner membrane is driven by ATP hydrolysis, while
the LolA mediated transport of lipoproteins across the periplasm, their
delivery to LolB and their insertion into the outer membrane does not require
any input of energy [18].
Figure 9. Sorting of lipoproteins in the periplasm. After translocation across the
inner membrane (IM) via the Sec-translocon (Sec), lipoproteins are modified (diacylation, processing by signal peptidase II and acylation) at the periplasmic side of the
(IM). IM lipoproteins have an aspartic acid (D) at position 2 that functions as a Lolavoidance signal and retains the proteins at the IM. Lipoproteins with any other
amino acid (X) at position 2 are sorted to the outer membrane (OM) by the Lol system. The ABC transporter, LolCDE, use ATP to release OM lipoproteins from the
inner membrane. The lipoprotein is bound by LolA, which carries the protein across
the periplasmic space to the OM. At the OM, the lipoprotein is transferred to LolB,
followed by incorporation into the outer membrane.
29
2.4 Insertion of β-barrel proteins into the outer
membrane
β-barrel outer membrane proteins (OMPs) are synthesized with a signal
peptide that directs them to the Sec-translocon. After translocation across the
inner membrane, the signal peptide is cleaved off, and the OMP is released
into the periplasm [152]. Several periplasmic chaperones involved in the
biogenesis of OMPs have been identified, e.g., Skp, SurA, DegP and FkpA
[152]. However, their precise functions and interplay are not yet known. Skp
may function as a periplasmic chaperone that receives OMPs after
translocation across the inner membrane since it selectively interacts with
OMPs [153], and has been crosslinked to the newly translocated PhoE
protein at the periplasmic side of the inner membrane [154]. Skp also binds
to LPS and this binding is required for association of the Skp-OMP complex
to the bilayers [14]. It has been suggested that Skp delivers OMPs to the
outer membrane and that the cycling between the free form of Skp and SkpOMP is modulated by LPS binding. However, cells lacking the skp gene are
still viable and have only mild defects in OMP biogenesis [14]. Mutants in
surA are viable but are affected in the folding of OMP monomers [152].
Double mutations in the skp and surA genes yield synthetic lethality,
indicating that Skp and SurA either act in parallel pathways or sequentially
in the same pathway [152]. DegP is a protease that degrades unfolded or
misfolded proteins in the periplasm. However, DegP also has chaperone
activity [155]. A combination of mutations in surA and degP also results in
synthetic lethality [152]. This may be an effect of increased accumulation of
misfolded OMPs in the periplasm in the absence of the proteolytic activity of
DegP.
A complex consisting of the OMP YeaT and the lipoproteins YfgL, YfiO
and NlpB was recently identified as an insertion-machinery for OMPs into
the outer membrane of E. coli [156]. The gene encoding yfgL was identified
in a genetic screen using ‘chemical conditionality’ to find suppressors of a
leaky mutant (imp4213) [157]. YeaT, YfiO and NlpB were subsequently
identified by co-immunoprecipitation experiments using YfgL as the bait
[156]. Omp85, a homologue of YeaT in Neisseria meningitidis had
previously been identified as an essential protein involved in the biogenesis
of OMPs [158]. In E. coli, YeaT and YfiO are essential for viability, while
YfgL and NlpB are not [14]. However, mutations in the yfgL and nlpB
30
genes cause outer membrane defects [14]. The precise function of the
individual components in the OMP insertion machinery is not yet clear.
Figure 10. OMPs are translocated
across the inner membrane (IM) by the
Sec-translocon (Sec). In the periplasm,
OMPs are transported to the outer
membrane (OM) by an unknown
mechanism, although the periplasmic
chaperones Skp, DegP and SurA have
been implicated. At the OM, the YaeT
/ YfgL / YfiO / NlpB complex
assembles OMPs by an unknown
mechanism (adapted from [14]).
31
3. Objectives
Approximately 10% of the ORFs in the E. coli genome encodes secretory
proteins and 20% encodes IMPs [8]. Secretory proteins and IMPs depend on
a protein sorting machineries to reach their correct cellular compartment.
Using various genetic, biochemical and structural approaches much has been
learned about the components involved in these processes. However, an
extremely limited set of model proteins has been used to study substratepathway relationships. It is not clear how representative the biogenesis
requirements for these model proteins are. In fact, previous work from our
laboratory and others suggest that protein biogenesis is an unexpectedly
versatile process and what is first viewed as an exception can very well turn
out to be the general rule [135, 159]. Therefore, the main objective of all the
papers presented in this thesis has been to extend our knowledge of
substrate-pathway relationships by determining the biogenesis requirements
of more proteins. In papers I and II, this was done using focused approaches.
Selected model proteins (lipoproteins and putative OMPs) were expressed
from plasmids and their targeting and translocation were analysed in vitro by
crosslinking experiments and/or in vivo by pulse-chase analysis in different
E. coli mutant strains. In papers III and IV, the aim was to study protein
biogenesis using a ‘global’ approach. Cells depleted of SecB and SecE were
compared to control cells by sub-cellular fractionation and comparative
proteomics. In contrast to more focused approaches, a global approach does
not depend on pre-selection of model substrates, or plasmid based
ovexpression that might induce protein aggregation, saturate a pathway
and/or force proteins to use a pathway that is not used under normal
conditions.
32
4. Summary of papers
4.1 Paper I - New Escherichia coli outer membrane
proteins identified through prediction and experimental
verification
It is often difficult to predict OMPs from amino acid sequence. Compared to
the transmembrane segments of α-helical IMPs, the transmembrane βstrands of OMPs are relatively short and much less hydrophobic [8]. It has
been estimated that the E. coli genome encodes approximately 75 OMPs
[160], but only 58 are currently listed in the EcoCyc database [8]. Recent
proteomics studies (e.g., paper III) have experimentally verified the
localization of several outer membrane proteins. However, not all proteins
are expressed under the conditions used in these studies. Therefore, a
complementary approach was used in paper I. Eight proteins predicted by
the Hunter predictor as putative OMPs [160], were expressed from a plasmid
and their localization was experimentally determined. Plasmid based
expression can lead to formation of protein aggregates that co-sediment with
outer membranes during density centrifugation. Therefore, a method based
on urea extraction was developed to distinguish between proteins localized
in the outer membrane and/or aggregates. Proteins that are properly
integrated into the outer membrane are resistant to urea treatment, while
proteins in aggregates are dissolved. Thus, upon urea treatment of the outer
membrane fraction, properly integrated proteins can be pelleted by
centrifugation, while aggregated proteins are found in the supernatant. Five
proteins (YftM, YaiO, YfaZ, CsgF, and YliL) were shown to localize to the
outer membrane, confirming the original prediction. Two proteins (YhjY and
YagZ) were susceptible to urea extraction and their localization could
therefore not be verified. Our data suggest that one of the proteins, YfaL, is
an autotransporter.
The SecA and SecB dependencies of all proteins, except YfaL, were tested
in vivo by pulse-chase analysis. All seven proteins were shown to depend on
SecA for translocation. Translocation of YftM, YliL, YaiO, YhjY, YfaZ and
CsgF was hampered in the absence of SecB. The degree of SecB dependence
33
varied between the different proteins. Notably, the translocation of some of
the proteins tested was slow, also when SecB was expressed. This fits well
with the observation that a fraction of the expressed proteins aggregate at
37ºC. Indeed, the proteins with the slowest translocation (YliL, YhjY, and
YagZ) were the ones that were most sensitive to urea extraction after
expression at 37ºC.
4.2 Paper II - Targeting and translocation of two
lipoproteins in Escherichia coli via the SRP/Sec/YidC
pathway
Lipoproteins are synthesized with an N-terminal signal sequence that targets
them for translocation across the inner membrane. Although little
experimental evidence exists, it is generally assumed that lipoproteins are
targeted to the Sec-translocon in a post-translational fashion. In paper II, a
combined in vitro crosslinking and in vivo depletion study was carried out to
determine the targeting and translocation requirements of two lipoproteins Lpp and BRP. Surprisingly, both proteins are targeted to the inner membrane
via the SRP-dependent co-translational targeting pathway. Depletion of both
the 4.5S RNA and Ffh impaired translocation. The signal sequences of Lpp
and BRP were efficiently crosslinked to Ffh and L23/L29, ribosomal
components located at the exit tunnel. In vivo depletion experiments
indicated that Lpp and BRP do not require SecB for efficient translocation.
However, it should be noted that the precursor form of endogenously
expressed Lpp was identified in aggregates isolated from a secB null mutant
strain (paper III). Thus, it is possible that a fraction of the newly synthesized
Lpp can make use of both the SRP and the SecB targeting pathways.
As expected, translocation of both Lpp and BRP was dependent on the Sectranslocon and SecA. However, our data showed that YidC also plays an
important role in the biogenesis of Lpp and BRP. The unprocessed form of
both proteins accumulated when cells were depleted of YidC and
furthermore, the signal sequences of Lpp and BRP could be crosslinked to
YidC. Thus, YidC appears to be involved in the translocation of Lpp and
BRP. One possibility is that YidC facilitate the lateral transfer of
lipoproteins into the inner membrane via their signal sequences. To our
knowledge, this is the first time that it has been shown that YidC can interact
with secretory proteins and plays a role in their biogenesis.
34
4.3 Paper III - Defining the role of the Escherichia coli
chaperone SecB using comparative proteomics
In E. coli, most secretory proteins are believed to cross the inner membrane
via the Sec-translocon in a post-translational manner. A common assumption
is that the targeting of these proteins is facilitated by the cytoplasmic
chaperone SecB. However, SecB dependence has only been shown for a few
selected model proteins, including the ones used in paper I. Several other
secretory proteins, including the two lipoproteins studied in paper II, appear
to be efficiently translocated in the absence of SecB. Although a SecB
binding motif has been identified, it can not be used to predict whether SecB
is required for efficient translocation since it is commonly found in both
SecB dependent and independent secretory proteins, as well as in
cytoplasmic proteins. In paper III, we chose a comparative proteomics
approach to characterize a secB null mutant strain and identify novel
substrates of the SecB-pathway. The secB null mutant strain was compared
to a control strain that expresses normal levels of SecB. Whole cell lysates,
protein aggregates and outer membrane fractions from these two strains were
isolated and analysed using one-dimensional electrophoresis (1DE) and twodimensional electrophoresis (2DE) in combination with mass spectrometry
and immunoblotting.
Comparative 2DE analysis showed that levels of the processed forms of
several secretory proteins were reduced in the secB null strain, suggesting
that these proteins depend on SecB for efficient translocation. The analysis
also showed that the levels of the σ32- regulated cytoplasmic chaperones
DnaK, GroEL/ES, ClpB, IbpA/B, and HslU were increased in the secB null
strain. This suggested that miss-targeting of secretory proteins in the absence
of SecB may lead to protein aggregation in the cytoplasm. Indeed, protein
aggregates containing mostly secretory proteins could be isolated from the
secB null mutant, but were virtually absent in the control strain. It should be
noted that the amount of protein found in aggregates corresponded to only
0.5% of the total protein content of the cell. Interestingly, pulse-chase
analysis showed that cytoplasmic OmpA was removed from the aggregate
fraction over time, indicating that miss-targeted OmpA is either degraded or
reactivated for translocation.
The outer membrane proteome consists of OMPs and lipoproteins, which are
all potential substrates of the SecB targeting pathway. To study the SecB
dependence of the outer membrane proteome, we developed a protocol for
35
2DE analysis of membranes isolated from [35S]methionine-labeled cells.
Proteins were visualized both by protein staining and by phosphorimaging.
This allowed us to study protein steady state levels and outer membrane
protein insertion kinetics on the same set of gels. The steady state levels of
the iron-siderophore transporter FhuA, the ferrichrome-iron receptor FhuE,
and peptidoglycan-associated lipoprotein (Pal) were decreased 2-fold in the
absence of SecB. In contrast, the level of the ferrienterobactin receptor FepA
was doubled, possibly compensating for the decrease of the FhuA and FhuE
levels. Steady state levels of all the other outer membrane proteins identified
were unchanged, suggesting that SecB is not required for their targeting to
the Sec-translocon. Importantly, the analysis of [35S]methionine-labeled
proteins revealed that many OMPs, such as BtuB, FhuA, FhuE, FadL, OmpT,
OmpX, and TolC need more time to reach the outer membrane in the secB
null mutant compared to the control. This demonstrates that SecB is not
strictly required but improves the secretion efficiency of these proteins.
Collectively, the analysis of whole cell lysates, aggregates and outer
membranes from the secB null mutant and the control strain lead to the
identification of 19 secretory proteins that were affected in the absence of
SecB. We hypothesized that these proteins are substrates of the SecBtargeting pathway. To test if this is indeed the case, the SecB dependence of
12 of these proteins was tested by classical pulse-chase analysis. Strikingly,
the translocation of all 12 proteins was hampered in the secB null mutant
strain.
4.4 Paper IV - Effects of SecE depletion on the inner
and outer membrane proteomes of E. coli
The Sec-translocon is a protein-conducting channel involved in the
translocation of secretory proteins across, and the insertion of IMPs into, the
inner membrane of E coli. Sec-translocon requirements are usually studied
using focused approaches and a very limited set of model proteins. To
characterize the Sec-translocon dependence of secretory and IMPs in a
global way, we performed a comparative sub-proteome analysis of SecE
depleted and non-depleted cells. Both the steady-state proteomes and the
proteome dynamics were evaluated using comparative 1DE and 2DE,
followed by mass spectrometry based protein identification and extensive
immunoblotting. This analysis resulted in several testable hypotheses and
new substrates to further discover guiding principles for protein translocation
36
and insertion in E. coli.
Depletion of SecE resulted in a 90% (10-fold) reduction of SecE and a 50%70% (2-fold) reduction of the SecY,G translocon components. Onedimensional (1D) Blue-Native (BN) polyacrylamide gel-electrophoresis
(PAGE) combined with immunoblotting showed that the level of the
SecYEG-complex was strongly reduced upon SecE-depletion. Furthermore,
the total level of SecA was increased 1.7 fold, consistent with insufficient
Sec-translocon capacity. SecE depletion did not affect the SecB and SRPtargeting pathway capacity.
Analysis of whole cell lysates by 2DE and immunoblotting showed that the
levels of the mature forms of several secretory proteins were reduced, while
precursors accumulated in the cytoplasm upon SecE depletion. Protein
aggregates containing secretory proteins were isolated from SecE depleted
cells. The cytoplasmic σ32-stress response was activated in these cells,
leading to increased levels of the chaperones DnaK, GroEL, GroES, ClpB
and IbpA/B.
To study the effect of SecE depletion on the insertion and composition of the
membrane proteomes, inner and outer membrane fractions were isolated
from cells labeled with [35S]methionine. The outer membrane proteome was
analysed by 2DE and the inner membrane proteome was analysed by twodimensional BN-PAGE. The insertion and steady state levels of most outer
membrane proteins were reduced upon SecE depletion. However, substantial
translocation activity was still observed, indicating that translocation across
the inner membrane is hampered but not abolished. The components of the
outer membrane proteome were differentially affected by the depletion of
SecE. OmpA, FadL, and Pal were unaffected or increased in the outer
membrane upon SecE depletion, while most other OMPs and outer
membrane lipoproteins were reduced to different extents.
Interestingly, the depletion has a differential effect on components of the
inner membrane proteome. Steady state levels and insertion of
approximately 30 IMPs were reduced, while a similar number of proteins
were not affected, or even increased, in the membrane upon SecE depletion.
The IMPs that were unaffected or increased lacked large translocated
domains, and/or consisted of only one or two transmembrane segments.
Interestingly, all established substrates of the Sec-translocon
independent/YidC dependent pathway share similar features. Thus, it is
37
tempting to speculate that the proteins that are unaffected or increased in the
membrane of SecE depleted cells are substrates of the YidC only pathway.
38
5. Concluding remarks
The work presented in this thesis has brought several unexpected findings
and new questions that are burning to be answered. In paper II, both the SRP
and YidC were found to be involved in the biogenesis of the lipoproteins
Lpp and BRP. Although it is commonly assumed that secretory proteins including lipoproteins - follow the SecB-pathway to the Sec-translocon,
translocation of these two lipoproteins was not detectably affected by the
absence of SecB. To date, it is not clear whether SRP and YidC have a
general role in the biogenesis of lipoproteins. Therefore, targeting and
translocation of more lipoproteins have to be studied. In paper III, we found
that the steady state composition of the outer membrane proteome was
hardly affected in the absence of SecB. However, using a combination of
global and focused methods, several novel SecB-substrates were identified
(papers I and III). Although not absolutely required, SecB was shown to
facilitate translocation of these proteins. In paper IV, we used a proteomics
approach to study the effects of SecE depletion. Given the central role of the
Sec-translocon in the biogenesis of OMPs, IMPs and lipoproteins, SecE
depletion was expected to have large effects on the outer and inner
membrane proteomes. The outer membrane proteome was indeed strongly
affected upon SecE depletion. However, the effect was differential,
suggesting that some outer membrane proteins may have superior access to
the Sec-translocon. The mechanism behind this observation needs to be
studied further. In addition, the levels of a many IMPs were unaffected or
even increased in the inner membrane of SecE depleted cells. It is tempting
to speculate that these proteins are substrates of a Sec-translocon
independent pathway, e.g., the YidC pathway. If this is indeed the case, Sectranslocon independent insertion of IMPs is far more common in E. coli than
thought previously. Notably, papers III and IV show that proteomics
approaches have great potential helping us to further our knowledge of
protein targeting, translocation and insertion. Luckily for newcomers in the
field, there is still plenty of work to be done!
39
6. Sammanfattning på svenska – Hur hittar
proteiner rätt i en cell?
Inuti alla celler tillverkas en mängd olika proteiner med varierande
funktioner. Exempel på olika typer av proteiner är enzymer, hormoner,
antikroppar och muskelfibrer. När ett protein har bildats måste det sorteras
och transporteras till rätt ställe i eller utanför cellen för att fungera. Det kan
till och med vara skadligt för cellen om denna sortering inte fungerar
eftersom proteiner till exempel kan aggregera (klumpa ihop sig) om de
hamnar på fel plats. Den här avhandlingen syftar till att klargöra hur
proteiner sorteras och transporteras i en cell. Bakterien E. coli har använts
som modellorganism, men många av de principer som styr
proteinsorteringen har bevarats under evolutionens gång. Detta innebär att
upptäckter som görs i E. coli bakterier ofta kan överföras till specialiserade
celler i högre organismer och vice versa.
Proteiner kan delas in i två grupper beroende på vilka lösningsegenskaper de
har. Globulära proteiner är vattenlösliga medan membranproteiner inte är
det. Membranproteiner är istället lösliga i den tunna film av lipider (fett) som
bildar stommen i alla cellmembran. E. coli bakterien kan delas upp i fyra
avdelningar som alla innehåller proteiner: cytoplasman och periplasman
som innehåller globulära proteiner, och det inre och det yttre membranet
som innehåller membranproteiner omgivna av lipider. Cytoplasman kapslas
in av innermembranet, som i sin tur omges av yttermembranet. Periplasman
finns i hålrummet mellan de båda membranen (för bild av E. coli bakteriens
struktur, se Figur 3, sida 17). I cytoplasman finns bakteriens DNA som
innehåller koden (ritningen) för hur alla proteiner ska se ut. När ett visst
protein ska bildas så tillverkas först en sorts arbetskopia, ett så kallat mRNA,
av den lilla del av DNA:t som innehåller koden för just det proteinet.
Därefter tillverkas proteinet av så kallade ribosomer, som kan läsa av koden i
mRNA:t. I E. coli bakterien sker allt detta i cytoplasman. Proteiner som hör
hemma i periplasman eller i yttermembranet måste därför transporteras till
och över innermembranet för att nå sina respektive avdelningar i cellen.
Proteiner som hör hemma i innermembranet måste transporteras till
40
membranet för att sedan sättas in i det. Enligt rådande modeller sker i de
flesta fall både insättning i och transport över innermembranet med hjälp av
det så kallade Sec-translokonet som bildar en kanal inuti membranet. För
att nybildade proteiner ska nå fram till Sec-translokonet innehåller
cytoplasman olika sorteringsfaktorer. Dessa arbetar enligt samma princip
som en brevbärare: de känner igen specifika egenskaper (adresslappar) hos
nybildade proteiner och levererar dem till Sec-translokonet.
I den här avhandlingen har vi undersökt en av E. coli bakteriens sorterings
faktorer, proteinet SecB. I artiklarna I, II och III har vi använt olika metoder
för att testa vilka proteiner som behöver SecB för att nå Sec-translokonet och
transporteras över membranet. Vi identifierade en rad olika proteiner i
periplasman och yttermembranet som använder SecB. Resultaten i artikel III
visar dock att även om proteintransporten är fördröjd i muterade E. coli
celler som helt saknar SecB, så når de flesta proteiner som hör hemma i
yttermembranet faktiskt fram ändå.
I artikel IV undersökte vi vilka proteiner som behöver, respektive inte
behöver, Sec-translonet genom att testa vilka proteiner som påverkas i
muterade E. coli celler som har mindre än en tiondel så många Sectranslokon som vanliga E. coli celler. Vi upptäckte att nivån av många
proteiner i innermembranet inte påverkades, eller till och med ökade, när
mängden Sec-translokon minskades. En möjlig förklaring till detta är att
vissa proteiner sätts in i membranet med hjälp av något annat
insättningsystem än Sec-translokonet. En annan möjlighet är att vissa
proteiner har egenskaper som ger dem förtur till Sec-translokonet, och därför
inte påverkas när mängden Sec-translokon i cellen minskas. Sammantaget
visar resultaten att insättningen av proteiner är mer komplicerad än man
tidigare trott, och att de rådande modellerna därför måste omprövas. Mycket
arbete återstår för att nå full förståelse för hur proteiner sorteras och
transporteras.
41
7. Acknowledgements
Jan-Willem, it all started with an email from Katmandu… Since then, we
have discovered the “do’s and don’ts” of proteomics together. I think we can
both agree it has been a bumpy, but rewarding journey! Thank you for your
patience, spelling corrections and dedication to science. A special thanks for
the colorful metaphors - those poor ants!!!
Klaas, thank you for welcoming me into your lab, for introducing me to
proteomics and for stimulating and rewarding discussions. I had a great time
playing with your nice machines! To all members of the KvW lab, thanks
for making me feel at home, for being so helpful and nice to work with.
Speciellt tack till Jimmy, det skulle inte gått utan dig!
Gunnar, thanks for introducing me to the wonderful world of membrane
proteins! It has been great working in the “GvH cluster”.
Stefan N, thank you for being such an excellent boss of the DBB PhD
program. To everyone at DBB, tank you for making the department such a
happy place!
Linda, it was excellent working with you and I still miss you in the lab.
Thanks for setting up the lab, for “breaking in” JW and most of all - for
being such a friend. David D, thank you for the good times, and for helping
me with the cloning for the SecB project. Samuel, I still wonder if I can’t
take you with me to my next position! How will I ever manage without
you?! Thank you for all the help (including life saving chocolate) and for
breaking more glass-ware than me. David W, thank you for all the laughs
and weird discussions in the lab. It has been a pure pleasure working with
you! However, I will never do your laundry again. Mirjam, soon it will be
your job to “hålla killarna på mattan”. I trained you well, don’t let me down!
Thanks for all the nice chats, and for eating more chocolate then me ;-).
Joy and Dan, thank you both for organizing journal clubs etc! It has been
great working with you. I especially appreciate both of you sharing your
‘senior’ experiences and opinions on science and career with me. Carolina
and Filippa, thank you for keeping track of birthdays etc! Ing-Marie, thanks
for taking care of so many things…freezers, gel dryers etc. Kalle S, thanks
42
for help with the SecE westerns! Kalle E, for organizing cake-clubs, thanks!
Marie, coffee breaks are more fun when you are around! Mikaela, thank
you for calling me organized! It has been great fun working with you. Hope
to see you on a rock or a wall in the future. Mirjam L, I truly enjoy the
heated discussions during coffee breaks, it is great to have you here!
Susanna, for your contagious enthusiasm, thanks! Tara, we shared misery
at KÖL and good times at Corsica. Thank you for being a good friend. Good
luck in the new lab.
Eva Severinsson och alla i Biomedicinska Forskarskolan 99/00, tack för
ett fantastiskt år! Speciellt tack till Hanna, Julian, Maria K, Maria S och
Martin G, Martin L och Camilla, och Sofia för trevliga middagar och
fester. Maria E, Henke och Holger (coolaste killen!), bonus-tack för alla
trevliga midsommarfester på Gräskö. Maria, jag saknar våra vandrings
strapatser, hoppas de blir fler!
Lisa and Andrea, my dear Italians who saved my life and sanity in not so
gorgeous Ithaca! You are both (yes, you too Andrea) fantastic people and I
feel blessed to have you as friends.
Jenny, Linnea och Tesan (alias brudarna babes), som tur är har vi inte bara
ett glödande intresse för böcker gemensamt... Tack för en underbar vänskap
och för alla roliga dagar, kvällar och nätter.
Frida och Kristina, tack för er vänskap som betyder så mycket för mig.
Tack för att ni finns och för att ni har så stort tålamod. Vi måste ses oftare,
puss!
Mattias, för att du säkrar mig när jag faller, och för alla berg vi kommer att
bestiga tillsammans - tack. Du gör mig lycklig.
Pappa, Marie, Vincent, Leo och Adde, det blir alldeles för sällan men det
är alltid kul när vi väl ses! Tack för allt stöd och för roliga studer på Öland, i
skidbacken och vid middagsbordet.
Carl-Fredrik, du är min idol, det du inte kan fixa är inte värt att fixa! Tack
för att du är - och alltid har varit - en sådan fantastisk storebror. Liza, tack
för alla gourmé middagar och trevliga stunder. Emma, Julia och Marcus, ni
är världens sötaste och jag blir så glad av att vara med er!
Mamma, du stöttar mig i vått och torrt. Stort tack för all din kärlek,
uppmuntran och omsorg.
43
8. References
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
44
Engelman, D.M., (2005) Membranes are more mosaic than fluid.
Nature 438: 578-580.
White, S.H., (2003) Translocons, thermodynamics, and the folding
of membrane proteins. FEBS Lett 555: 116-121.
Mouritsen, O., Life- As a matter of fat, the emerging science of
lipidomics. The frontier collection, ed. D. Dragoman, Dragoman,
M., Elitzur, A.C., Silverman, M.P., Tuszymski, J., Zeh, H.D. . 2005,
Heidelberg: Springer.
Singer, S.J. and G.L. Nicolson, (1972) The fluid mosaic model of
the structure of cell membranes. Science 175: 720-731.
Lee, A.G., (2004) How lipids affect the activities of integral
membrane proteins. Biochim Biophys Acta 1666: 62-87.
Killian, J.A. and G. von Heijne, (2000) How proteins adapt to a
membrane-water interface. Trends Biochem Sci 25: 429-434.
Ulmschneider, M.B., M.S. Sansom, and A. Di Nola, (2005)
Properties of integral membrane protein structures: derivation of an
implicit membrane potential. Proteins 59: 252-265.
Elofsson, A. and G. Heijne, (2007) Membrane Protein Structure:
Prediction versus Reality. Annu Rev Biochem 76: 125-140.
von Heijne, G., (1986) The distribution of positively charged
residues in bacterial inner membrane proteins correlates with the
trans-membrane topology. EMBO J 5: 3021-3027.
Nilsson, I. and G. von Heijne, (1990) Fine-tuning the topology of a
polytopic membrane protein: role of positively and negatively
charged amino acids. Cell 62: 1135-1141.
Nilsson, J., B. Persson, and G. von Heijne, (2005) Comparative
analysis of amino acid distributions in integral membrane proteins
from 107 genomes. Proteins 60: 606-616.
Blattner, F.R., G. Plunkett, C.A. Bloch, N.T. Perna, V. Burland, M.
Riley, J. Collado-Vides, J.D. Glasner, C.K. Rode, G.F. Mayhew, J.
Gregor, N.W. Davis, H.A. Kirkpatrick, M.A. Goeden, D.J. Rose, B.
Mau, and Y. Shao, (1997) The complete genome sequence of
Escherichia coli K-12. Science 277: 1453-1462.
Lugtenberg, E.J. and R. Peters, (1976) Distribution of lipids in
cytoplasmic and outer membranes of Escherichia coli K12. Biochim
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
Biophys Acta 441: 38-47.
Ruiz, N., D. Kahne, and T.J. Silhavy, (2006) Advances in
understanding bacterial outer-membrane biogenesis. Nat Rev
Microbiol 4: 57-66.
Zimmerman, S.B. and S.O. Trach, (1991) Estimation of
macromolecule concentrations and excluded volume effects for the
cytoplasm of Escherichia coli. J Mol Biol 222: 599-620.
Krogh, A., B. Larsson, G. von Heijne, and E. Sonnhammer, (2001)
Predicting transmembrane protein topology with a hidden Markov
model. Application to complete genomes. J Mol Biol 305: 567-580.
Messens, J. and J.F. Collet, (2006) Pathways of disulfide bond
formation in Escherichia coli. Int J Biochem Cell Biol 38: 10501062.
Tokuda, H. and S. Matsuyama, (2004) Sorting of lipoproteins to the
outer membrane in E. coli. Biochim Biophys Acta 1694: IN1-9.
Sugawara, E. and H. Nikaido, (1992) Pore-forming activity of
OmpA protein of Escherichia coli. J Biol Chem 267: 2507-2511.
Dong, C., K. Beis, J. Nesper, A.L. Brunkan-Lamontagne, B.R.
Clarke, C. Whitfield, and J.H. Naismith, (2006) Wza the translocon
for E. coli capsular polysaccharides defines a new class of
membrane protein. Nature 444: 226-229.
Nandakumar, M.P., A. Cheung, and M.R. Marten, (2006) Proteomic
analysis of extracellular proteins from Escherichia coli W3110. J
Proteome Res 5: 1155-1161.
McBroom, A.J. and M.J. Kuehn, (2007) Release of outer membrane
vesicles by Gram-negative bacteria is a novel envelope stress
response. Mol Microbiol 63: 545-558.
Blobel, G. and B. Dobberstein, (1975) Transfer of proteins across
membranes. I. Presence of proteolytically processed and
unprocessed nascent immunoglobulin light chains on membranebound ribosomes of murine myeloma. J Cell Biol 67: 835-851.
Luirink, J., G. von Heijne, E. Houben, and J.W. de Gier, (2005)
Biogenesis of inner membrane proteins in Escherichia coli. Annu
Rev Microbiol 59: 329-355.
Van den Berg, B., W.M. Clemons, Jr., I. Collinson, Y. Modis, E.
Hartmann, S.C. Harrison, and T.A. Rapoport, (2004) X-ray structure
of a protein-conducting channel. Nature 427: 36-44.
Kiefer, D. and A. Kuhn, (2007) YidC as an essential and
multifunctional component in membrane protein assembly. Int Rev
Cytol 259: 113-138.
Facey, S.J. and A. Kuhn, (2003) The sensor protein KdpD inserts
into the Escherichia coli membrane independent of the Sec
translocase and YidC. Eur J Biochem 270: 1724-1734.
Lee, P.A., D. Tullman-Ercek, and G. Georgiou, (2006) The bacterial
twin-arginine translocation pathway. Annu Rev Microbiol 60: 373395.
Dalbey, von Heijne., Protein Targeting Transport and
45
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
46
Translocation, 2002, Elsevier Science.
von Heijne, G., (1983) Patterns of amino acids near signal-sequence
cleavage sites. Eur J Biochem 133: 17-21.
von Heijne, G., (1985) Signal sequences. The limits of variation. J
Mol Biol 184: 99-105.
Berks, B.C., (1996) A common export pathway for proteins binding
complex redox cofactors? Mol Microbiol 22: 393-404.
Cristóbal, S., J.W. de Gier, H. Nielsen, and G. von Heijne, (1999)
Competition between Sec- and TAT-dependent protein translocation
in Escherichia coli. EMBO J 18: 2982-2990.
Bogsch, E., S. Brink, and C. Robinson, (1997) Pathway specificity
for a delta pH-dependent precursor thylakoid lumen protein is
governed by a 'Sec-avoidance' motif in the transfer peptide and a
'Sec-incompatible' mature protein. Embo J 16: 3851-3859.
von Heijne, G., (1997) Getting greasy: How transmembrane
polypeptide segments integrate into the lipid bilayer. Mol Microbiol
24: 249-253.
Luirink, J. and I. Sinning, (2004) SRP-mediated protein targeting:
structure and function revisited. Biochim Biophys Acta 1694: 17-35.
Lee, H.C. and H.D. Bernstein, (2002) Trigger factor retards protein
export in Escherichia coli. J Biol Chem 277: 43527-43535.
Randall, L.L. and S.J. Hardy, (2002) SecB, one small chaperone in
the complex milieu of the cell. Cell Mol Life Sci 59: 1617-1623.
Dekker, C., B. de Kruijff, and P. Gros, (2003) Crystal structure of
SecB from Escherichia coli. J Struct Biol 144: 313-319.
Xu, Z., J.D. Knafels, and K. Yoshino, (2000) Crystal structure of the
bacterial protein export chaperone secB. Nat Struct Biol 7: 11721177.
Knoblauch, N.T., S. Rudiger, H.J. Schonfeld, A.J. Driessen, J.
Schneider-Mergener, and B. Bukau, (1999) Substrate specificity of
the SecB chaperone. J Biol Chem 274: 34219-34225.
Powers, E.L. and L.L. Randall, (1995) Export of periplasmic
galactose-binding protein in Escherichia coli depends on the
chaperone SecB. J Bacteriol 177: 1906-1907.
Gannon, P.M., P. Li, and C.A. Kumamoto, (1989) The mature
portion of Escherichia coli maltose-binding protein (MBP)
determines the dependence of MBP on SecB for export. J Bacteriol
171: 813-818.
Randall, L.L., T.B. Topping, and S. Hardy, (1990) No Specific
Recognition of Leader Peptide by SecB, a Chaperone Involved in
Protein Export. Science 248: 860-863.
Topping, T.B. and L.L. Randall, (1994) Determination of the
binding frame within a physiological ligand for the chaperone SecB.
Prot Sci 3: 730-736.
Khisty, V.J., G.R. Munske, and L.L. Randall, (1995) Mapping of the
binding frame for the chaperone SecB within a natural ligand,
galactose-binding protein. J Biol Chem 270: 25920-25927.
47.
48.
49.
50.
51.
52.
53.
54.
55.
56.
57.
58.
59.
Smith, V.F., S.J.S. Hardy, and L.L. Randall, (1997) Determination
of the binding frame of the chaperone SecB within the physiological
ligand oligopeptide-binding protein. Protein Sci 6: 1746-1755.
Zhou, J. and Z. Xu, (2005) The structural view of bacterial
translocation-specific chaperone SecB: implications for function.
Mol Microbiol 58: 349-357.
Watanabe, M. and G. Blobel, (1989) SecB functions as a cytosolic
signal recognition factor for protein export in E. coli. Cell 58: 695705.
Lecker, S., R. Lill, T. Ziegelhoffer, C. Georgopoulos, P.J. Bassford,
Jr., C.A. Kumamoto, and W. Wickner, (1989) Three pure chaperone
proteins of Escherichia coli -SecB, trigger factor and GroEL-form
soluble complexes with precursor proteins in vitro. EMBO J 8:
2703-2709.
den Blaauwen, T., E. Terpetschnig, J.R. Lakowicz, and A.J.
Driessen, (1997) Interaction of SecB with soluble SecA. FEBS Lett
416: 35-38.
Fekkes, P., C. van der Does, and A.J. Driessen, (1997) The
molecular chaperone SecB is released from the carboxy-terminus of
SecA during initiation of precursor protein translocation. EMBO J
16: 6105-6113.
Hartl, F.U., S. Lecker, E. Schiebel, J.P. Hendrick, and W. Wickner,
(1990) The Binding Cascade of SecB to SecA to SecY/E Mediates
Preprotein Targeting to the E. Coli Plasma Membrane. Cell 63: 269279.
Woodbury, R.L., T.B. Topping, D.L. Diamond, D. Suciu, C.A.
Kumamoto, S.J. Hardy, and L.L. Randall, (2000) Complexes
between protein export chaperone SecB and SecA. Evidence for
separate sites on SecA providing binding energy and regulatory
interactions. J Biol Chem 275: 24191-24198.
Randall, L.L., J.M. Crane, G. Liu, and S.J. Hardy, (2004) Sites of
interaction between SecA and the chaperone SecB, two proteins
involved in export. Protein Sci 13: 1124-1133.
Randall, L.L., J.M. Crane, A.A. Lilly, G. Liu, C. Mao, C.N. Patel,
and S.J. Hardy, (2005) Asymmetric binding between SecA and SecB
two symmetric proteins: implications for function in export. J Mol
Biol 348: 479-489.
Schiebel, E., A.J. Driessen, F.U. Hartl, and W. Wickner, (1991)
Delta mu H+ and ATP function at different steps of the catalytic
cycle of preprotein translocase. Cell 64: 927-939.
Economou, A. and W. Wickner, (1994) SecA promotes preprotein
translocation by undergoing ATP-driven cycles of membrane
insertion and deinsertion. Cell 78: 835-843.
Economou, A., J.A. Pogliano, J. Beckwith, D.B. Oliver, and W.
Wickner, (1995) SecA membrane cycling at SecYEG is driven by
distinct ATP binding and hydrolysis events and is regulated by SecD
and SecF. Cell 83: 1171-1181.
47
60.
61.
62.
63.
64.
65.
66.
67.
68.
69.
70.
71.
72.
73.
48
Miller, A., L. Wang, and D.A. Kendall, (2002) SecB modulates the
nucleotide-bound state of SecA and stimulates ATPase activity.
Biochemistry 41: 5325-5332.
Driessen, A.J.M., E.H. Manting, and C. van der Does, (2001) The
structural basis of protein targeting and translocation in bacteria. Nat
Struct Biol 8: 492-498.
Park, S., G. Liu, T.B. Topping, W.H. Cover, and L.L. Randall,
(1988) Modulation of folding pathways of exported proteins by the
leader sequence. Science 239: 1033-1035.
Liu, G.P., T.B. Topping, and L.L. Randall, (1989) Physiological
Role During Export for the Retardation of Folding by the Leader
Peptide of Maltose-Binding Protein. Proc Natl Acad Sci U S A 86:
9213-9217.
Wild, J., W.A. Walter, C.A. Gross, and E. Altman, (1993)
Accumulation of secretory protein precursors in Escherichia coli
induces the heat shock response. J Bacteriol 175: 3992-3997.
Ito, K., Y. Akiyama, T. Yura, and K. Shiba, (1986) Diverse effects
of the MalE-LacZ hybrid protein on Escherichia coli cell
physiology. J Bacteriol 167: 201-204.
Ullers, R.S., J. Luirink, N. Harms, F. Schwager, C. Georgopoulos,
and P. Genevaux, (2004) SecB is a bona fide generalized chaperone
in Escherichia coli. Proc Natl Acad Sci U S A 101: 7583-7588.
Muller, J.P., (1996) Influence of impaired chaperone or secretion
function on SecB production in Escherichia coli. J Bacteriol 178:
6097-6104.
Valent, Q.A., J.W. de Gier, G. von Heijne, D.A. Kendall, C.M. ten
Hagen-Jongman, B. Oudega, and J. Luirink, (1997) Nascent
membrane and presecretory proteins synthesized in Escherichia coli
associate with signal recognition particle and trigger factor. Mol
Microbiol 25: 53-64.
Schaffitzel, C., M. Oswald, I. Berger, T. Ishikawa, J.P. Abrahams,
H.K. Koerten, R.I. Koning, and N. Ban, (2006) Structure of the
Escherichia coli signal recognition particle bound to a translating
ribosome. Nature 444: 503-506.
(Gill and Salmond, B., R., E. Crooke, K. Shiba, W. Wickner, and K.
Ito, (1986) The secY protein can act post-translationally to promote
bacterial protein export. J Biol Chem 261: 12907-12910.
Luirink, J., C.M. ten Hagen-Jongman, C.C. van der Weijden, B.
Oudega, S. High, B. Dobberstein, and R. Kusters, (1994) An
alternative protein targeting pathway in Escherichia coli: studies on
the role of FtsY. EMBO J 13: 2289-2296.
Angelini, S., S. Deitermann, and H.G. Koch, (2005) FtsY, the
bacterial signal-recognition particle receptor, interacts functionally
and physically with the SecYEG translocon. EMBO Rep 6: 476-481.
Valent, Q.A., P.A. Scotti, S. High, J.W. de Gier, G. von Heijne, G.
Lentzen, W. Wintermeyer, B. Oudega, and J. Luirink, (1998) The
Escherichia coli SRP and SecB targeting pathways converge at the
74.
75.
76.
77.
78.
79.
80.
81.
82.
83.
84.
85.
86.
translocon. EMBO J 17: 2504-2512.
Herskovits, A.A. and E. Bibi, (2000) Association of Escherichia coli
ribosomes with the inner membrane requires the signal recognition
particle receptor but is independent of the signal recognition particle.
Proc Natl Acad Sci U S A 97: 4621-4626.
Shan, S.O. and P. Walter, (2005) Co-translational protein targeting
by the signal recognition particle. FEBS Lett 579: 921-926.
Mitra, K., C. Schaffitzel, T. Shaikh, F. Tama, S. Jenni, C.L. Brooks,
3rd, N. Ban, and J. Frank, (2005) Structure of the E. coli proteinconducting channel bound to a translating ribosome. Nature 438:
318-324.
Scotti, P.A., Q.A. Valent, E.H. Manting, M.L. Urbanus, A.J.
Driessen, B. Oudega, and J. Luirink, (1999) SecA is not required for
signal recognition particle-mediated targeting and initial membrane
insertion of a nascent inner membrane protein. J Biol Chem 274:
29883-29888.
Bowers, C.W., F. Lau, and T.J. Silhavy, (2003) Secretion of LamBLacZ by the signal recognition particle pathway of Escherichia coli.
J Bacteriol 185: 5697-5705.
Kim, J., S. Rusch, J. Luirink, and D.A. Kendall, (2001) Is Ffh
required for export of secretory proteins? FEBS Lett 505: 245-248.
Lee, H.C. and H.D. Bernstein, (2001) The targeting pathway of
Escherichia coli presecretory and integral membrane proteins is
specified by the hydrophobicity of the targeting signal. Proc Natl
Acad Sci U S A 98: 3471-3476.
Schierle, C.F., M. Berkmen, D. Huber, C. Kumamoto, D. Boyd, and
J. Beckwith, (2003) The DsbA signal sequence directs efficient,
cotranslational export of passenger proteins to the Escherichia coli
periplasm via the signal recognition particle pathway. J Bacteriol
185: 5706-5713.
Huber, D., D. Boyd, Y. Xia, M.H. Olma, M. Gerstein, and J.
Beckwith, (2005) Use of thioredoxin as a reporter to identify a
subset of Escherichia coli signal sequences that promote signal
recognition particle-dependent translocation. J Bacteriol 187: 29832991.
Nakatogawa, H. and K. Ito, (2001) Secretion monitor, SecM,
undergoes self-translation arrest in the cytosol. Mol Cell 7: 185-192.
Sijbrandi, R., M.L. Urbanus, C.M. ten Hagen-Jongman, H.D.
Bernstein, B. Oudega, B.R. Otto, and J. Luirink, (2003) Signal
recognition particle (SRP)-mediated targeting and sec- dependent
translocation of an extracellular Escherichia coli protein. J Biol
Chem 278: 4654-4659.
Drew, D., L. Fröderberg, L. Baars, and J.W. de Gier, (2003)
Assembly and overexpression of membrane proteins in Escherichia
coli. Biochim Biophys Acta 1610: 3-10.
Newitt, J.A., N.D. Ulbrandt, and H.D. Bernstein, (1999) The
structure of multiple polypeptide domains determines the signal
49
87.
88.
89.
90.
91.
92.
93.
94.
95.
96.
97.
98.
99.
100.
50
recognition particle targeting requirement of Escherichia coli inner
membrane proteins. J Bacteriol 181: 4561-4567.
Valent, Q.A., P.A. Scotti, S. High, J.W. de Gier, G. von Heijne, G.
Lentzen, W. Wintermeyer, B. Oudega, and J. Luirink, (1998) The
Escherichia coli SRP and SecB targeting pathways converge at the
translocon. EMBO J 17: 2504-2512.
Osborne, A.R., T.A. Rapoport, and B. van den Berg, (2005) Protein
translocation by the Sec61/SecY channel. Annu Rev Cell Dev Biol
21: 529-550.
Rusch, S.L. and D.A. Kendall, (2007) Oligomeric states of the SecA
and SecYEG core components of the bacterial Sec translocon.
Biochim Biophys Acta 1768: 5-12.
Deitermann, S., G.S. Sprie, and H.G. Koch, (2005) A dual function
for SecA in the assembly of single spanning membrane proteins in
Escherichia coli. J Biol Chem 280: 39077-39085.
Manting, E.H. and A.J. Driessen, (2000) Escherichia coli
translocase: the unravelling of a molecular machine. Mol Microbiol
37: 226-238.
Brundage, L., J.P. Hendrick, E. Schiebel, A.J. Driessen, and W.
Wickner, (1990) The purified E. coli integral membrane protein
SecY/E is sufficient for reconstitution of SecA-dependent precursor
protein translocation. Cell 62: 649-657.
Akimaru, J., S.I. Matsuyama, H. Tokuda, and S. Mizushima, (1991)
Reconstitution of a Protein Translocation System Containing
Purified SecY, SecE, and SecA from Escherichia coli. Proc Natl
Acad Sci USA 88: 6545-6549.
Nakatogawa, H. and K. Ito, (2001) Secretion monitor, SecM,
undergoes self-translation arrest in the cytosol. Mol Cell 7: 185-192.
Osborne, A.R. and T.A. Rapoport, (2007) Protein translocation is
mediated by oligomers of the SecY complex with one SecY copy
forming the channel. Cell 129: 97-110.
Hessa, T., H. Kim, K. Bihlmaier, C. Lundin, J. Boekel, H.
Andersson, I. Nilsson, S.H. White, and G. von Heijne, (2005)
Recognition of transmembrane helices by the endoplasmic reticulum
translocon. Nature 433: 377-381.
Kihara, A., Y. Akiyama, and K. Ito, (1995) FtsH is required for
proteolytic elimination of uncomplexed forms of SecY, an essential
protein translocase subunit. Proc Natl Acad Sci USA 92: 4532-4536.
Schatz, P.J., K.L. Bieker, K.M. Ottemann, T.J. Silhavy, and J.
Beckwith, (1991) One of three transmembrane stretches is sufficient
for the functioning of the SecE protein, a membrane component of
the E. coli secretion machinery. EMBO J 10: 1749-1757.
Saparov, S.M., K. Erlandson, K. Cannon, J. Schaletzky, S.
Schulman, T.A. Rapoport, and P. Pohl, (2007) Determining the
conductance of the SecY protein translocation channel for small
molecules. Mol Cell 26: 501-509.
Li, W., S. Schulman, D. Boyd, K. Erlandson, J. Beckwith, and T.A.
101.
102.
103.
104.
105.
106.
107.
108.
109.
110.
111.
112.
113.
114.
Rapoport, (2007) The plug domain of the SecY protein stabilizes the
closed state of the translocation channel and maintains a membrane
seal. Mol Cell 26: 511-521.
Harris, C.R. and T.J. Silhavy, (1999) Mapping an interface of SecY
(PrlA) and SecE (PrlG) by using synthetic phenotypes and in vivo
cross-linking. J Bacteriol 181: 3438-3444.
Flower, A.M., (2007) The SecY translocation complex: convergence
of genetics and structure. Trends Microbiol 15: 203-210.
White, S.H. and G. von Heijne, (2004) The machinery of membrane
protein assembly. Curr Opin Struct Biol 14: 397-404.
Hamman, B., J.-C. Chen, E. Johnson, and A. Johnson, (1997) The
aqueous pore through the translocon has a diameter of 40-60 Å
during cotrsnaltional protein translocation at the ER membrane. Cell
89: 535-544.
Sadlish, H., D. Pitonzo, A.E. Johnson, and W.R. Skach, (2005)
Sequential triage of transmembrane segments by Sec61alpha during
biogenesis of a native multispanning membrane protein. Nat Struct
Mol Biol 12: 870-878.
Breyton, C., W. Haase, T.A. Rapoport, W. Kühlbrandt, and I.
Collinson, (2002) Three-dimensional structure of the bacterial
protein-translocation complex SecYEG. Nature 418: 662-665.
Bostina, M., B. Mohsin, W. Kuhlbrandt, and I. Collinson, (2005)
Atomic model of the E. coli membrane-bound protein translocation
complex SecYEG. J Mol Biol 352: 1035-1043.
Woodbury, R.L., S.J. Hardy, and L.L. Randall, (2002) Complex
behavior in solution of homodimeric SecA. Protein Sci 11: 875-882.
Or, E., A. Navon, and T. Rapoport, (2002) Dissociation of the
dimeric SecA ATPase during protein translocation across the
bacterial membrane. EMBO J 21: 4470-4479.
Benach, J., Y.T. Chou, J.J. Fak, A. Itkin, D.D. Nicolae, P.C. Smith,
G. Wittrock, D.L. Floyd, C.M. Golsaz, L.M. Gierasch, and J.F.
Hunt, (2003) Phospholipid-induced monomerization and signalpeptide-induced oligomerization of SecA. J Biol Chem 278: 36283638.
Hunt, J.F., S. Weinkauf, L. Henry, J.J. Fak, P. McNicholas, D.B.
Oliver, and J. Deisenhofer, (2002) Nucleotide control of interdomain
interactions in the conformational reaction cycle of SecA. Science
297: 2018-2026.
Vassylyev, D.G., H. Mori, M.N. Vassylyeva, T. Tsukazaki, Y.
Kimura, T.H. Tahirov, and K. Ito, (2006) Crystal structure of the
translocation ATPase SecA from Thermus thermophilus reveals a
parallel, head-to-head dimer. J Mol Biol 364: 248-258.
Zimmer, J., W. Li, and T.A. Rapoport, (2006) A novel dimer
interface and conformational changes revealed by an X-ray structure
of B. subtilis SecA. J Mol Biol 364: 259-265.
Alami, M., K. Dalal, B. Lelj-Garolla, S.G. Sligar, and F. Duong,
(2007) Nanodiscs unravel the interaction between the SecYEG
51
115.
116.
117.
118.
119.
120.
121.
122.
123.
124.
125.
126.
127.
128.
129.
52
channel and its cytosolic partner SecA. EMBO J 26: 1995-2004.
Driessen, A.J., (1993) SecA, the peripheral subunit of the
Escherichia coli precursor protein translocase, is functional as a
dimer. Biochemistry 32: 13190-13197.
de Keyzer, J., E.O. van der Sluis, R.E. Spelbrink, N. Nijstad, B. de
Kruijff, N. Nouwen, C. van der Does, and A.J. Driessen, (2005)
Covalently dimerized SecA is functional in protein translocation. J
Biol Chem 280: 35255-35260.
Jilaveanu, L.B. and D. Oliver, (2006) SecA dimer cross-linked at its
subunit interface is functional for protein translocation. J Bacteriol
188: 335-338.
McFarland, L., O. Francetic, and C.A. Kumamoto, (1993) A
mutation of Escherichia coli SecA protein that partially compensates
for the absence of SecB. J Bacteriol 175: 2255-2262.
Or, E. and T. Rapoport, (2007) Cross-linked SecA dimers are not
functional in protein translocation. FEBS Lett 581: 2616-2620.
Chen, Y., P.C. Tai, and S.F. Sui, (2007) The active ring-like
structure of SecA revealed by electron crystallography:
Conformational change upon interaction with SecB. J Struct Biol
159: 149-153.
Duong, F. and W. Wickner, (1997) Distinct catalytic roles of the
SecYE, SecG and SecDFyajC subunits of preprotein translocase
holoenzyme. EMBO J 16: 2756-2768.
Scotti, P.A., M.L. Urbanus, J. Brunner, J.W. de Gier, G. von Heijne,
C. van der Does, A.J. Driessen, B. Oudega, and J. Luirink, (2000)
YidC, the Escherichia coli homologue of mitochondrial Oxa1p, is a
component of the Sec translocase. Embo J 19: 542-549.
Nouwen, N. and A.J. Driessen, (2002) SecDFyajC forms a
heterotetrameric complex with YidC. Mol Microbiol 44: 1397-1405.
Matsuyama, S., Y. Fujita, K. Sagara, and S. Mizushima, (1992)
Overproduction, purification and characterization of SecD and SecF,
integral membrane components of the protein translocation
machinery of Escherichia coli. Biochim Biophys Acta 1222: 77-84.
Pogliano, J.A. and J. Beckwith, (1994) SecD and SecF facilitate
protein export in Escherichai coli. EMBO J 13: 554-561.
Yi, L., N. Celebi, M. Chen, and R.E. Dalbey, (2004) Sec/SRP
requirements and energetics of membrane insertion of subunits a, b,
and c of the Escherichia coli F1F0 ATP synthase. J Biol Chem 279:
39260-39267.
Matsuyama, S., Y. Fujita, and S. Mizushima, (1993) SecD Is
Involved in the Release of Translocated Secretory Proteins from the
Cytoplasmic Membrane of Escherichia coli. EMBO J 12: 265-270.
Duong, F. and W. Wickner, (1997) The SecDFyajC domain of
preprotein translocase controls preprotein movement by regulating
SecA membrane cycling. EMBO J 16: 4871-4879.
Albers, S.V., Z. Szabo, and A.J. Driessen, (2006) Protein secretion
in the Archaea: multiple paths towards a unique cell surface. Nat
130.
131.
132.
133.
134.
135.
136.
137.
138.
139.
140.
141.
142.
Rev Microbiol 4: 537-547.
van de Vossenberg, J.L., S.V. Albers, C. van der Does, A.J.
Driessen, and W. van Klompenburg, (1998) The positive inside rule
is not determined by the polarity of the delta psi (transmembrane
electrical potential) [letter]. Mol Microbiol 29: 1125-1127.
Urbanus, M.L., L. Froderberg, D. Drew, P. Bjork, J.W. de Gier, J.
Brunner, B. Oudega, and J. Luirink, (2002) Targeting, insertion, and
localization of Escherichia coli YidC. J Biol Chem 277: 1271812723.
Pohlschroder, M., E. Hartmann, N.J. Hand, K. Dilks, and A.
Haddad, (2005) Diversity and evolution of protein translocation.
Annu Rev Microbiol 59: 91-111.
Samuelson, J.C., M. Chen, F. Jiang, I. Moller, M. Wiedmann, A.
Kuhn, G.J. Phillips, and R.E. Dalbey, (2000) YidC mediates
membrane protein insertion in bacteria. Nature 406: 637-641.
Chen, M., J.C. Samuelson, F. Jiang, M. Müller, A. Kuhn, and R.E.
Dalbey, (2002) Direct interaction of YidC with the Sec-independent
Pf3 coat protein during its membrane protein insertion. J Biol Chem
277: 7670-7675.
Froderberg, L., E. Houben, J.C. Samuelson, M.Y. Chen, S.K. Park,
G.J. Phillips, R. Dalbey, J. Luirink, and J.W.L. de Gier, (2003)
Versatility of inner membrane protein biogenesis in Escherichia
coli. Mol Microbiol 47: 1015-1027.
van der Laan, M., M.L. Urbanus, C.M. Ten Hagen-Jongman, N.
Nouwen, B. Oudega, N. Harms, A.J. Driessen, and J. Luirink,
(2003) A conserved function of YidC in the biogenesis of
respiratory chain complexes. Proc Natl Acad Sci U S A 100: 58015806.
Van Der Laan, M., P. Bechtluft, S. Kol, N. Nouwen, and A.J.
Driessen, (2004) F1F0 ATP synthase subunit c is a substrate of the
novel YidC pathway for membrane protein biogenesis. J Cell Biol
165: 213-222.
van Bloois, E., G. Jan Haan, J.W. de Gier, B. Oudega, and J.
Luirink, (2004) F(1)F(0) ATP synthase subunit c is targeted by the
SRP to YidC in the E. coli inner membrane. FEBS Lett 576: 97-100.
Facey, S.J., S.A. Neugebauer, S. Krauss, and A. Kuhn, (2007) The
mechanosensitive channel protein MscL is targeted by the SRP to
the novel YidC membrane insertion pathway of Escherichia coli. J
Mol Biol 365: 995-1004.
Scotti, P.A., M.L. Urbanus, J. Brunner, J.W.L. de Gier, G. von
Heijne, C. van der Does, A.J.M. Driessen, B. Oudega, and J.
Luirink, (2000) YidC, the Escherichia coli homologue of
mitochondrial Oxa1p, is a component of the Sec translocase. EMBO
J 19: 542-549.
Facey, S.J. and A. Kuhn, (2004) Membrane integration of E. coli
model membrane proteins. Biochim Biophys Acta 1694: 55-66.
Urbanus, M.L., P.A. Scotti, L. Froderberg, A. Saaf, J.W. de Gier, J.
53
143.
144.
145.
146.
147.
148.
149.
150.
151.
152.
153.
154.
155.
54
Brunner, J.C. Samuelson, R.E. Dalbey, B. Oudega, and J. Luirink,
(2001) Sec-dependent membrane protein insertion: sequential
interaction of nascent FtsQ with SecY and YidC. EMBO Rep 2: 524529.
van der Laan, M., E.N. Houben, N. Nouwen, J. Luirink, and A.J.
Driessen, (2001) Reconstitution of Sec-dependent membrane protein
insertion: nascent FtsQ interacts with YidC in a SecYEG-dependent
manner. EMBO Rep 2: 519-523.
Houben, E.N., P.A. Scotti, Q.A. Valent, J. Brunner, J.L. de Gier, B.
Oudega, and J. Luirink, (2000) Nascent Lep inserts into the
Escherichia coli inner membrane in the vicinity of YidC, SecY and
SecA. FEBS Lett 476: 229-233.
van Bloois, E., G.J. Haan, J.W. de Gier, B. Oudega, and J. Luirink,
(2006) Distinct requirements for translocation of the N-tail and Ctail of the Escherichia coli inner membrane protein CyoA. J Biol
Chem 281: 10002-10009.
Nagamori, S., I.N. Smirnova, and H.R. Kaback, (2004) Role of
YidC in folding of polytopic membrane proteins. J Cell Biol 165:
53-62.
Froderberg, L., E.N. Houben, L. Baars, J. Luirink, and J.W. de Gier,
(2004) Targeting and translocation of two lipoproteins in
Escherichia coli via the SRP/Sec/YidC pathway. J Biol Chem 279:
31026-31032.
Yen, M.R., Y.H. Tseng, E.H. Nguyen, L.F. Wu, and M.H. Saier, Jr.,
(2002) Sequence and phylogenetic analyses of the twin-arginine
targeting (Tat) protein export system. Arch Microbiol 177: 441-450.
Sargent, F., B.C. Berks, and T. Palmer, (2006) Pathfinders and
trailblazers: a prokaryotic targeting system for transport of folded
proteins. FEMS Microbiol Lett 254: 198-207.
Rodrigue, A., A. Chanal, K. Beck, M. Muller, and L.F. Wu, (1999)
Co-translocation of a periplasmic enzyme complex by a hitchhiker
mechanism through the bacterial tat pathway. J Biol Chem 274:
13223-13228.
Hatzixanthis, K., T. Palmer, and F. Sargent, (2003) A subset of
bacterial inner membrane proteins integrated by the twin-arginine
translocase. Mol Microbiol 49: 1377-1390.
Bos, M.P., V. Robert, and J. Tommassen, (2006) Biogenesis of the
Gram-Negative Bacterial Outer Membrane. Annu Rev Microbiol.
Chen, R. and U. Henning, (1996) A periplasmic protein (Skp) of
Escherichia coli selectively binds a class of outer membrane
proteins. Mol Microbiol 19: 1287-1294.
Harms, N., G. Koningstein, W. Dontje, M. Muller, B. Oudega, J.
Luirink, and H. de Cock, (2001) The early interaction of the outer
membrane protein PhoE with the periplasmic chaperone Skp occurs
at the cytoplasmic membrane. J Biol Chem 276: 18804-18811.
Spiess, C., A. Beil, and M. Ehrmann, (1999) A temperaturedependent switch from chaperone to protease in a widely conserved
156.
157.
158.
159.
160.
heat shock protein. Cell 97: 339-347.
Wu, T., J. Malinverni, N. Ruiz, S. Kim, T.J. Silhavy, and D. Kahne,
(2005) Identification of a multicomponent complex required for
outer membrane biogenesis in Escherichia coli. Cell 121: 235-245.
Ruiz, N., B. Falcone, D. Kahne, and T.J. Silhavy, (2005) Chemical
conditionality: a genetic strategy to probe organelle assembly. Cell
121: 307-317.
Voulhoux, R. and J. Tommassen, (2004) Omp85, an evolutionarily
conserved bacterial protein involved in outer-membrane-protein
assembly. Res Microbiol 155: 129-135.
Hegde, R.S. and H.D. Bernstein, (2006) The surprising complexity
of signal sequences. Trends Biochem Sci 31: 563-571.
Casadio, R., P. Fariselli, G. Finocchiaro, and P.L. Martelli, (2003)
Fishing new proteins in the twilight zone of genomes: the test case
of outer membrane proteins in Escherichia coli K12, Escherichia
coli O157:H7, and other Gram-negative bacteria. Protein Sci 12:
1158-1168.
55
Downloaded from www.proteinscience.org on August 6, 2007
New Escherichia coli outer membrane proteins
identified through prediction and experimental
verification
PAOLA MARANI,1,2,4 SAMUEL WAGNER,1,4 LOUISE BAARS,1
PIERRE GENEVAUX,3 JAN-WILLEM DE GIER,1 INGMARIE NILSSON,1
RITA CASADIO,2 AND GUNNAR VON HEIJNE1
1
2
3
Department of Biochemistry and Biophysics, Stockholm University, SE-106 91 Stockholm, Sweden
Laboratory of Biocomputing, CIRB/Department of Biology, University of Bologna, Bologna, Italy
Department of Microbiology and Molecular Medicine, Centre Médical Universitaire, CH-1211 Geneva, Switzerland
(RECEIVED October 5, 2005; FINAL REVISION December 23, 2005; ACCEPTED December 23, 2005)
Abstract
Many new Escherichia coli outer membrane proteins have recently been identified by proteomics
techniques. However, poorly expressed proteins and proteins expressed only under certain conditions
may escape detection when wild-type cells are grown under standard conditions. Here, we have taken
a complementary approach where candidate outer membrane proteins have been identified by
bioinformatics prediction, cloned and overexpressed, and finally localized by cell fractionation experiments. Out of eight predicted outer membrane proteins, we have confirmed the outer membrane
localization for five—YftM, YaiO, YfaZ, CsgF, and YliI—and also provide preliminary data indicating that a sixth—YfaL—may be an outer membrane autotransporter.
Keywords: outer membrane protein; bioinformatics; SecB; autotransporter
From the known high-resolution structures of transmembrane proteins, only two basic architectures have
been identified so far: the helix bundle and the b-barrel
(von Heijne 1999). Helix bundle proteins have been
extensively studied, both from an experimental and
from a bioinformatics perspective, and rather reliable
prediction methods exist for their identification from
sequence data alone (Chen et al. 2002; Melén et al.
2003). b-Barrel proteins have received comparatively
less attention, and only a few methods have been proposed for identification and topology prediction of such
proteins (Casadio et al. 2003; Bagos et al. 2004; Berven
et al. 2004; Bigelow et al. 2004).
An important use of bioinformatics prediction schemes
is to guide the experimentalist toward targets that
are highly likely to correspond to true instances of
the particular kind of gene or protein of interest. Here,
we have used the recently developed Hunter predictor
(Casadio et al. 2003) to select likely outer membrane
proteins among the nonannotated part of the Escherichia coli proteome, and have experimentally verified the
predicted outer membrane localization of five hitherto
uncharacterized proteins: YftM, YaiO, YfaZ, CsgF, and
YliI. We further provide data indicating that a sixth protein, YfaL, is an outer membrane autotransporter.
Results
4
These authors contributed equally to this work.
Reprint requests to: Gunnar Von Heijne, Department of Biochemistry and Biophysics, Stockholm University, SE-106 91 Stockholm,
Sweden; e-mail: [email protected]; fax: +46-8-15-36-79.
Article published online ahead of print. Article and publication date are
at http://www.proteinscience.org/cgi/doi/10.1110/ps.051889506.
884
Selection of target proteins
From the list of 18 new outer membrane proteins predicted by the Hunter predictor in the E. coli proteome
(see Table 3 in Casadio et al. 2003), we initially chose 11
Protein Science (2006), 15:884–889. Published by Cold Spring Harbor Laboratory Press. Copyright 2006 The Protein Society
Downloaded from www.proteinscience.org on August 6, 2007
New outer membrane proteins
proteins, characterized by different lengths and different
numbers of predicted b-strands, for further analysis.
Despite repeated attempts, only eight of these genes
could be cloned in our vector system. Therefore, we
focused our experimental analysis on this set of putative
outer membrane proteins (Table 1).
Cloning and expression of target proteins
The eight target genes were cloned into the pING vector
(Johnston et al. 1985), and a hemagglutinin (HA) tag was
added to the C terminus of the gene products for immunodetection. Induction with arabinose and labeling with
[35S]-Met in all cases gave rise to a protein product that
could be immunoprecipitated by an HA antibody and was
of the expected molecular weight (data not shown).
In initial [35S]-Met labeling experiments, we noted that
seven of the eight proteins appeared as doublets (Fig. 1),
possibly reflecting inefficient removal of the signal peptide
(because of its large size, small molecular weight differences
could not be detected in pulse-chase experiments with
YfaL). To study this possibility further, we blocked SecAdependent translocation through the inner membrane
SecYEG translocon by adding sodium azide 30 sec prior
to the addition of [35S]-Met (Oliver et al. 1990). As seen in
Figure 1, after a 1-min pulse, significantly more of the higher
molecular-weight form was seen in the presence than in the
absence of azide for all seven proteins, strongly suggesting
that these proteins are translocated across the inner membrane in a SecA- and translocon-dependent process.
Outer membrane localization of target proteins
To assay the possible outer membrane localization of the
target proteins, we separated outer and inner membranes by successive two- and six-step sucrose gradient
Table 1. Proteins included in the study
UniProt code
Molecular
weight
(processed, Urea
SecB
HA tagged) pellet dependence
CSGF_ECOLI
14.2
+
+
YHJY_ECOLI
YTFM_ECOLI
25.1
63.8
–
++
+
+
YFAL_ECOLI
129.7
N.D.
N.D.
YAIO_ECOLI
YLII_ECOLI
28.3
40.0
++
+
+
+
YFAZ_ECOLI
YAGZ_ECOLI
17.9
19.1
+
–
+
+
UniProt
annotation
Biogenesis of
curli organelles
Lipase 1
Hypothetical
protein
Putative
autotransporter
–
Putative glucose
dehydrogenase
–
Fimbrillin
Figure 1. Translocation of the target proteins across the inner membrane is SecA dependent. Protein expression was induced for 5 min
with arabinose, followed by labeling for 1 min with [35S]-Met (YliI was
labeled for 3 min). Sodium azide was added to a final concentration of
2 mM (+ lanes) 30 sec prior to radio-labeling. Proteins were immunoprecipitated with antisera against the HA-tag. Precursor (p) and
mature (m) forms of the proteins are indicated.
centrifugation. The purity of the inner and outer membrane fractions was determined by Western blotting
against the inner and outer membrane marker proteins
Lep and OmpA, respectively.
All target proteins were found in the outer membrane
fraction, together with the control outer membrane protein OmpA (Fig. 2). A higher molecular-weight form
migrating slightly slower than the 150-kDa standard
was seen for YtfM, possibly representing an SDS-resistant dimeric form of the protein. For the relatively
strongly expressed YaiO and YliI proteins, trace amounts
were also present in the inner membrane fraction, most
likely due to cross-contamination.
It was recently shown that cytosolic aggregates of misfolded proteins cosediment with the outer membrane fraction upon sucrose density gradient centrifugation, and
that the inclusion-body binding proteins IbpA and IbpB
can be used as a marker for these aggregates (Laskowska
et al. 2004). To evaluate if our overexpressed target proteins are inserted into the outer membrane and do not
simply copurify in aggregates, we developed a protocol in
which the purified outer membrane fraction is washed
with 5 M urea to dissolve potential aggregates but leave
membranes intact. Similar procedures are often used to
demonstrate the correct insertion of helix bundle proteins
into membranes (Chen et al. 2003). Western blots against
IbpA,B showed that aggregates are solubilized by urea
treatment, whereas the major outer membrane protein
OmpA remains in the membrane pellet (Fig. 3A).
As shown in Figure 3B, only YaiO and YftM remained
totally in the urea-resistant outer membrane fraction (the
latter gave rise to two additional lower molecular weight
www.proteinscience.org
885
Downloaded from www.proteinscience.org on August 6, 2007
Marani et al.
We considered the possibility that the 55-kDa fragment
is the cleaved translocator domain of the autotransporter,
which would be similar in size to the AIDA-I translocator
domain (47.5 kDa). As many autotransporters are serine/
threonine proteases, we tested if the cleavage of the putative translocator domain could be inhibited by the serine/
threonine protease inhibitor Pefabloc SC. Indeed, when
YfaL was expressed in the presence of Pefabloc SC, the
55-kDa band disappeared (Fig. 4). The putative cleaved
75-kDa N-terminal passenger domain does not contain a
HA-tag and thus cannot be detected by Western blotting.
Figure 2. Membrane fractionation. Cells were grown at 37C, and
expression of HA-tagged target proteins was induced with arabinose
at an OD600 of 0.4–0.6. Cells were harvested 45 min after induction and
lysed by French pressing. Inner and outer membrane fractions were
prepared by sucrose density gradient centrifugation, separated by SDSPAGE, and probed by immunodecoration of the HA-tagged proteins.
Western blots of the outer membrane marker OmpA and the inner
membrane marker Lep are also shown.
bands upon urea treatment; the identity of these bands is
unknown). CsgF and YfaZ were also largely urea-resistant,
while YfaL, YliI, YhjY, and YagZ were to a greater or
lesser extent removed from the outer membrane fraction.
Since the formation of inclusion bodies is often reduced
at lower temperature, YfaL, YliL, YhjY, and CsgF were
expressed also at 30C (Fig. 3C). The amounts of CsgF and
YliI in the outer membrane fraction increased, whereas
YhjY was still mainly extracted. Expression of YfaL was
too low for detection under these conditions. Our results
strongly suggest that at least five of the eight proteins (YftM,
YaiO, YfaZ, CsgF, YliI) are localized to the outer membrane.
The 130-kDa protein YfaL has been predicted to be
an autotransporter (Yen et al. 2002). Supporting this,
the C-terminal part shows considerable homology with
the AIDA-I autotransporter of E. coli O126:H27, which
mediates binding to an integral membrane glycoprotein
on HeLa cells (Laarmann and Schmidt 2003). In general,
autotransporters consist of an outer membrane C-terminal translocator domain and a globular N-terminal
passenger domain that mediates the ultimate function of
the protein. After translocation through the outer membrane, the passenger domain is (often autolytically)
cleaved off (Henderson et al. 1998).
As noted above, we could not detect YfaL in the outer
membrane after cell fractionation and urea extraction. A
substantial amount of the expressed protein seems to
end up in aggregates, which is in concert with the fact
that the inclusion body binding protein IbpB is up-regulated upon overexpression of YfaL (data not shown).
Interestingly, when whole cells were subjected to SDSPAGE and Western blotting against the C-terminal HAtag, an additional band at 55 kDa appeared (Fig. 4).
886
Protein Science, vol. 15
SecB dependence
It is generally assumed that the cytoplasmic chaperone
SecB facilitates the export of precursor polypeptides by
Figure 3. Urea wash of outer membrane fractions. Outer membrane
fractions from MC1061 cells overexpressing the different target proteins
were prepared as described in Figure 2. After fractionation, membranes
were washed with PBS plus 5 M urea for 1 h in the cold. Urea-treated
membranes (+ wash) and untreated controls (– wash) were analyzed by
Western blotting. (A) Western blots of the inclusion body binding protein
B (IbpB; the cells in this experiment were induced for expression of YfaL)
and the major outer membrane protein OmpA. (B) Western blots of the
indicated HA-tagged target proteins expressed at 37C. (C) Western
blots of the indicated HA-tagged target proteins expressed at 30C.
Downloaded from www.proteinscience.org on August 6, 2007
New outer membrane proteins
Figure 4. Analysis of the putative outer membrane autotransporter
YfaL. Cells were grown in LB medium at 30C, and expression of
HA-tagged YfaL was induced with arabinose at an OD600 of 0.4–0.6
for 2 h. Induced cells were grown in the presence or the absence of
Pefabloc SC. Cells were harvested and subsequently analyzed by Western blotting against the HA-tag. Full-length YfaL (I) and the putative
55-kDa translocator domain (II) are indicated.
maintaining them in a translocation competent conformation and by delivering them to SecA (Randall and Hardy
2002). However, up until recently, it has been shown for
only six proteins (PhoE, LamB, MBP, GBP, OmpF, and
OmpA) that their export is facilitated by SecB (Kumamoto and Beckwith 1985; de Cock et al. 1992; Powers and
Randall 1995), whereas four proteins (PhoA, Lpp, RbsB,
and AmpC) do not seem to require SecB (Knoblauch et al.
1999; Xu et al. 2000; Randall and Hardy 2002; Dekker et
al. 2003). Twelve additional proteins were recently identified as SecB substrates in a proteomics screen (Baars et al.
2006).
In an attempt to expand the list even further, we
expressed HA-tagged CsgF, YfaZ, YagZ, YhjY, YaiO,
YliI, and YftM in the secB null strain MC4100DsecB (no
expression was seen for YfaL) and the control strain
MC4100 (Fig. 5). With the possible exception of YhjY,
the relative amount of precursor was clearly increased in
the secB null strain compared with wild type for all seven
proteins, indicating that SecB facilitates their targeting.
As expected, the uncleaved precursor form pro-OmpA
accumulated in the transformed secB null strains but not
in the transformed control strains (data not shown).
identification of bacterial outer membrane proteins, and
may be particularly effective for low-abundance proteins or
proteins that are expressed only under certain conditions.
The outer membrane localization of the five proteins was
experimentally verified by sucrose density gradient centrifugation and urea treatment of the outer membranes. Only
YhjY gave ambiguous results: We could not determine if it
is located in the outer membrane as it is difficult to overexpress and ends up mostly in inclusion bodies. YagZ could be
extracted completely from the outer membrane fraction
with 5 M urea and is thus not embedded in the outer membrane. YagZ is identical with the protein MatB (meningitisassociated and temperature-regulated), which has been
shown recently to be the major fimbrillin of the Mat fimbria
(and thus not directly inserted in the outer membrane)
(Pouttu et al. 2001). MatB is expressed in some pathogenic
strains (MENEC) but not in the laboratory strain K12.
Finally, our results indicate that YfaL is an autotransporter
with a cleavable 55-kDa translocator domain.
In common with previously studied outer membrane
proteins, we find that targeting of the outer membrane
proteins identified here is facilitated by the SecB chaperone, suggesting that SecB dependence may be a common
characteristic of outer membrane proteins.
Materials and methods
Enzymes and chemicals
Unless otherwise stated, all enzymes were from Promega or
New England Biolabs. [35S]-Met and [14C]-methylated marker
Discussion
Until recently, identification of bacterial outer membrane
proteins by computational approaches (other than standard sequence similarity searches) has been a neglected
field in bioinformatics. Here, we have experimentally verified that at least five of the top candidate outer membrane
proteins identified by the Hunter predictor among the
unannotated portion of the E. coli proteome (Casadio et
al. 2003)—YftM, YaiO, YfaZ, CsgF, and YliI—are in fact
localized in the outer membrane. Target protein selection
based on bioinformatics predictions followed by experimental verification is thus a viable alternative to largescale proteomics approaches (Molloy et al. 2000) for the
Figure 5. Analysis of SecB dependence. The indicated HA-tagged
target proteins were expressed in the secB null mutant MC4100DsecB
(– SecB) and the wild-type strain MC4100 (+ SecB). Protein expression was induced with arabinose for 5 min, followed by labeling for 1
min with [35S]-Met. Proteins were immunoprecipitated with antisera
against the HA-tag. Precursor (p) and mature (m) forms of the proteins
are indicated. Note the slow cleavage of YliI, which is complete only
after a 3-min pulse in wild-type cells (Fig. 1).
www.proteinscience.org
887
Downloaded from www.proteinscience.org on August 6, 2007
Marani et al.
proteins were from Amersham-Pharmacia Biotech. Protein A–
Sepharose and sodium azide were from Sigma Chemical. Pansorbin was from Calbiochem Biochemicals &Immunochemicals.
BigDye Terminator v1.1 Cycle Sequencing Kit was from AB
Applied Biosystems,, and oligonucleotides were from CyberGene AB. The QuikChange site-directed mutagenesis kit was
from Stratagene. The Expand Long Template PCR System was
from Roche Diagnostics GmbH, and the QIAquick PCR purification kit was from Qiagen. All mutants were confirmed by
sequencing of plasmid DNA at BM Labbet AB. Rabbit polyclonal anti-HA-tag (influenza HA-epitope) antibody was from
Abcam Limited. BCA protein concentration assay was from
Pierce, and Pefabloc was from Biomol.
DNA techniques
The genes encoding the eight E. coli target proteins were amplified from E. coli strains MG1655 (Blattner et al. 1997) or
MC1061 using Expand Long Polymerase. For cloning into and
in vivo expression from the pING1 plasmid (see Whitley et al.
1994), both ends of the gene were modified during PCR amplification by introducing a XhoI site and an initiator ATG codon
encoded by the 5¢ primer, and by changing the 3¢ end of the gene
by a reverse primer encoding a HA-tag, YPYDVPDYA, two
stop codons (TAA TAG), and a SmaI site. Thus, the 5¢ region of
the gene was modified to …CTCGAGTATG… (XhoI site and
initiator codon underlined). The resulting fragment was cloned
into the pING vector behind the ara promoter using an XhoI site
and a SmaI site introduced by site-specific mutagenesis.
Strains, plasmids, culture conditions, and pulse
experiments
Experiments were performed in E. coli strain MC1061 (Dalbey
and Wickner 1986), MC4100 (Casadaban and Cohen 1979),
and MC4100DsecB (R.S. Ullers., F. Schwager, D. Ang, C. Georgopoulos, and P. Genevaux, in prep.). Constructs were expressed
from the pING plasmid (Johnston et al. 1985) by induction with
L-arabinose.
E. coli strains were transformed with the pING vector carrying the relevant constructs under control of the arabinose
promoter were grown at 37C in M9 minimal medium supplemented with 100 mg/mL ampicillin, 0.5% (w/v) fructose, 100
mg/mL thiamine, and all amino acids (50 mg/mL each) except
methionine. An overnight culture was diluted 1:25 in fresh
medium, shaken for 3.5 h at 37C, induced with arabinose
(0.2% [w/v]) for 5 min, labeled with [35S]-Met (75 mCi/mL)
for 1 min, and put on ice. Sodium azide (final concentration
2 mM) was added 30 sec before radiolabeling. Samples were
acid-precipitated with trichloroacetic acid (TCA) (10% [v/v]
final concentration), resuspended, and then analyzed by immunoprecipitation with HA-antiserum combined with SDSPAGE as described previously (Fröderberg et al. 2004). Proteins were visualized in a Fuji FLA-3000 PhosphorImager
using the Image Reader V1.8J/Image Gauge V 3.45 software.
Separation of outer and inner membrane fractions
Cell fractionation was carried out essentially as described in
Laskowska et al. (2004) using two subsequent sets of sucrose
density gradients. Samples (1000 mL) of strain MC1061 transformed with a pING vector harboring the different OMPs were
888
Protein Science, vol. 15
grown at 37C and 30C, respectively. Expression of the outer
membrane proteins was induced by the addition of 0.1% arabinose at an OD600 of 0.4–0.6. Cells were harvested 45 min after
induction at 6000g using a Beckman 8.1000 rotor.
The 1000 OD600 units of cells were resuspended in 6 mL
buffer K (50 mM triethanolamine [TEA], 250 mM sucrose, 1
mM EDTA, 1 mM dithiothreitol [DTT], 0.1 mg/mL Pefabloc
at pH 7.5) and lysed by two cycles of French pressing (18,000
psi). The lysate was clarified of unbroken cells by 20-min
centrifugation at 8,000g. The supernatant was transferred on
top of a two-step sucrose gradient: bottom to top, 1 mL 55%
(w/w), 5.5 mL 9% (w/w). All sucrose gradients were prepared
in buffer M: 50 mM TEA, 1 mM EDTA, and 1 mM DTT (pH
7.5). The gradients were spun for 2.5 h at 210,000g in a Beckman SW 40 rotor, and the membrane fraction was collected
from the top of the 55% sucrose step. This fraction, which
contains the entire membranes, was diluted 1:1 with buffer M
and subjected to a six-step sucrose gradient to obtain pure
inner and outer membrane fractions. The assembly of this
second gradient was as follows (from bottom to top): 0.8 mL
at 55%; 2.0 mL at 50%, 45%, 40%, and 35%; 0.8 mL at 30% (all
w/w) and 3.3 mL of the sample. The gradients were spun for 15
h at 210,000g in a Beckman SW 40 rotor, and the inner and
outer membrane fractions were collected from the top of the
40% and 50% sucrose steps, respectively.
The purity of the fractions was confirmed by Western blotting against Lep and OmpA as inner and outer membrane
markers, respectively. The protein concentration of the fractions was determined by a BCA assay according to the instructions of the manufacturer (Pierce).
Aggregate removal
Outer membranes containing 50 mg of protein were resuspended in PBS/5 M urea and washed by rotating for 1 h in
the cold room. Membranes were collected in Beckman TLA
100.3 at 194,000g for 20 min. Urea-washed membranes and
unwashed control membranes were analyzed by Western blotting against the HA-tag. Blotting against OmpA was used as a
control for a protein that is properly inserted into the outer
membrane and that cannot be washed away. Blotting against
IbpB was used to show that the aggregates could be washed
away by the 5 M urea treatment.
Immunoblot analysis
The expression of the target proteins (with HA-tag fused to the
C terminus), Lep, OmpA, and IbpA,B (the IbpB antiserum
cross-reacts with IbpA) in the inner/outer membranes and
aggregates was monitored by immunoblot analysis. Cells
were cultured as described above. Purified inner/outer membranes or aggregates (5 mg protein) were solubilized in Laemmli solubilization buffer and were separated by SDS-PAGE.
Proteins were transferred from the polyacrylamide gel to a
polyvinylidene fluoride (PVDF) membrane (Millipore). Subsequently, membranes were blocked with 5% milk and decorated
with antisera to the components listed above. Proteins were
visualized with secondary HRP-conjugated antibodies (BioRad) using the ECL system according to the instructions
of the manufacturer (Amersham Pharmacia) and a Fuji LAS
1000-Plus CCD camera. Blots were quantified using the Image
Gauge software (version 3.4). Experiments were repeated at
least twice. If the membrane had to be tested with more than
Downloaded from www.proteinscience.org on August 6, 2007
New outer membrane proteins
one antibody, it was washed with 5 M urea and 10 mM DTT
overnight at 37C, blocked, and reused as before (Terzi et al.
2004).
Protease inhibition assay for YfaL
MC1061 transformed with a pING vector harboring the yfaL
gene fused to a C-terminal HA-tag was grown at 30C in LB
medium supplemented with 100 mg/mL ampicillin. Expression
was induced by the addition of 0.1% arabinose at an OD600 of
0.4–0.6 in the presence or absence of 1 mg/mL Pefabloc serine
protease inhibitor. Cells were harvested 2 h after induction,
and 0.15 OD600 units of whole cells/well was run on an SDSPAGE and analyzed by Western blotting as described above.
Acknowledgments
We thank B. Bukau for gift of IbpB antiserum, and C. Georgopoulos, in whose laboratory part of the work was performed. This work was supported by grants from the Swedish
Research Council and the Marianne and Marcus Wallenberg
Foundation to G.v.H.; from FIRB and the European Community BioSapiens and Functional Genomics programs to R.C.;
from Bologna University, CNR, and AIRBBC to P.M.; and
from the Swiss National Science Foundation (FN-31-65403) to
P.G.
References
Baars, L., Ytterberg, J., Drew, D., Wagner, S., Thilo, C., van Wijk, K.-J.,
and de Gier, J.-W. 2006. Defining the role of the E. coli chaperone SecB
using comparative proteomics. J. Biol. Chem. (in press).
Bagos, P.G., Liakopoulos, T.D., Spyropoulos, I.C., and Hamodrakas, S.J.
2004. PRED-TMBB: A web server for predicting the topology of bbarrel outer membrane proteins. Nucleic Acids Res. 32: W400–W404.
Berven, F.S., Flikka, K., Jensen, H.B., and Eidhammer, I. 2004. BOMP: A
program to predict integral b-barrel outer membrane proteins encoded
within genomes of Gram-negative bacteria. Nucleic Acids Res. 32:
W394–W399.
Bigelow, H.R., Petrey, D.S., Liu, J., Przybylski, D., and Rost, B. 2004.
Predicting transmembrane b-barrels in proteomes. Nucleic Acids Res.
32: 2566–2577.
Blattner, F.R., Plunkett, G., Bloch, C.A., Perna, N.T., Burland, V., Riley,
M., Collado-Vides, J., Glasner, J.D., Rode, C.K., Mayhew, G.F., et al.
1997. The complete genome sequence of Escherichia coli K-12. Science
277: 1453–1462.
Casadaban, M.J. and Cohen, S.N. 1979. Lactose genes fused to exogenous
promoters in one step using a Mu-lac bacteriophage: In vivo probe for
transcriptional control sequences. Proc. Nat. Acad. Sci. 76: 4530–4533.
Casadio, R., Fariselli, P., Finocchiaro, G., and Martelli, P.L. 2003.
Fishing new proteins in the twilight zone of genomes: The test
case of outer membrane proteins in Escherichia coli K12, Escherichia
coli O157:H7, and other Gram-negative bacteria. Protein Sci. 12:
1158–1168.
Chen, C.P., Kernytsky, A., and Rost, B. 2002. Transmembrane helix predictions revisited. Protein Sci. 11: 2774–2791.
Chen, Y., Song, J., Sui, S.F., and Wang, D.N. 2003. DnaK and DnaJ
facilitated the folding process and reduced inclusion body formation of
magnesium transporter CorA overexpressed in Escherichia coli. Protein
Expr. Purif. 32: 221–231.
Dalbey, R.E. and Wickner, W. 1986. The role of the polar, carboxylterminal domain of Escherichia coli leader peptidase in its translocation
across the plasma membrane. J. Biol. Chem. 261: 13844–13849.
de Cock, H., Overeem, W., and Tommassen, J. 1992. Biogenesis of outer
membrane protein PhoE of Escherichia coli: Evidence for multiple
SecB-binding sites in the mature portion of the PhoE protein. J. Mol.
Biol. 224: 369–379.
Dekker, C., de Kruijff, B., and Gros, P. 2003. Crystal structure of SecB
from Escherichia coli. J. Struct. Biol. 144: 313–319.
Fröderberg, L., Houben, E.N., Baars, L., Luirink, J., and de Gier, J.W.
2004. Targeting and translocation of two lipoproteins in Escherichia
coli via the SRP/Sec/YidC pathway. J. Biol. Chem. 279: 31026–
31032.
Henderson, I.R., Navarro-Garcia, F., and Nataro, J.P. 1998. The great
escape: Structure and function of the autotransporter proteins. Trends
Microbiol. 6: 370–378.
Johnston, S., Lee, J.H., and Ray, D.S. 1985. High-level expression of M13
gene II protein from an inducible polycistronic messenger RNA. Gene
34: 137–145.
Knoblauch, N., Rüdiger, S., Schönfeld, H.-J., Driessen, A., SchneiderMergener, J., and Bukau, B. 1999. Substrate specificity of the SecB
chaperone. J. Biol. Chem. 274: 34219–34225.
Kumamoto, C.A. and Beckwith, J. 1985. Evidence for specificity at an
early step in protein export in Escherichia coli. J. Bacteriol. 163: 267–
274.
Laarmann, S. and Schmidt, M.A. 2003. The Escherichia coli AIDA autotransporter adhesin recognizes an integral membrane glycoprotein as
receptor. Microbiology 149: 1871–1882.
Laskowska, E., Bohdanowicz, J., Kuczynska-Wisnik, D., Matuszewska,
E., Kedzierska, S., and Taylor, A. 2004. Aggregation of heat-shock –
denatured, endogenous proteins and distribution of the IbpA/B and
Fda marker-proteins in Escherichia coli WT and grpE280 cells. Microbiology 150: 247–259.
Melén, K., Krogh, A., and von Heijne, G. 2003. Reliability measures for
membrane protein topology prediction algorithms. J. Mol. Biol. 327:
735–744.
Molloy, M.P., Herbert, B.R., Slade, M.B., Rabilloud, T., Nouwens,
A.S., Williams, K.L., and Gooley, A.A. 2000. Proteomic analysis
of the Escherichia coli outer membrane. Eur. J. Biochem. 267: 2871–
2881.
Oliver, D.B., Cabelli, R.J., Dolan, K.M., and Jarosik, G.P. 1990. Azideresistant mutants of Escherichia coli alter the SecA protein, an azidesensitive component of the protein export machinery. Proc. Natl. Acad.
Sci. 87: 8227–8231.
Pouttu, R., Westerlund-Wikstrom, B., Lang, H., Alsti, K., Virkola, R.,
Saarela, U., Siitonen, A., Kalkkinen, N., and Korhonen, T.K. 2001.
matB, a common fimbrillin gene of Escherichia coli, expressed in a
genetically conserved, virulent clonal group. J. Bacteriol. 183: 4727–
4736.
Powers, E.L. and Randall, L.L. 1995. Export of periplasmic galactosebinding protein in Escherichia coli depends on the chaperone SecB. J.
Bacteriol. 177: 1906–1907.
Randall, L.L. and Hardy, S.J. 2002. SecB, one small chaperone in the
complex milieu of the cell. Cell. Mol. Life Sci. 59: 1617–1623.
Terzi, L., Pool, M.R., Dobberstein, B., and Strub, K. 2004. Signal recognition particle Alu domain occupies a defined site at the ribosomal
subunit interface upon signal sequence recognition. Biochemistry 43:
107–117.
von Heijne, G. 1999. Recent advances in the understanding of membrane
protein assembly and structure. Q. Rev. Biophys. 32: 285–307.
Whitley, P., Nilsson, I., and von Heijne, G. 1994. De novo design of
integral membrane proteins. Nat. Struct. Biol. 1: 858–862.
Xu, Z., Knafels, J.D., and Yoshino, K. 2000. Crystal structure of the
bacterial protein export chaperone secB. Nat. Struct. Biol. 7: 1172–
1177.
Yen, M.R., Peabody, C.R., Partovi, S.M., Zhai, Y., Tseng, Y.H., and
Saier, M.H. 2002. Protein-translocating outer membrane porins of
Gram-negative bacteria. Biochim. Biophys. Acta 1562: 6–31.
www.proteinscience.org
889
THE JOURNAL OF BIOLOGICAL CHEMISTRY
© 2004 by The American Society for Biochemistry and Molecular Biology, Inc.
Vol. 279, No. 30, Issue of July 23, pp. 31026 –31032, 2004
Printed in U.S.A.
Targeting and Translocation of Two Lipoproteins in
Escherichia coli via the SRP/Sec/YidC Pathway*
Received for publication, March 23, 2004, and in revised form, May 12, 2004
Published, JBC Papers in Press, May 12, 2004, DOI 10.1074/jbc.M403229200
Linda Fröderberg‡§, Edith N. G. Houben¶, Louise Baars‡, Joen Luirink¶,
and Jan-Willem de Gier‡储
From the ‡Department of Biochemistry and Biophysics, Arrhenius Laboratories, Stockholm University, SE-106 91
Stockholm, Sweden and the ¶Department of Microbiology, BioCentrum Amsterdam, De Boelelaan 1087,
1081 HV Amsterdam, The Netherlands
In the bacterium Escherichia coli, the SecB pathway targets
a subset of secretory proteins to the Sec translocase (1). The
chaperone SecB keeps secretory proteins in a translocationcompetent state. The SecB-preprotein complex is targeted at a
late stage during translation or after translation to the Sec
translocase in the inner membrane. The signal recognition
particle (SRP)1 pathway targets IMPs to the same or a very
similar Sec translocase in a cotranslational mechanism (2, 3).
The SRP, which consists of the Ffh protein and the 4.5 S RNA,
binds to the first transmembrane segment (TM) of an IMP
* This work was supported by grants from the Swedish Research
Council, the Carl Tryggers Stiftelse, the European Molecular Biology
Organization (EMBO) Young Investigator Program (to J. W. dG.), and
the Dutch Research Council (to J.L.). The costs of publication of this
article were defrayed in part by the payment of page charges. This
article must therefore be hereby marked “advertisement” in accordance
with 18 U.S.C. Section 1734 solely to indicate this fact.
§ Recipient of an EMBO short term fellowship.
储 To whom correspondence should be addressed. Tel.: 46-8-164389
(laboratory)/162420 (office); Fax: 46-8-153679; E-mail: degier@dbb.
su.se.
1
The abbreviations used are: SRP, signal recognition particle; IMP,
inner membrane protein; IMV, inverted membrane vesicle; IPTG, isopropyl-1-thio-␤-D-galactopyranoside; LP, mature lipoprotein; OmpA,
outer membrane protein A; TF, trigger factor; TM, transmembrane
segment; (Tmd)Phe, L-[3-(trifluoromethyl)-3-diazirin-3H-yl]phenylalanine; U-PLP, unmodified prolipoprotein; HA, hemagglutinin; BRP, bacteriocin release protein; Lpp, murein lipoprotein; SPase, signal
peptidase.
when it becomes exposed outside the ribosome. Upon contact of
the SRP with its receptor, FtsY, the nascent IMP dissociates
from the SRP and enters the Sec translocase.
The core of the Sec translocase consists of the IMPs, SecY
and SecE, and the peripheral subunit, SecA (1). SecY and SecE
form a protein-conducting channel (4), and SecA drives proteins in an ATP-dependent process through this channel (1).
The Sec translocase catalyzes linear transport of secretory
proteins and of periplasmic domains of IMPs across the membrane. In addition, TMs of IMPs are recognized in the Sec
translocase and laterally transferred into the lipid bilayer. The
Sec translocase-associated form of YidC appears to assist in
this lateral transfer of TMs (5, 6). YidC, which is present in
excess over the Sec translocase, is also involved in the integration of some small SRP/Sec-independent IMPs (7, 8). Thus far,
no evidence has been obtained that points to a role of YidC in
the targeting/translocation of secretory proteins across the inner membrane (7, 9).2
It should be noted that in E. coli targeting and translocation
of only a handful of secretory proteins and IMPs have been
studied thoroughly. Hardly anything is known about the targeting and translocation of lipoproteins, which in most cases
are secretory proteins. A lipoprotein is synthesized as a preprotein with an N-terminal signal sequence. Lipoproteins contain a conserved sequence, the “lipobox,” that includes the
signal peptidase II (SPase II) cleavage site (10). The cysteine
located just after the SPaseII cleavage site is diacylglycerated
upon translocation; subsequently the signal sequence is clipped
off by SPaseII, yielding an apolipoprotein. Finally, the aminomodified cysteine is fatty acylated, giving rise to the mature
lipoprotein (11). Secretory lipoproteins can, depending on the
sequence of the early mature region, remain associated to the
outer leaflet of the inner membrane or be transported by
the Lol system to the outer membrane (12).
To identify the components involved in the targeting and
translocation of secretory lipoproteins, we have analyzed the
maturation of two model secretory lipoproteins. Maturation of
lipoproteins has been studied in vivo using strains that are
mutated in targeting and translocation factors and in vitro
using a translation/cross-linking system. One of the two model
lipoproteins is the murein lipoprotein (Lpp), which is the most
abundant protein in E. coli. Lpp is attached to the inner leaflet
of the outer membrane of E. coli and forms stable trimers (13).
One of three Lpp molecules is covalently linked to the peptidoglycan layer (13). The other model lipoprotein is the
pCloCF13-encoded bacteriocin release protein (BRP) (14). The
BRP is essential for the translocation of the bacteriocin cloacin
2
31026
E. N. G. Houben and J. Luirink, unpublished observations.
This paper is available on line at http://www.jbc.org
Downloaded from www.jbc.org at Stockholm Universitetsbibliotek on August 6, 2007
In Escherichia coli, two main protein targeting pathways to the inner membrane exist: the SecB pathway for
the essentially posttranslational targeting of secretory
proteins and the SRP pathway for cotranslational targeting of inner membrane proteins (IMPs). At the inner
membrane both pathways converge at the Sec translocase, which is capable of both linear transport into the
periplasm and lateral transport into the lipid bilayer.
The Sec-associated YidC appears to assist the lateral
transport of IMPs from the Sec translocase into the lipid
bilayer. It should be noted that targeting and translocation of only a handful of secretory proteins and IMPs
have been studied. These model proteins do not include
lipoproteins. Here, we have studied the targeting and
translocation of two secretory lipoproteins, the murein
lipoprotein and the bacteriocin release protein, using a
combined in vivo and in vitro approach. The data indicate that both murein lipoprotein and bacteriocin release protein require the SRP pathway for efficient targeting to the Sec translocase. Furthermore, we show
that YidC plays an important role in the targeting/translocation of both lipoproteins.
BRP and LPP Targeting and Translocation
DF13, a bactericidal protein, across the cell envelope (14).
Notably, the BRP signal sequence is very stable after cleavage
from the preprotein, in contrast to other signal sequences, and
plays a yet undefined role in cloacin DF13 export (14).
Our combined in vivo and in vitro studies indicate that the
SRP pathway plays an important role in the targeting of Lpp
and BRP to the Sec translocase. Surprisingly, YidC is also
shown to function in the targeting/translocation of both secretory lipoproteins.
MATERIALS AND METHODS
and the pEH3 vector in strain WAM121. For all experiments cells were
grown to mid-log phase. Expression of the constructs was induced for 3
min with either L-arabinose (0.2%) or IPTG (1 mM). When indicated, the
SPaseII inhibitor globomycin (final concentration 100 ␮g ml⫺1) was
added 5 min before induction. Cells were labeled with [35S]methionine
(60 ␮Ci/ml, Ci ⫽ 37 GBq) for 30 s before precipitation with trichloroacetic acid (final concentration 10%). Subsequently, the samples were
washed with acetone, resuspended in 10 mM Tris/2% SDS, and immunoprecipitated with anti-HA and anti-OmpA serum. Anti-HA immunoprecipitations were analyzed by means of Tricine SDS-PAGE (16.5%
peptide criterion gels from Bio-Rad) and anti-OmpA immunoprecipitations by means of standard SDS-PAGE. Gels were scanned by Fuji
FLA-3000 phosphorimaging using the Image Reader V1.8J/Image Gauge
V 3.45 software.
E. coli Strains, Plasmids, and Growth Conditions for in Vitro Studies—E. coli strain MC1061 grown in Luria Bertani medium supplemented with CaCl2 (10 mM) was used for all plasmid constructions. The
QuikChange site-directed mutagenesis kit was used for the construction of all point mutations. Strain MRE600 was used to prepare a lysate
for translation of in vitro synthesized mRNA and suppression of UAG
stop codons in the presence of (Tmd)Phe-tRNAsup. Inverted membrane
vesicles (IMVs) were prepared from strain MC4100 grown in Luria
Bertani medium.
Plasmid pC4Meth55LppTAG11 (Fig. 1) was constructed by plasmid
PCR and site-directed mutagenesis using pGEM42-Lpp as template
(17). Plasmid pC4Meth55BRPTAG10 was obtained by plasmid PCR
and site-directed mutagenesis using pJL28 as a template (16).
pC4Meth55LppTAG11 encodes truncated Lpp, and pC4Meth55BRPTAG10 encodes truncated BRP. A methionine has been introduced at position 16 in the Lpp signal sequence and at position 17 in the
BRP signal sequence, and both Lpp and BRP have been fused to a
C-terminal 4 ⫻ methionine tag to improve labeling efficiency. Plasmid
pC4Meth55LppTAG11 contains an amber mutation at position 11 in
the Lpp gene, and plasmid pC4Meth55BRPTAG10 contains an amber
mutation (TAG) at position 10 in the BRP gene to enable tRNAsup
photocross-linking (Fig. 1). Where appropriate, ampicillin (100 ␮g ml⫺1)
was added to the medium.
In Vitro Transcription, Translation, Targeting, and Cross-linking—
Truncated mRNA was prepared as described previously from HindIIIlinearized Lpp and BRP derivative plasmids. For photocross-linking,
(Tmd)Phe was site-specifically incorporated into nascent chains by suppression of a UAG stop codon using (Tmd)Phe-tRNAsup in an E. coli in
vitro translation system containing [35S]methionine to label the nascent
chains. This procedure has been described previously (5, 27). Targeting
to IMVs, photocross-linking, and carbonate extraction (to separate soluble and peripheral membrane proteins from integral membrane proteins) was carried out as described previously (27). Carbonate-soluble
and -insoluble fractions were either trichloroacetic acid precipitated or
immunoprecipitated. The material used for immunoprecipitation was
2-fold the amount used for trichloroacetic acid precipitation. Release of
nascent chains from the ribosome was provoked by incubating the
nascent chains after translation for 10 min at 37 °C with EDTA (25
mM). Samples were analyzed using 15% Laemmli SDS-PAGE and phosphorimaging as described previously (27).
RESULTS
Model Lipoproteins—We have used the murein Lpp and the
pCloDF13-encoded BRP as model proteins to study the targeting and translocation of secretory lipoproteins in E. coli (13,
14). To improve the labeling of Lpp and BRP with [35S]methionine, both lipoproteins were slightly modified. A methionine
was introduced in the signal sequences of both Lpp and BRP
(Fig. 1). This does not have a significant impact on the predicted hydrophobicity of the Lpp and BRP signal sequences,
and we felt confident that the introduction of a methionine
would affect Lpp and BRP signal sequence interactions only
marginally at the most. Actually, in the ColA and ColN BRPs,
which are homologous to the pCloDF13 BRP, a methionine
naturally occurs at this position. To further improve labeling, a
stretch of 4 methionines was attached to the very C terminus of
both Lpp and BRP. To be able to immunoprecipitate Lpp and
BRP in the in vivo experiments, the 9-amino acid-long influenza virus HA epitope tag preceded by a flexible linker (ProGly-Gly) was attached to their C termini (Fig. 1) (20).
The modified pCloDF13-encoded BRP causes lysis upon over-
Downloaded from www.jbc.org at Stockholm Universitetsbibliotek on August 6, 2007
Reagents, Enzymes, and Sera—All restriction enzymes, T4 DNA
ligase, and alkaline phosphatase were purchased from Invitrogen. The
Expand long template PCR kit was from Roche Applied Science. The
QuikChange site-directed mutagenesis kit was from Stratagene, and
the Megashort script T7 transcription kit was from Ambion Inc.
[35S]methionine and protein A-Sepharose were from Amersham Biosciences. Pansorbin was obtained from Merck. All other chemicals were
supplied by Sigma. Antiserum against hemagglutinin (HA) tag was
purchased from Sigma and AbCam. Antisera against L23 and L29 were
kind gifts from R. Brimacombe. The antisera against TF and SecA were
gifts from W. Wickner. Antisera against Ffh, YidC, and SecY were from
our own collection.
E. coli Strains, Plasmids, and Growth Conditions for in Vivo Targeting and Translocation Studies—E. coli strain TOP10F⬘ grown in Luria
Bertani medium was used for all plasmid constructions. In all lipoprotein expression vectors, the pCloDF13-encoded T1 terminator, which
regulates the expression of the pCloDF13-encoded BRP, was cloned
upstream of the genes encoding Lpp and BRP to prevent any background expression (15). Upon induction of expression, the T1 terminator is “overruled.” The pCloDF13-encoded T1 terminator region was
amplified using pJL28 as a template (16). Primers were designed so
that both an NcoI site and a stop codon were introduced upstream and
an EcoRI site and a ribosome binding site were introduced downstream
of the T1 terminator region. The T1 terminator region was cloned
NcoI-EcoRI into pET21d, yielding pET21d-T1. Plasmids pET21d-T1Lpp
and pET21d-T1BRP were obtained by plasmid PCR and site-directed
mutagenesis using pGEM42-Lpp and pJL28, respectively, as templates
(17). A methionine codon was introduced at position 16 in the Lpp signal
sequence and position 17 in the BRP signal sequence (Fig. 1). In addition, a C-terminal 4 ⫻ methionine tag was attached to both Lpp and
BRP to increase labeling efficiency. Subsequently, T1Lpp and T1BRP
were NcoI-HindIII-cloned into pEH1 (18), pEH3 (18), and pBAD24 (19),
yielding pEH1-T1Lpp, pEH3-T1Lpp, pBAD24-T1Lpp, pEH1-T1BRP,
pEH3-T1BRP, and pBAD24-T1BRP. Finally, the genetic information
coding for an HA tag with a stop codon at its 3⬘ prime end (20), preceded
by a flexible linker (Pro-Gly-Gly) was fused to the 4 ⫻ methionine-tagged
Lpp and BRP, using BamHI and HindIII sites. This yielded pEH1T1LppHA, pEH3-T1LppHA, pBAD24-T1LppHA, pEH1-T1BRPHA,
pEH3-T1BRPHA, and pBAD24-T1BRPHA. The nucleotide sequences of
all constructs were verified by DNA sequencing.
The Ffh conditional strain WAM121 was cultured in M9 minimal
medium supplemented with 0.2% arabinose as described previously
(21). To deplete cells for Ffh, cells were grown to mid-log phase in the
absence of arabinose. The 4.5 S RNA conditional strain FF283 was
cultured in M9 minimal medium supplemented with 1 mM IPTG as
described previously (21). To deplete cells for 4.5 S RNA, cells were
grown to mid-log phase in the absence of IPTG. The temperaturesensitive amber suppressor SecA depletion strain BA13 and the control
strain DO251 were cultured in M9 minimal medium at 30 °C as described previously (22, 23). To deplete cells for SecA, cells were grown to
mid-log phase at 41 °C. The SecE depletion strain CM124 was cultured
in M9 minimal medium supplemented with 0.2% glucose and 0.2%
L-arabinose as described previously (24, 25). To deplete cells for SecE,
cells were grown to mid-log phase in the absence of L-arabinose. The
temperature-sensitive amber suppressor YidC depletion strain
KO1672, along with its wild-type control strain KO1670 (26), were
cultured in M9 minimal medium at 30 °C overnight. To deplete KO1672
cells for YidC, cells were grown to mid-log phase at 42 °C. Where
appropriate, ampicillin (100 ␮g ml⫺1), chloramphenicol (30 ␮g ml⫺1),
kanamycin (50 ␮g ml⫺1), and tetracyclin (12.5 ␮g ml⫺1) were added to
the medium.
In Vivo Assay for Targeting and Translocation—The model lipoproteins Lpp and BRP were expressed by L-arabinose induction from the
pBAD24 vector in strains TOP10F⬘, FF283, BA13, DO251, KO1672, and
KO1670 and by IPTG induction from the pEH1 vector in strain CM124
31027
31028
BRP and LPP Targeting and Translocation
FIG. 1. The model lipoproteins Lpp and BRP. The amino acid sequence of BRP, Lpp, and their derivatives used in this study: wild-type BRP
(WTBRP), in vitro construct 55BRPTAG10, in vivo construct BRP, wild-type Lpp (WTLPP), in vitro construct 54LPPTAG11, in vivo construct Lpp
(LPP). The SPaseII cleavage site is indicated by an arrow, the lipid modifiable cystein by C, inserted/added methionines by M, and attached
influenza virus hemagglutinin epitope tags with a black background. Between the methionine stretch and the HA tag there is a flexible linker
(PGG). Amber codons and stop codons are marked with an asterisk.
FIG. 2. Translocation of Lpp and BRP is affected by the SPaseII inhibitor globomycin and is Sec translocase-dependent.
TOP10F⬘ cells harboring either pBAD24-T1LppHA (top panel, lanes 1
and 2) or pBAD24-T1BRPHA (bottom panel, lanes 1 and 2) were cultured in M9 medium and pulse-labeled in the absence and presence of
the SPaseII inhibitor, globomycin. CM124 (PAra-secE) cells harboring
either pEH1-T1LppHA (top panel, lanes 3 and 4) or pEH1-T1BRPHA
(bottom panel, lanes 3 and 4) were cultured in M9 medium with (⫹SecE)
or without (⫺SecE) L-arabinose. BA13 (SecA depletion strain, ⫺SecA)
cells and DO251 (control strain, ⫹SecA) cells harboring either pBAD24T1LppHA (top panel, lanes 5 and 6) or pBAD24-T1BRPHA (bottom
panel, lanes 5 and 6) were cultured in M9 medium at 41 °C (SecA
depletion conditions, ⫺SecA). Cells were pulse-labeled and processed as
described under ‘‘Materials and Methods.’’ Lpp and BRP were immunoprecipitated with anti-HA antiserum. The processing of Lpp and BRP
was not affected by Me2SO (DMSO) in which globomycin was dissolved.
The processing of the SpaseII-independent OmpA was not affected in
the presence of globomycin, and depletion of SecE and SecA was
checked by monitoring the accumulation of OmpA (results not shown).
No effect of depletion of the SRP components could be detected,
confirming that the observed accumulation of U-PLP is not
because of more general secondary effects of SRP depletion.
Together, the results indicate that a functional SRP pathway is
required for efficient targeting of the BRP and, albeit to a lesser
extent, of the Lpp.
Efficient in Vivo Translocation of Lpp and BRP Requires
YidC—All IMPs studied so far require YidC for efficient assembly into the inner membrane. It has been suggested that YidC
assists the transfer of TMs from the Sec translocase into the
lipid bilayer (5). So far, no evidence has been obtained pointing
to a role of YidC in the translocation of secretory proteins (6, 7,
27, 37, 38). However, the unexpected role of the SRP in the
targeting of BRP and Lpp prompted us to evaluate the role of
YidC in the translocation of these proteins using a temperature-sensitive strain that is conditional for YidC expression
(Fig. 4). To our surprise, depletion of YidC by growth at the
non-permissive temperature resulted in the accumulation of
unmodified precursor forms of BRP (most pronounced) and Lpp
(less pronounced). Again, the processing of pro-OmpA that is
translocated independent of YidC (7) was monitored as a control and appeared unaffected. The combined data suggest that
YidC plays a differential role in the translocation of both
lipoproteins.
Downloaded from www.jbc.org at Stockholm Universitetsbibliotek on August 6, 2007
expression, just like the wild-type version of the protein, and
the clipped off signal sequence is stable (results not shown). For
the sake of clarity, we refer in the rest of this report to the
modified lipoproteins as Lpp and BRP.
Translocation of Lpp and BRP across the Inner Membrane Is
Sec Translocase-dependent—To test whether the introduction
of the extra methionines and the HA tag interfere with the in
vivo processing and maturation of Lpp and BRP, the constructs
were studied in the E. coli TOP10F⬘ strain and Sec translocase
mutant strains. Maturation of lipoproteins occurs in three
steps: 1) the unmodified prolipoprotein (U-PLP) is converted
into a diacylated prolipoprotein (M-PLP) upon translocation
across the inner membrane, 2) cleavage of the signal sequence
by SPaseII yields the apolipoprotein, and 3) acylation of the
lipoprotein gives rise to the mature lipoprotein (11). Both Lpp
and BRP were expressed in the TOP10F⬘ strain in the absence
and presence of globomycin, which inhibits SPaseII. Inhibition
of SPaseII causes the accumulation of the M-PLP form of lipoproteins (Fig. 2, lanes 1 and 2) (28). This was confirmed in
experiments where [3H]palmitate was used rather than
[35S]methionine to label cells in the presence of globomycin
(results not shown). Taken together, the modifications improve
labeling and facilitate immunoprecipitation of Lpp and BRP
without affecting their maturation.
Lpp and BRP were also expressed in SecE and SecA depletion strains. Both SecE and SecA are key components of the Sec
translocase. Upon depletion of SecE, SecY is rapidly degraded
by the FtsH protease. Therefore, SecE depletion results in the
loss of the SecY/E core of the Sec translocase (29). SecA depletion results in the loss of the motor of the Sec translocase (22).
Upon SecE and SecA depletion, the U-PLP form of both Lpp
and BRP accumulate (Fig. 2, lanes 3– 6). This points to a key
role of the Sec translocase in the translocation of Lpp and BRP
across the inner membrane, corroborating previous studies
using Sec-conditional mutant strains (30 –33).
Efficient in Vivo Targeting of Lpp and BRP Requires SRP—
How are Lpp and BRP targeted to the Sec translocase? The
SecB and the SRP pathways are the two main targeting pathways to the Sec translocase (34). In the absence of SecB, targeting of both Lpp and BRP is not significantly affected (results
not shown).
We next investigated the role of the SRP in the targeting of
Lpp and BRP to the Sec translocase. The E. coli SRP consists of
the protein component Ffh and the RNA component 4.5 S RNA.
Both Ffh and 4.5 S RNA are essential for viability, and depletion of either of the SRP components compromises the SRP
targeting pathway, thereby preventing the targeting of many
IMPs (3, 35, 36). Targeting of Lpp and BRP was studied both
under 4.5 S RNA and Ffh depletion conditions (Fig. 3, A and B).
Depletion of 4.5 S RNA (Fig. 3A) and of Ffh (Fig. 3B) both
resulted in accumulation of the unmodified precursor forms of
BRP (most pronounced) and Lpp (less pronounced). As a control, the processing of pro-OmpA, an outer membrane protein
that is targeted by SecB, was monitored in the same samples.
BRP and LPP Targeting and Translocation
FIG. 4. YidC is required for efficient targeting of Lpp and BRP.
KO1670 (control strain, ⫹YidC) cells and KO1672 (YidC depletion
strain, ⫺YidC) cells were cultured in M9 medium at 42 °C (YidC depletion conditions) and 30 °C (non-depletion conditions). Cells were
pulse-labeled and processed as described under ‘‘Materials and Methods.’’ Lpp and BRP were immunoprecipitated with anti-HA antiserum
and OmpA with OmpA antiserum. OmpA processing was not affected
upon YidC depletion. YidC depletion was monitored by means of Western blotting and resulted, as expected, in a strong PspA response (Ref.
56 and results not shown).
Nascent Lpp and BRP Synthesized in Vitro Cross-link to Ffh,
SecA, SecY, and YidC—To study the targeting and translocation of Lpp and BRP in more detail, we have used an in vitro
translation/photo cross-linking approach. In this assay, the
interactions of nascent (ribosome-associated) polypeptides with
cytosolic and membrane components are fixed and analyzed.
[35S]Methionine-radiolabeled nascent chains of Lpp and BRP
were synthesized in an E. coli cell-free extract from truncated
mRNA to a length of 55 amino acids. Assuming that the ribosome covers ⬃35 amino acids, the Lpp and BRP signal sequences are expected to be exposed just outside the ribosome
(37). To specifically probe the molecular environment of the
signal sequence in the nascent Lpp and BRP species, a single
amber stop codon (TAG) was introduced in the center of the
hydrophobic core in the Lpp (position 11) and BRP (position 10)
signal sequences (Fig. 1). The amber stop codons were suppressed during the in vitro translation by addition of (Tmd)Phe-tRNAsup, an amber suppressor tRNA that is amino-acylated with the photo cross-linker (Tmd)Phe. In all constructs,
the TAGs were efficiently suppressed by (Tmd)Phe-tRNAsup
(data not shown). Purified inverted IMVs were added from the
start of the translation reaction to allow cotranslational membrane targeting and interaction of the translation intermediates with the membrane. After the translation/insertion reaction, one half of each sample was irradiated with UV light to
induce cross-linking; the other half was kept in the dark to
serve as a control. The samples were extracted with carbonate
to separate soluble and peripherally membrane-associated material from membrane-integrated components. Cross-linking
partners were identified by immunoprecipitation. Without UV
irradiation, no cross-linking products were detected using both
Lpp and BRP nascent chains (data not shown). Upon translation but prior to UV irradiation, the samples were divided in
two, and EDTA was added to one aliquot to provoke the release
of the nascent chains from the ribosome. This allowed us to
assess the importance of the context of the ribosome for crosslinking to the truncated Lpp and BRP species. EDTA has been
shown to disassemble ribosomes (37).
Both nascent Lpp and BRP were efficiently targeted to the
IMVs, judging from the relatively high (⬃50%) carbonate resistance. When the carbonate supernatant of UV-irradiated
samples was analyzed, Lpp nascent chains were shown to
cross-link Ffh, albeit inefficiently (Fig. 5A, lane 3), and the
chaperone trigger factor (TF, lane 4). The ⬃30-kDa cross-linking adducts represented cross-linking to a breakdown product
of Ffh, as observed before (39).
Cross-linking to both Ffh and TF appeared dependent on the
context of the ribosome (lanes 8, 9) consistent with earlier
studies (39, 40). In contrast, cross-linking to SecA was hardly
detectable unless the nascent chains were released from the
ribosomes prior to cross-linking (lanes 7, 10). Strong crosslinking specific for the ribosome-associated Lpp was observed
to the ribosomal proteins L23 and L29 (lanes 5, 6). L23 and L29
are located near the exit site of the large ribosomal tunnel that
runs from the peptidyl transferase center to the surface of the
large ribosomal subunit (41). In the carbonate pellet, crosslinking to SecY and (weakly) to YidC was observed that appeared dependent on the ribosomal context (lanes 13, 14, 16,
17), again consistent with earlier studies (5, 37). In addition, in
the pellet fractions SecA cross-linking was detectable only upon
release of nascent Lpp from the ribosome. Together, the data
suggest that nascent Lpp leaves the ribosome via the major exit
tunnel near L23 and L29. Most likely, the signal sequence of a
small fraction of nascent Lpp contacts the SRP. However, the
majority of nascent Lpp is close to TF. Consistent with this
explanation, both the SRP and TF dock near L23/L29 on the
ribosome, probably in a mutually exclusive manner. Upon
forced release of the nascent chains from the ribosome, the
signal peptide loses contact with L23, L29, TF, and Ffh and is
free to bind SecA, part of which is carbonate-resistant. Nascent
Lpp is primarily targeted to SecY but also contacts YidC.
Qualitatively, very similar results were obtained when nascent BRP was used instead of Lpp (Fig. 5B). However, crosslinking to Ffh appeared much more prominent at the expense
of cross-linking to TF (lanes 1, 3, 7), especially when considering the small amount of SRP present in cells as compared with
TF (42, 43). Upon treatment with EDTA, cross-linking to Ffh
was no longer detectable (lane 8), but another unknown factor
Downloaded from www.jbc.org at Stockholm Universitetsbibliotek on August 6, 2007
FIG. 3. The SRP pathway is required for efficient targeting of
Lpp and BRP. A, FF283 (PIPTG-4.5 S RNA) cells were cultured in M9
medium with (⫹4.5 S RNA) or without (⫺4.5 S RNA) IPTG. Cells were
pulse-labeled and processed as described under ‘‘Materials and Methods.’’ Lpp and BRP were immunoprecipitated with anti-HA antiserum
and OmpA with OmpA antiserum. OmpA processing was not affected
upon 4.5 S RNA depletion. B, WAM121 (Para-ffh) cells harboring either
pEH3T1LppHA or pEH3T1BRPHA were cultured in M9 medium with
(⫹Ffh) or without (⫺Ffh) arabinose. Cells were pulse-labeled and processed as described under ‘‘Materials and Methods.’’ Lpp and BRP were
immunoprecipitated with anti-HA antiserum and OmpA with OmpA
antiserum. OmpA processing was not affected upon Ffh depletion.
31029
31030
BRP and LPP Targeting and Translocation
of about the same molecular mass (⬃50 kDa) was cross-linked
(lane 2). Other notable differences were that ribosome-associated BRP also detectably cross-linked SecA (lane 1, 6), and
there was stronger cross-linking of targeted nascent BRP to
YidC (lane 11). Finally, a cross-linking adduct of 75 kDa was
observed in the carbonate pellet fractions both before and after
release of nascent BRP from the ribosome, which remains to be
identified (lanes 11, 12). The more prominent contacts with the
SRP and YidC suggest a more important role for these factors
in the targeting and translocation of BRP, corroborating the
in vivo data.
DISCUSSION
Here, we have studied in E. coli the targeting and translocation of two secretory lipoproteins, Lpp and BRP, using a
combined in vivo and in vitro approach. Surprisingly, the signal
peptides of both nascent BRP and, to a lesser extent, Lpp show
cross-linking to the targeting factor SRP and to the Sec-associated YidC, which are thought to function in the membrane
targeting and integration of integral IMPs. Consistent with
Downloaded from www.jbc.org at Stockholm Universitetsbibliotek on August 6, 2007
FIG. 5. Interactions of nascent Lpp
and BRP. A, 55LppTAG11 was translated in the presence of IMVs and (Tmd)Phe-tRNAsup as described under ‘‘Materials and Methods.’’ After translation, the
nascent chains were treated with EDTA
when indicated, UV irradiated, and subsequently extracted with sodium carbonate (Supernatant, lanes 1–10; Pellet,
lanes 11–18). UV-irradiated fractions
were immunoprecipitated using antiserum against Ffh, TF, L23, L29, SecA,
SecY, and YidC (lanes 3–10, 13–18). All
cross-linking adducts are indicated with
⬎. B, in vitro translation, EDTA treatment, cross-linking, and carbonate extraction of nascent 55BRPTAG10 were
carried out as described under panel A.
these in vitro results, BRP and, to a lesser extent, Lpp depend
on the presence of the SRP and YidC for efficient targeting to
and translocation across the inner membrane in vivo.
In vitro, Lpp and BRP nascent chains with a length of 55
amino acids, carrying a UV-inducible cross-linker in the middle
of the signal sequence, are also cross-linked to the ribosomal
components L23 and L29 and to the ribosome-associated chaperone TF. L23 and L29 are located near the exit of the presumed ribosomal tunnel (41) and have been cross-linked to
short nascent chains of other origin before (39, 44). L23 has
recently been shown to function as an attachment site for both
TF and SRP (39, 45, 46). Cross-link studies have identified TF
as the first chaperone to interact generically with nascent
polypeptides (47) unless they carry a particularly hydrophobic
targeting signal that has a high affinity for the SRP (39, 40, 48).
The mechanism that underlies the interplay between TF and
SRP at the nascent chain exit site and how this interplay
influences the mode of membrane targeting and insertion of a
particular nascent protein are still unresolved issues (39, 44).
BRP and LPP Targeting and Translocation
3
L. Baars and J.-W. de Gier, manuscript in preparation.
in vivo depletion of YidC affects the maturation of BRP and
Lpp. In the context of the Sec translocon, YidC is considered to
facilitate the transfer of TMs of IMPs from the Sec translocase
into the lipid bilayer without being essential for this process.
The role of YidC in the targeting/translocation of Lpp and BRP
is enigmatic. It is possible that YidC facilitates the lateral
movement of the Lpp and BRP signal sequences into the lipid
bilayer or that YidC chaperones Lpp and BRP to the SPaseII/
Lol system. It is also possible that there is a direct connection
between SRP dependence and YidC dependence. It has been
shown that the chloroplast homologues of SRP and YidC participate in targeting complexes (53). Therefore, it cannot be
excluded that YidC plays a role in the SRP cycle, perhaps in the
reception of the SRP or FtsY. Interestingly, membranes isolated from cells depleted of YidC show decreased levels of the
lipoprotein CyoA, which in contrast to Lpp and BRP is an
integral IMP (54, 55). This may point to a more general and not
yet understood role of YidC in the targeting/translocation of
lipoproteins. In conclusion, our results indicate that the SRP/
Sec/YidC pathway is used by the secretory lipoproteins Lpp
and BRP.
REFERENCES
1. Manting, E. H., and Driessen, A. J. (2000) Mol. Microbiol. 37, 226 –238
2. Herskovits, A. A., Bochkareva, E. S., and Bibi, E. (2000) Mol. Microbiol. 38,
927–939
3. de Gier, J. W., and Luirink, J. (2001) Mol. Microbiol. 40, 314 –322
4. Van den Berg, B., Clemons, W. M., Jr., Collinson, I., Modis, Y., Hartmann, E.,
Harrison, S. C., and Rapoport, T. A. (2004) Nature 427, 36 – 44
5. Urbanus, M. L., Scotti, P. A., Froderberg, L., Saaf, A., de Gier, J. W., Brunner,
J., Samuelson, J. C., Dalbey, R. E., Oudega, B., and Luirink, J. (2001)
EMBO Rep. 2, 524 –529
6. Beck, K., Eisner, G., Trescher, D., Dalbey, R. E., Brunner, J., and Müller, M.
(2001) EMBO Rep. 2, 709 –714
7. Samuelson, J. C., Chen, M., Jiang, F., Moller, I., Wiedmann, M., Kuhn, A.,
Phillips, G. J., and Dalbey, R. E. (2000) Nature 406, 637– 641
8. Serek, J., Bauer-Manz, G., Struhalla, G., Van Den Berg, L., Kiefer, D., Dalbey,
R., and Kuhn, A. (2004) EMBO J. 23, 294 –301
9. Sijbrandi, R., Urbanus, M. L., Ten Hagen-Jongman, C. M., Bernstein, H. D.,
Oudega, B., Otto, B. R., and Luirink, J. (2003) J. Biol. Chem. 278,
4654 – 4659
10. Juncker, A. S., Willenbrock, H., Von Heijne, G., Brunak, S., Nielsen, H., and
Krogh, A. (2003) Protein Sci. 12, 1652–1662
11. Sankaran, K., and Wu, H. C. (1994) J. Biol. Chem. 269, 19701–19706
12. Hara, T., Matsuyama, S., and Tokuda, H. (2003) J. Biol. Chem. 278,
40408 – 40414
13. Shu, W., Liu, J., Ji, H., and Lu, M. (2000) J. Mol. Biol. 299, 1101–1112
14. van der Wal, F. J., Luirink, J., and Oudega, B. (1995) FEMS Microbiol. Rev. 17,
381–399
15. van den Elzen, P. J., Walters, H. H., Veltkamp, E., and Nijkamp, H. J. (1983)
Nucleic Acids Res. 11, 2465–2477
16. Luirink, J., Duim, B., de Gier, J. W., and Oudega, B. (1991) Mol. Microbiol. 5,
393–399
17. Valent, Q. A., Kendall, D. A., High, S., Kusters, R., Oudega, B., and Luirink, J.
(1995) EMBO J. 14, 5494 –5505
18. Hashemzadeh-Bonehi, L., Mehraein-Ghomi, F., Mitsopoulos, C., Jacob, J. P.,
Hennessey, E. S., and Broome-Smith, J. K. (1998) Mol. Microbiol. 30,
676 – 678
19. Siegele, D. A., and Hu, J. C. (1997) Proc. Natl. Acad. Sci. U. S. A. 94,
8168 – 8172
20. Chen, Y. T., Holcomb, C., and Moore, H. P. (1993) Proc. Natl. Acad. Sci.
U. S. A. 90, 6508 – 6512
21. de Gier, J. W., Mansournia, P., Valent, Q. A., Phillips, G. J., Luirink, J., and
von Heijne, G. (1996) FEBS Lett. 399, 307–309
22. Qi, H. Y., and Bernstein, H. D. (1999) J. Biol. Chem. 274, 8993– 8997
23. Froderberg, L., Houben, E., Samuelson, J. C., Chen, M., Park, S. K., Phillips,
G. J., Dalbey, R., Luirink, J., and De Gier, J. W. (2003) Mol. Microbiol. 47,
1015–1027
24. Traxler, B., and Murphy, C. (1996) J. Biol. Chem. 271, 12394 –12400
25. de Gier, J. W., Scotti, P. A., Sääf, A., Valent, Q. A., Kuhn, A., Luirink, J., and
von Heijne, G. (1998) Proc. Natl. Acad. Sci. U. S. A. 95, 14646 –14651
26. Hansen, F. G., Hansen, E. B., and Atlung, T. (1985) Gene 38, 85–93
27. Scotti, P. A., Urbanus, M. L., Brunner, J., de Gier, J. W., von Heijne, G., van
der Does, C., Driessen, A. J., Oudega, B., and Luirink, J. (2000) EMBO J.
19, 542–549
28. Inukai, M., Takeuchi, M., Shimizu, K., and Arai, M. (1978) J. Antibiot. (Tokyo)
31, 1203–1205
29. Akiyama, Y., Kihara, A., Tokuda, H., and Ito, K. (1996) J. Biol. Chem. 271,
31196 –31201
30. Hayashi, S., and Wu, H. C. (1985) J. Bacteriol. 161, 949 –954
31. Oudega, B., Mol, O., van Ulsen, P., Stegehuis, F., van der Wal, F. J., and
Luirink, J. (1993) J. Bacteriol. 175, 1543–1547
32. Sugai, M., and Wu, H. C. (1992) J. Bacteriol. 174, 2511–2516
33. Watanabe, T., Hayashi, S., and Wu, H. C. (1988) J. Bacteriol. 170, 4001– 4007
34. Valent, Q. A., Scotti, P. A., High, S., de Gier, J. W., von Heijne, G., Lentzen, G.,
Downloaded from www.jbc.org at Stockholm Universitetsbibliotek on August 6, 2007
The SRP is primarily used for the targeting of IMPs that are
thought to benefit from a cotranslational insertion mechanism
to prevent aggregation of hydrophobic domains in the cytoplasm. Then why does the SRP appear to play such an important role in the targeting of the secretory lipoprotein BRP?
First, the BRP signal sequence is very hydrophobic, more hydrophobic than many signal anchor sequences of SRP-dependent IMPs and prone to aggregation if unprotected. Interaction
of the signal sequence with the SRP and cotranslational targeting of BRP may be the best way to prevent uncontrolled
insertion of the hydrophobic BRP signal sequence into the
inner membrane and aggregation of BRP in the cytoplasm.
Recently, it has been suggested that basic amino acids in the N
region of a signal sequence contribute to SRP binding, probably
through the formation of salt bridges between the 4.5 S RNA
and positively charged amino acids in the N region (49). Strikingly, the BRP signal sequence has 3 lysines in its N region,
which may enhance even more the affinity of the already very
hydrophobic BRP signal sequence for the SRP. These features
may compensate for the relatively short time window in which
the SRP can interact with nascent BRP, given that the BRP is
a very small protein. It should be noted that the BRP signal
peptide is peculiar in the sense that it is stable in the inner
membrane, whereas other cleaved signal peptides are rapidly
degraded upon cleavage from the precursor protein. Apparently, the BRP signal peptide is recognized as a signal anchor
sequence of an IMP: it binds the SRP in the cytosol, inserts at
the Sec translocon, and is subsequently transferred to the lipid
bilayer as a stably folded unit assisted by YidC (see below).
The signal sequence of Lpp is not very hydrophobic, and its N
region contains only one positively charged amino acid. However, it does show (weak) SRP cross-linking and SRP dependence in vivo. There are recent precedents of relatively nonhydrophobic signal peptides that are yet able to funnel
passenger proteins into the SRP pathway (50). Therefore, it is
not unlikely that there are yet unknown features of a signal
sequence that can provoke SRP binding. There may be an
important biological reason for a preference of Lpp for the
SRP-targeting pathway. Lpp is the most abundant protein in E.
coli and, therefore, highly expressed. When the mature part of
Lpp is expressed in the cytoplasm it forms stable trimers (13).
It is conceivable that cotranslational targeting of Lpp via the
SRP pathway prevents the formation of Lpp trimers in the
cytoplasm, which would make it incompetent for translocation.
The partial effect of SRP depletion suggests that Lpp can also
travel via SecB. The lack of SecB did not significantly affect the
kinetics of Lpp translocation (32).3 However, pre-Lpp could be
detected in cytosolic aggregates isolated from a SecB null
strain.3 A similar flexibility in targeting has recently been
demonstrated for the autotransporter Hbp (9). Therefore, it is
not unlikely that when the capacity of the SRP pathway is
insufficient, the SecB pathway gets a more prominent role in
the targeting of Lpp and, perhaps, also BRP. Unfortunately, we
have not succeeded in studying the targeting of Lpp and BRP in
an Ffh depletion/SecB null mutant background. Therefore,
other modes of targeting, like spontaneous or mRNA targeting,
cannot be excluded at present (3, 51).
Cross-linking of Lpp and BRP nascent chains to SecA and
SecY and the almost completely blocked maturation of both
lipoproteins under SecA and SecE depletion conditions clearly
show that both Lpp and BRP are translocated by the Sec
translocase across the inner membrane, corroborating previous
in vivo studies (30 –33, 52). Surprisingly, both BRP and, to
some extent, Lpp nascent chains are cross-linked to YidC, and
31031
31032
BRP and LPP Targeting and Translocation
Wintermeyer, W., Oudega, B., and Luirink, J. (1998) EMBO J. 17,
2504 –2512
35. Ribes, V., Romisch, K., Giner, A., Dobberstein, B., and Tollervey, D. (1990) Cell
63, 591– 600
36. Phillips, G. J., and Silhavy, T. J. (1992) Nature 359, 744 –746
37. Houben, E. N., Urbanus, M. L., Van Der Laan, M., Ten Hagen-Jongman, C. M.,
Driessen, A. J., Brunner, J., Oudega, B., and Luirink, J. (2002) J. Biol.
Chem. 277, 35880 –35886
38. Drew, D., Fröderberg, L., Baars, L., and de Gier, J. W. (2003) Biochim Biophys
Acta 1610, 3–10
39. Ullers, R. S., Houben, E. N., Raine, A., ten Hagen-Jongman, C. M., Ehrenberg,
M., Brunner, J., Oudega, B., Harms, N., and Luirink, J. (2003) J. Cell Biol.
161, 679 – 684
40. Beck, K., Wu, L. F., Brunner, J., and Müller, M. (2000) EMBO J. 19, 134 –143
41. Ban, N., Nissen, P., Hansen, J., Moore, P. B., and Steitz, T. A. (2000) Science
289, 905–920
42. Lill, R., Crooke, E., Guthrie, B., and Wickner, W. (1988) Cell 54, 1013–1018
43. Jensen, C. G., and Pedersen, S. (1994) J. Bacteriol. 176, 7148 –7154
44. Eisner, G., Koch, H. G., Beck, K., Brunner, J., and Muller, M. (2003) J. Cell
Biol. 163, 35– 44
45. Gu, S. Q., Peske, F., Wieden, H. J., Rodnina, M. V., and Wintermeyer, W.
(2003) RNA 9, 566 –573
46. Kramer, G., Rauch, T., Rist, W., Vorderwulbecke, S., Patzelt, H., Schulze-
Specking, A., Ban, N., Deuerling, E., and Bukau, B. (2002) Nature 419,
171–174
47. Patzelt, H., Rudiger, S., Brehmer, D., Kramer, G., Vorderwulbecke, S., Schaffitzel, E., Waitz, A., Hesterkamp, T., Dong, L., Schneider-Mergener, J.,
Bukau, B., and Deuerling, E. (2001) Proc. Natl. Acad. Sci. U. S. A. 98,
14244 –14249
48. Valent, Q. A., de Gier, J. W., von Heijne, G., Kendall, D. A., ten HagenJongman, C. M., Oudega, B., and Luirink, J. (1997) Mol. Microbiol. 25,
53– 64
49. Peterson, J. H., Woolhead, C. A., and Bernstein, H. D. (2003) J. Biol. Chem.
50. Schierle, C. F., Berkmen, M., Huber, D., Kumamoto, C., Boyd, D., and Beckwith, J. (2003) J. Bacteriol. 185, 5706 –5713
51. Bernstein, H. D., and Hyndman, J. B. (2001) J. Bacteriol. 183, 2187–2197
52. Tian, G., Wu, H. C., Ray, P. H., and Tai, P. C. (1989) J. Bacteriol. 171,
1987–1997
53. Moore, M., Goforth, R. L., Mori, H., and Henry, R. (2003) J. Cell Biol. 162,
1245–1254
54. Chepuri, V., Lemieux, L., Au, D. C., and Gennis, R. B. (1990) J. Biol. Chem.
265, 11185–11192
55. Ma, J., Katsonouri, A., and Gennis, R. B. (1997) Biochemistry 36, 11298 –11303
56. van der Laan, M., Urbanus, M. L., Ten Hagen-Jongman, C. M., Nouwen, N.,
Oudega, B., Harms, N., Driessen, A. J., and Luirink, J. (2003) Proc. Natl.
Acad. Sci. U. S. A. 100, 5801–5806
Downloaded from www.jbc.org at Stockholm Universitetsbibliotek on August 6, 2007
Supplemental Material can be found at:
http://www.jbc.org/cgi/content/full/M509929200/DC1
THE JOURNAL OF BIOLOGICAL CHEMISTRY VOL. 281, NO. 15, pp. 10024 –10034, April 14, 2006
© 2006 by The American Society for Biochemistry and Molecular Biology, Inc. Printed in the U.S.A.
Defining the Role of the Escherichia coli Chaperone SecB
Using Comparative Proteomics*□
S
Received for publication, September 8, 2005, and in revised form, December 9, 2005 Published, JBC Papers in Press, December 13, 2005, DOI 10.1074/jbc.M509929200
Louise Baars‡, A. Jimmy Ytterberg§, David Drew‡, Samuel Wagner‡, Claudia Thilo¶, Klaas Jan van Wijk§,
and Jan-Willem de Gier‡1
From the ‡Department of Biochemistry and Biophysics, Arrhenius Laboratories, Stockholm University SE-106 91 Stockholm,
Sweden, the §Department of Plant Biology, Cornell University, Ithaca, New York 14853, and the ¶Center for Infectious Medicine,
Karolinska Institute, Karolinska University Hospital Huddinge, SE-141 86 Stockholm, Sweden
The periplasmic and outer membrane proteins in the Gram-negative
bacterium Escherichia coli need to cross the cytoplasmic membrane
to reach their final destination. The vast majority of these secretory
proteins are translocated through the cytoplasmic membrane via the
Sec-translocase (1, 2). The core of the Sec-translocase is comprised
of integral membrane proteins SecY and SecE, which form a protein
conducting channel (3). The peripheral subunit SecA drives polypeptide chains in an ATP-dependent manner into and through the
Sec-translocase (1).
It is generally assumed that secretory proteins in E. coli are targeted to
the Sec-translocase by the cytoplasmic protein SecB in a mostly posttranslational fashion (4 – 8). However, direct evidence for SecB dependence is only established for six secretory proteins (PhoE, LamB, MBP,
* This work was supported by grants from the Swedish Research Council, Carl Tryggers
Stiftelse, Marianne and Marcus Wallenberg Foundation, and the EMBO Young Investigator Programme (to J.-W. d. G.), a grant from The Swedish Foundation for International Cooperation in Research and Higher Education (STINT) (to J.-W. d. G. and
K. J. v. W.), and proteomics infrastructure was supported by a grant from New York
State Office of Science, Technology and Academic Research (to K. J. v. W.). The costs of
publication of this article were defrayed in part by the payment of page charges. This
article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C.
Section 1734 solely to indicate this fact.
□
S
The on-line version of this article (available at http://www.jbc.org) contains supplemental Tables S1–S4.
1
To whom correspondence should be addressed. Tel.: 46-8-162420; Fax: 46-8-153679;
E-mail: [email protected].
10024 JOURNAL OF BIOLOGICAL CHEMISTRY
OmpF, GBP, and OmpA), whereas four secretory proteins (PhoA, Lpp,
RbsB, and ␤-lac) do not seem to require SecB (9 –12, 55). SecB also has
the capacity to assist the chaperone DnaK in the folding of proteins, as
shown in vitro with luciferase as a model substrate (12). This indicates
that SecB has the potential to assist the folding of cytoplasmic proteins.
The successful complementation of a DnaK/trigger factor (TF)2 double
mutant strain by overexpression of SecB, and cross-linking of SecB to
nascent chains of both secretory and cytoplasmic proteins in SecBenriched lysates support this notion (13).
SecB does not bind to signal sequences and peptide library screens
suggested a very loosely defined SecB binding “motif ” (12). This motif,
which is ⬃9 residues long, is enriched in aromatic and basic residues,
whereas acidic residues are disfavored. It theoretically occurs every
20 –30 residues in both secretory and cytoplasmic proteins and is too
unspecific to facilitate genome-wide prediction of SecB substrates (10 –
12). Thus experimentation is needed to identify novel SecB substrates.
To characterize the role of SecB in more detail and identify additional
SecB substrates, we analyzed a secB null mutant using comparative
proteomics. This analysis included flow cytometry, pulse labeling
combined with cell fractionation, one- and two-dimensional gel electrophoresis, and mass spectrometry (MS), complemented by immunoblotting. The comparative proteomics approach allowed us to investigate protein mistargeting, aggregation, and translocation kinetics, and
to determine changes in the proteome composition. Our analysis
showed that, although the secB null mutation did not affect cell growth,
there are significant differences at the proteome level. Most differences
pointed to protein targeting defects, resulting in a protein folding/aggregation problem in the cytoplasm. Careful analysis of the (sub)proteome(s) of the secB null mutant strain combined with a classical pulselabeling approach enabled us to more than triple the number of known
SecB-dependent secretory proteins.
EXPERIMENTAL PROCEDURES
Strains and Culture Conditions—We used E. coli strain EK413, which
is a MC4100 derivative that is ara⫹ (a kind gift from Ken-ichi Nishiyama), harboring plasmid pE63 as wild-type. Plasmid pE63 harbors the
gpsA gene, which encodes for sn-glycerol-3-phosphate dehydrogenase,
under control of an arabinose inducible promotor and has a pSC101
origin of replication and a ␤-lactamase resistance marker (14). Using P1
transduction, we moved the secB null mutation secB8 (15) from strain
HS101/pE63 into EK413/pE63, yielding an EK413/pE63-derived secB
2
The abbreviations used are: TF, trigger factor; MS, mass spectrometry; CHAPS, 3-[(3cholamidopropyl)dimethylammonio]-1-propanesulfonic acid; bis-Tris, 2-[bis(2-hydroxyethyl)amino]-2-(hydroxymethyl)propane-1,3-diol; Tricine, 2-{[2-hydroxy-1,1bis(hydroxymethyl)ethyl]amino}ethanesulfonic acid; Ibp, inclusion body associated
protein; MALDI-TOF, matrix-assisted laser desorption ionization time-of-flight; IPG,
immobilized pH gradient; SRP, signal recognition particle.
VOLUME 281 • NUMBER 15 • APRIL 14, 2006
Downloaded from www.jbc.org at Stockholm Universitetsbibliotek on August 6, 2007
To improve understanding and identify novel substrates of the
cytoplasmic chaperone SecB in Escherichia coli, we analyzed a secB
null mutant using comparative proteomics. The secB null mutation
did not affect cell growth but caused significant differences at the
proteome level. In the absence of SecB, dynamic protein aggregates
containing predominantly secretory proteins accumulated in the
cytoplasm. Unprocessed secretory proteins were detected in radiolabeled whole cell lysates. Furthermore, the assembly of a large fraction of the outer membrane proteome was slowed down, whereas its
steady state composition was hardly affected. In response to aggregation and delayed sorting of secretory proteins, cytoplasmic chaperones DnaK, GroEL/ES, ClpB, IbpA/B, and HslU were up-regulated severalfold, most likely to stabilize secretory proteins during
their delayed translocation and/or rescue aggregated secretory proteins. The SecB/A dependence of 12 secretory proteins affected by
the secB null mutation (DegP, FhuA, FkpA, OmpT, OmpX, OppA,
TolB, TolC, YbgF, YcgK, YgiW, and YncE) was confirmed by “classical” pulse-labeling experiments. Our study more than triples the
number of known SecB-dependent secretory proteins and shows
that the primary role of SecB is to facilitate the targeting of secretory
proteins to the Sec-translocase.
Defining the Role of E. coli SecB
APRIL 14, 2006 • VOLUME 281 • NUMBER 15
of cells. Lysis was achieved by incubation at 100 °C in a water bath for 2
min. The samples were then cooled on ice for 10 min before addition of
4 ␮l of DNase/RNase solution (1 mg/ml DNase I, 0.25 mg/ml RNase A,
476 mM Tris-HCl (pH 8), 24 mM Tris base, 50 mM MgCl2 and deionized
water) per 1 A600 unit of cells. The samples were incubated on ice for 10
min and were then immediately used for isoelectric focusing as
described below.
Preparation of Radiolabeled Membranes—Cells corresponding to
200 A600 units were cultured as described above. At the time of harvesting, an aliquot of 2 A600 units of cells was labeled with [35S]methionine
(60 ␮Ci/ml, Ci ⫽ 37 GBq) for 1 min. An excess of cold methionine (final
concentration 1 mg/ml) was added and cells were collected by centrifugation either directly after labeling or after a 10-min chase. The
remaining unlabeled cells were washed and collected by centrifugation.
Before breaking the cells, labeled and unlabeled cells from the same
culture were pooled back together resulting in a mixture of labeled and
unlabeled cells with a ratio of 1:100 that was then used for membrane
isolations. Carbonate-washed total membranes (i.e. a mixture of inner
and outer membranes) were isolated essentially as described by Molloy
et al. (23), with the exception that we used sonication rather than French
pressing to break cells. Protein concentrations were determined with
the BCA assay (Pierce) according to the instructions of the manufacturer.
Two-dimensional Gel Electrophoresis—The analysis of stained twodimensional electrophoresis gels of whole cell lysates was first done on
gels with a low protein load (0.5 A600 units of cells) to avoid saturation
and allow analysis of highly abundant proteins, and then on gels with a
high protein load (1 A600 unit of cells) for the analysis of low abundant
proteins. 1 A600 unit of cells was used for the analysis of [35S]methionine-labeled whole cell lysates. Whole cell lysates were solubilized in 9
M urea, 4% (w/v) CHAPS, 2 mM tributylphosphine, 0.5% (v/v) Triton
X-100, 5% glycerol, 2% (v/v) immobilized pH gradient gel (IPG) buffer
for pH 4 –7 (Amersham Biosciences) and bromphenol blue. For analysis
of the outer membrane proteome, 350 ␮g of protein was solubilized in 7
M urea, 2 M thiourea, 1% (w/v) ASB-14, 2 mM tributylphosphine, 5%
glycerol, 2% (v/v) IPG buffer for pH 4 –7 (Amersham Biosciences) and
bromphenol blue (23). Unsolubilized material was removed by centrifugation at 14,000 ⫻ g for 30 min. The clarified protein solution was used
to re-swell Immobilin DryStrips, pH 4 –7 (Amersham Biosciences),
overnight at room temperature. Isoelectric focusing was subsequently
performed at 20 °C in a Multiphor II apparatus (Amersham Biosciences); whole cell samples at 80 kVh and membrane samples at 60
kVh at a maximum 3,500 V. Proteins were separated in the second
dimension on 10% duracrylamide (Genomic Solutions) gels (10% acrylamide monomer and 1% bisacrylamide) containing 1 M Tris-HCl (pH
8.45), 0.1% (w/v) SDS, and 20% (v/v) glycerol. After focusing, proteins in
the IPG strips were reduced and alkylated, as described before (24). The
strips were loaded on top of the second dimension gel by submerging
the strips in warm agarose solution (1% (w/v) low melting agarose, 0.2%
SDS, 150 mM bis-Tris, 80 mM HCl and bromphenol blue). Electrophoresis was performed with Tricine-SDS buffer system (25) in a DALTON
tank (Amersham Biosciences) at 30 – 60 mA/gel for ⬃48 h, until the dye
front reached the bottom of the gel. Gels used for comparative analysis
were stained with high sensitivity silver stain (26) and gels containing
radiolabeled proteins were dried on filter paper. Preparative gels used
for identification of proteins by mass spectrometry were stained with
Coomassie Brilliant Blue R-250 or with mass spectrometry compatible
silver stain.
Several proteins were found in multiple spots at different pI values,
but with the same molecular weight. This was also observed in the outer
membrane maps of E. coli constructed by Molloy et al. (23). Most of
JOURNAL OF BIOLOGICAL CHEMISTRY
10025
Downloaded from www.jbc.org at Stockholm Universitetsbibliotek on August 6, 2007
null mutant strain. As expected, the secB null mutant is unable to form
single colonies on LB plates in the absence of arabinose; i.e. when GspA
is not expressed (14). Hereafter, we will refer to this EK413/pE63-derived secB null mutant strain and EK413/pE63 as the secB null mutant
and the control strains, respectively.
Cells were cultured in standard M9 medium supplemented with thiamine (10 mM), all amino acids but methionine and cysteine, glucose
(0.2% w/v), arabinose (0.2% w/v), and ampicillin (100 ␮g/ml). Overnight
cultures were diluted 1:50 in pre-warmed medium and cultured at
37 °C. Growth was monitored by measuring the A600 with a Shimadzu
UV-1601 spectrophotometer. Under these conditions, we did not
observe differences in growth (as monitored by A600 measurements)
between the secB null mutant and the control (results not shown). For all
the experiments, cells were harvested at an A600 of 1.0 (i.e. in the early
exponential phase).
Flow Cytometry—Analysis of the secB null mutant and the control by
means of flow cytometry was done using a FACSCalibur (BD Biosciences) instrument. Cultures of the secB null mutant and the control
were immediately diluted in ice-cold phosphate-buffered saline to a
final concentration of ⬃106 cells per ml, and analyzed with an average
flow rate of 400 events/s. Forward and side scatters were measured and
used for comparison of cell morphology of the secB null mutant and control
(16). Propidium iodide staining was performed to assess viability (16).
Immunoblot Analysis—The protein accumulation of SecB, SecY,
SecE, SecA, Ffh, PspA, TF, GroEL, DnaK, and IbpB (it should be noted
that the IbpB antiserum cross-reacts with IbpA) in the secB null mutant
strain and the control strain were determined by immunoblot analysis.
Cells were cultured as described above. Cells (0.2 A600 units) or inner
membranes (5 ␮g of protein) isolated by sucrose gradient centrifugation
(17, 18) were solubilized in Laemmli solubilization buffer. Proteins were
separated by SDS-PAGE. Blotting, immunodecoration, detection, and
quantification of blots were done as described previously (19).
Protein Translocation Assays in Vivo—Protein translocation assays
were done with 1 ml of culture each. Cells were labeled with [35S]methionine (60 ␮Ci/ml, Ci ⫽ 37 GBq) for 45 s and subsequently precipitated in 10% trichloroacetic acid. Trichloroacetic acid-precipitated
samples were washed with acetone, resuspended in 10 mM Tris-HCl
(pH 7.5), 2% SDS, and immunoprecipitated with antisera to OmpA and
␤-lactamase, followed by standard SDS-PAGE analysis (21). Gels were
scanned in a Fuji FLA-3000 phosphorimager and quantified using the
Image Gauge software (version 3.4). Potential SecB-dependent secretory proteins were C-terminal hemagglutinin-tagged and expressed by
isopropyl 1-thio-␤-D-galactopyranoside induction from the pEH1 vector as described previously (20), and the protein translocation assay was
performed as described above except that antiserum to the hemagglutinin tag was used for immunoprecipitations.
Preparation of Whole Cell Lysates for Two-dimensional Gel Electrophoresis—Cells were cultured as described above. Radiolabeled cells
were cultured in the presence of [35S]methionine (60 ␮Ci/ml, Ci ⫽ 37
GBq) for 1 min followed by the addition of an excess of cold methionine
(final concentration 1 mg/ml). Cells were collected by centrifugation at
10,000 ⫻ g for 10 min at 4 °C, and were subsequently washed twice in
ice-cold M9 minimal medium. For labeled cells, cold methionine was
included in the wash steps. Cell pellets were snap-frozen and stored at
⫺80 °C. To prepare samples for isoelectric focusing, cells were lysed
essentially as described by VanBogelen and Neidhardt (22). Frozen cell
pellets were thawed on ice and then quickly resuspended in 40 ␮l of
solubilization solution (0.3% (w/v) SDS, 200 mM dithiothreitol, 28 mM
Tris-HCl (pH 8), 22 mM Tris base and deionized water) per 1 A600 unit
Defining the Role of E. coli SecB
these “trains of spots” are because of modifications induced during sample preparation (27), likely because of stepwise deamidation of residues
Asn and Gln, resulting in loss of 1 dalton and net loss of one positive
charge.3
Image Analysis and Statistics—Stained gels were scanned using a
GS-800 densitometer from Bio-Rad. Radiolabeled gels were scanned in
a Fuji FLA-3000 phosphorimager. Spots were detected, quantified,
matched, and compared using the two-dimensional analysis software
PDQuest (Bio-Rad). The analyses of silver-stained and radiolabeled
outer membrane proteins were done on the same set of gels. In all cases,
each analysis set consists of at least three gels in each replicate group (i.e.
secB null mutant and the control). All gels in a set represented independent samples (i.e. samples from different bacterial colonies, cultures, and
membrane preparations), which were subjected to two-dimensional
electrophoresis and image analysis in parallel, i.e. en group. Spot quantities were normalized using the “total density in gel image” method to
compensate for non-expression related variations in spot quantities
between gels. The PDQuest software was set to detect differences that
were found to be statistically significant using the Student’s t test and a
99 (whole cell lysates) or 95% (outer membrane) level of confidence,
including qualitative differences (“on-off responses”) present in all gels
in a group. Saturated spots were excluded from the analysis.
Protein Identification by Mass Spectrometry and Bioinformatics—
Stained protein spots or bands were excised, washed, digested with
modified trypsin and peptides extracted manually or automatically
(ProPic and Progest, Genomic Solutions, Ann Arbor, MI), and peptides
were applied to the MALDI target plates as described previously (28).
The mass spectra were obtained automatically by MALDI-TOF MS in
reflectron mode (Voyager-DE-STR; PerSeptive Biosystems, Framing3
V. Zabrouskov et al., unpublished results.
10026 JOURNAL OF BIOLOGICAL CHEMISTRY
ham, MA), followed by automatic internal calibration using tryptic peptides from autodigestion. The latest version of the NBCI non-redundant
data base (downloaded locally) were searched automatically with the
resulting peptide mass lists, using the search engine ProFound (29), as
part of Knexus (30). Criteria for positive identification by MALDI-TOF
MS peptide mass fingerprinting were at least four matching peptides
with an error distribution within ⫾25 ppm and at least 15% sequence
coverage. During the search, we only allowed one missed cleavage and
partially oxidized methionines. In the more complex samples, the peptides were also analyzed by nano-LC-ESI-MS/MS in automated mode
on a quadruple/orthogonal acceleration TOF tandem mass spectrometer (Q-TOF; Micromass, Manchester, UK) (see Ref. 31 for details). The
spectra were used to search the SwissProt 42.10 data base with the Mascot
search engine. All significant MS/MS identifications by Mascot were manually verified for spectral quality and matching y and b ion series.
Isolation of Protein Aggregates—Protein aggregates were isolated
essentially as described (32). 100 ml of culture with an A600 of 1.0 was
used for each aggregate isolation. The protein content of total cells and
aggregates was determined with the BCA assay according to the instructions of the manufacturer (Pierce). Aggregates were analyzed by SDSPAGE using 24-cm long 8 –16% acrylamide gradient gels. Proteins were
stained with Coomassie Brilliant Blue R-250 and identified by mass
spectrometry as described before.
For radiolabeling of aggregates, 100 A600 units of cells were labeled
with [35S]methionine (2500 ␮Ci/ml, Ci ⫽ 25 GBq) for 30 s and chased
for 1, 3, and 15 min by addition of an excess of cold methionine (final
concentration 1 mg/ml). Aggregates were isolated as described above,
solubilized in 10 mM Tris-HCl (pH 7.5), 2% SDS, and subsequently
processed using an OmpA antiserum as described under “Protein
Translocation Assays” (see above).
VOLUME 281 • NUMBER 15 • APRIL 14, 2006
Downloaded from www.jbc.org at Stockholm Universitetsbibliotek on August 6, 2007
FIGURE 1. Analysis of the E. coli secB null mutant by flow cytometry and immunoblotting. A, flow cytometric properties of secB null mutant (SecB⫺) and control (SecB⫹) cells. In
the first 2 panels, the size of the population (forward scatter FSC) is plotted versus granularity (side scatter, SSC) for both SecB⫹ and SecB⫺. To facilitate comparison of those 2
parameters for SecB⫹ and SecB⫺, histograms for size and granularity are shown in the third and fourth panels. One representative experiment of four is shown. Cells were cultured
and flow cytometry was performed as described under “Experimental Procedures.” B, Western blot analysis of SecB and components of the Sec-translocase (SecA, -Y, -E), Ffh, a
constituent of the SRP-targeting pathway, and PspA, of which the expression is up-regulated when the electrochemical potential is affected. Left panel, cells (0.2 A600 units) were
separated by means of SDS-PAGE and subsequently subjected to immunoblot analysis with antibodies to SecB, SecA, Ffh, and PspA. Right panel, inner membranes (5 ␮g of protein)
were separated by means of SDS-PAGE, and subsequently subjected to immunoblot analysis with antibodies to SecA, SecY, and SecE.
Defining the Role of E. coli SecB
Downloaded from www.jbc.org at Stockholm Universitetsbibliotek on August 6, 2007
FIGURE 2. Analysis of whole cell lysates of the secB null mutant by two-dimensional electrophoresis. A, comparative two-dimensional electrophoresis gel analysis of highly
abundant proteins in whole cell lysates of the secB null mutant (SecB⫺) and its control (SecB⫹). Cells were harvested when the cultures had reached an A600 of 1.0. 0.5 A600 units of
cells were solubilized and proteins were separated by two-dimensional electrophoresis. Proteins were visualized by silver stain and differences between secB null mutant and control
gels were analyzed using the PDQuest software (Bio-Rad). At least four independent samples from each strain were used for the analysis. Differential protein expression between the
secB null mutant and the control was analyzed using the Student’s t test and a 99% level of confidence (see “Experimental Procedures”). Proteins were identified by mass spectrometry
from spots excised from gels stained with Coomassie or mass spectrometry compatible silver stain (Table 1 and supplemental Table 1). Annotated spots have been matched onto the
silver-stained gels shown here using the PDQuest software (Bio-Rad). The levels of the highly abundant chaperones DnaK and GroEL are ⬃50% higher in the secB null mutant
(supplemental Table 1). In contrast, the level of TF is unaffected in the secB null mutant. B, quantitative immunoblotting of DnaK, GroEL, and TF in secB null mutant (SecB⫺) and control
(SecB⫹) cells. SDS-PAGE and blotting were performed as described under “Experimental Procedures.” The levels of the chaperones DnaK and GroEL are increased in the secB null
mutant, whereas the level of TF is unaffected. C, comparative two-dimensional electrophoresis gel analysis of low abundant proteins in whole cell lysates of the secB null mutant
(SecB⫺) and the control (SecB⫹). Proteins from 1 A600 unit of cells were visualized by silver staining and analyzed as described above. 46 additional spots are significantly changed
using the criteria described above (supplemental Table 1). Spots that are down-regulated in the secB null mutant are indicated in the “SecB⫹” gel and spots that are up-regulated are
indicated in the “SecB⫺” gel. Spots where proteins have been successfully identified are labeled with both the spot number and gene name (Table 1 and supplemental Table 1). The
level of GroES, co-chaperone, and regulator of GroEL, is 50% increased in the secB null mutant. The level of the chaperone ClpB is doubled and the level of the chaperone/protease
HslU is tripled. D, zooms of two-dimensional electrophoresis gels with radiolabeled whole lysates from the secB null mutant (SecB⫺) and the control (SecB⫹) visualized by autoradiography. Processed (m ⫽ mature) and precursor (p) forms of the secretory proteins OmpT, OmpA, and OppA are indicated.
APRIL 14, 2006 • VOLUME 281 • NUMBER 15
JOURNAL OF BIOLOGICAL CHEMISTRY
10027
Defining the Role of E. coli SecB
TABLE 1
Proteins identified in differentially regulated spots in two-dimensional electrophoresis gels of whole cell lysates of the secB null mutant and
the control
Whole cells were analyzed by two-dimensional electrophoresis (Fig. 2, A and C). Differentially regulated spots were excised from silver- or Coomassie-stained gels. Proteins
were identified by MALDI-TOF MS and/or by nano-LC-ESI-MS/MS as described under “Experimental Procedures” (supplemental Table I). Spots 1–2 are from whole cell
maps loaded with 0.5 A600 units of cells (Fig. 2A), and the remaining spots are from gels loaded with 1 A600 unit of cells (Fig. 2C).
Spot no.a
Gene name(s)b
Protein name
Localizationc
1
2
3
5
10
11
14
15
17
18
22
26
27
27
31
33
35
38
44
46
47
dnaK, grpF, groP, seg
groEL, groL, mopA
crr, csr, iex, tgs, treD
ompT
fhuA, tonA
luxS
deoB, drm, thyR
ribH, ribE
oppA
oppA
oppA
oppA
kdgA, eda, hga
ssb, exrB, lexC
groES, mopB
rplL
clpB
hslU, htpI
yjfR
yjfR
yjfR
Chaperone protein DnaK
60-kDa Chaperonin GroEL
PTS system, glucose-specific IIA component
Protease VII
Ferrichrome-iron receptor
S-Ribosylhomocysteinase
Phosphopentomutase
Riboflavin synthase ␤ chain
Periplasmic oligopeptide-binding protein
Periplasmic oligopeptide-binding protein
Periplasmic oligopeptide-binding protein
Periplasmic oligopeptide-binding protein
KHG/KDPG aldolase
Single-strand binding protein, helix-destabilizing protein
10-kDa Chaperonin, GroES protein
50 S ribosomal protein L7/L12, L8
Chaperone ClpB
ATP-dependent hsl protease ATP-binding subunit HslU
Hypothetical protein YjfR
Hypothetical protein YjfR
Hypothetical protein YjfR
Cytoplasmic
Cytoplasmic
Cytoplasmic
Outer membrane
Outer membrane
Cytoplasmic
Cytoplasmic
Cytoplasmic
Periplasmic
Periplasmic
Periplasmic
Periplasmic
Cytoplasmic
Cytoplasmic
Cytoplasmic
Cytoplasmic
Cytoplasmic
Cytoplasmic
PSORT:cytoplasmic
PSORT:cytoplasmic
PSORT: cytoplasmic
Change (secB null
mutant/control)d
-Fold
a
The numbering corresponds to the spots in two-dimensional electrophoresis gel images shown in Fig. 2, A and C.
Names in bold are used to label the corresponding spots in the gel shown in Fig. 2, A and C.
Localization according to the SwissProt database. The localization of unknown proteins was predicted using PSORT.
d
The PDQuest software was used to detect and calculate the -fold change, i.e. the ratio of the average intensity of spots in secB null mutant gels to the average intensity of matched
spots in the control gels (supplemental Table I).
b
c
RESULTS
Characterization of the secB Null Mutant Strain—Using P1 transduction we moved the secB null mutation secB8 (15) from strain HS101/
pE63 (14) into strain EK413/pE63 (a MC4100 derivative that is ara⫹; a
kind gift from Ken-ichi Nishiyama), yielding an EK413/pE63-derived
secB null mutant strain. Hereafter, we will refer to the secB null mutant
strain and EK413/pE63 as the secB null mutant and control, respectively.
The secB null mutant and control were cultured aerobically in M9 minimal medium. Under these conditions, we did not observe any differences in growth, as monitored by A600 measurements. In addition, propidium iodide staining (16) did not point to differences in viability
between the mutant and the control (results not shown). Early logphase cells were used in all the experiments described in this study. The
morphology of cells was analyzed by means of flow cytometry (16, 33).
Interestingly, we detected a small increase of both the forward scatter
and side scatter of secB null mutant cells (Fig. 1A). This indicates that
secB null mutant cells are slightly bigger than control cells and most
likely contain extra internal structures (i.e. extra membranes and/or
protein aggregates).
To verify the phenotype of the secB null mutant strain, we monitored
the targeting of the established SecB-dependent outer membrane protein OmpA (14) and SecB-independent periplasmic protein ␤-lactamase (34), using pulse-chase radiolabeling experiments in combination
with immunoprecipitations. As expected, the translocation of OmpA
was hampered in the secB null mutant, as evidenced by accumulation of
precursor protein, whereas the translocation of ␤-lactamase was not
affected (results not shown). The levels of SecA, -Y, and -E were determined by Western blotting for the secB null mutant and control,
because SecB delivers proteins to the SecYE protein-conducting channel through interaction with SecA (1). Protein levels of the SecAYEtranslocase components did not change in the absence of SecB (Fig. 1B).
Ffh is a core component of the SRP targeting pathway, which mainly
10028 JOURNAL OF BIOLOGICAL CHEMISTRY
targets inner membrane proteins to the Sec-translocase but may have
some overlap with the SecB targeting pathway (20, 35–37). It is not
known if the SRP targeting pathway can compensate for the absence of
SecB. Immunoblot analysis of secB null mutant and control showed that
Ffh levels were unchanged in the absence of SecB. It has been shown that
expression of PspA is up-regulated when the electrochemical potential
is affected (38). Because the electrochemical potential plays an important role in protein translocation we analyzed the levels of the PspA
protein by immunoblot analysis. In contrast to several other Sec
mutants (38), there is no PspA response in the secB null mutant (Fig. 1B).
Analysis of Whole Cell Lysates of the secB Null Mutant by Two-dimensional Electrophoresis—To identify potential SecB substrates and
compensatory mechanisms and/or stress responses in the secB null
mutant, we used a proteomics approach. Whole cell lysates of the
mutant and the control were analyzed by two-dimensional electrophoresis, using IPG strips with a pI range from 4 to 7. To allow for
quantitative analysis of highly abundant proteins, gels were loaded with
limited amounts of protein, such that staining of highly abundant proteins was not saturated. The comparative analysis was based on 4 gels
per strain (an independent culture was used for each gel). Gels were
stained with silver, scanned, and images were analyzed and compared
using the PDQuest software (Bio-Rad). Significance was determined
using Student’s t test (for details see “Experimental Procedures”). This
analysis demonstrated that the levels of both DnaK and GroEL were
increased by about 50% (p ⬍ 0.01) in the secB null mutant (Fig. 2A, Table
1, and supplemental Table 1). Higher levels of DnaK and GroEL are
consistent with increased synthesis rates of these proteins in a SecB
knock-out strain (39). The level of TF, the first cytoplasmic chaperone
that interacts with ribosome-associated nascent peptides, was not
affected. These results were confirmed by immunoblot analysis
(Fig. 2B).
VOLUME 281 • NUMBER 15 • APRIL 14, 2006
Downloaded from www.jbc.org at Stockholm Universitetsbibliotek on August 6, 2007
1.5
1.6
0.3
0.3
0.5
0.5
0.5
0.5
0.5
0.6
0.7
0.7
0.7
0.7
1.5
1.7
2.3
3.8
⬎100
⬎100
⬎100
Defining the Role of E. coli SecB
APRIL 14, 2006 • VOLUME 281 • NUMBER 15
FIGURE 3. Characterization of aggregates isolated from E. coli secB null mutant. A,
aggregates from the secB null mutant (SecB⫺) and the control (SecB⫹) were isolated
from 100-ml cultures in M9 minimal media. The protein content of the aggregates was
analyzed by SDS-PAGE on a 24-cm long 8 –16% gradient gel stained by Coomassie Brilliant Blue R-250 and proteins were identified by mass spectrometry as described under
“Experimental Procedures” (Table 2 and supplemental Table 2). Proteins that were identified with a signal sequence are marked with an asterisk (*). Known or predicted localizations are indicated by: om, outer membrane; om lp, outer membrane lipoprotein; cyt,
cytoplasmic; per, periplasmic; or sec, secretory (predicted by PSORT to be either periplasmic or outer membrane protein). B, quantitative immunoblotting of IbpA/B in secB null
mutant (SecB⫺) and control (SecB⫹) whole cell lysates. The expression levels of IbpA/B
are considerably higher in the secB null mutant. C, OmpA was isolated from aggregates
prepared from secB null mutant (SecB⫺) control (SecB⫹) cells. Cells were labeled with
[35S]methionine and chased with cold methionine for different times as indicated in the
figure. Aggregates were prepared and OmpA was subsequently isolated by means of
immunoprecipitation as described under “Experimental Procedures.” To be able to distinguish between the precursor (pro-OmpA) and processed forms of OmpA, the immunoprecipitation was also performed on [35S]methionine-labeled whole cells from E. coli
strain CM124 (PAra-secE) cultured in the presence (SecE⫹) and absence (SecE⫺) of arabinose. In SecE⫹ cells there is a fully functional Sec-translocase and OmpA is translocated
across the cytoplasmic membrane and subsequently processed, whereas in SecE⫺ cells
there is not a functional Sec-translocase and the precursor form of OmpA accumulates in
the cytoplasm (54).
Furthermore, MS/MS data revealed that at least two of the secretory
proteins, OmpA and the murein lipoprotein (Lpp or MulI), identified in
the aggregates contained an uncleaved signal sequence (results not
shown), again pointing to aggregation of proteins in the cytoplasm
rather than in the periplasm.
To study the localization and dynamics of the aggregates in more
detail, we isolated OmpA by immunoprecipitation from aggregates iso-
JOURNAL OF BIOLOGICAL CHEMISTRY
10029
Downloaded from www.jbc.org at Stockholm Universitetsbibliotek on August 6, 2007
To detect differences in accumulation of low abundant proteins, we
repeated the two-dimensional electrophoresis proteome analysis with
higher protein loading (Fig. 2C). In total 48 spots were found to be
significantly (with p ⬍ 0.01) altered. Although most of these spots were
weakly stained, we were able to identify 16 non-redundant proteins by
MS (Table 1; image analysis and MS data are summarized in supplementary Table 1). Some proteins were identified in multiple spots
located in “trains” next to each other and in total 20 spots could be
annotated. In addition to DnaK and GroEL, the levels of the chaperones
GroES, ClpB, and HslU were increased by ⬃50, ⬃200, and ⬃400%,
respectively. GroES is the regulator and co-chaperone of GroEL and the
GroEL/ES chaperone system is essential for proper folding and maturation of proteins in the cytoplasm (40). The chaperone ClpB has been
shown to act in concert with DnaK and inclusion body proteins (Ibp) to
extract and refold proteins from aggregates (41). The protein HslU can
function either on its own as a chaperone (42) or in a complex together
with HslV (ClpY) as a protease (43). However, we did not find any spots
that were differentially regulated in the region of the gel where HslV
should migrate. Notably, the genes encoding the DnaK, GroEL/ES,
ClpB, and HslU proteins are all regulated by transcription factor ␴32 (44,
45). The ␴32-induced response, better known as the “heat shock
response,” is activated in response to protein misfolding/aggregation in
the cytoplasm (44, 45).
The levels of the processed forms of the outer membrane proteins
ferrichrome-iron receptor (FhuA), the protease OmpT, and the
periplasmic oligopeptide-binding protein (OppA) were decreased by
⬃50, ⬃70, and ⬃30%, respectively, in the secB null mutant strain. Interestingly, the homolog of OppA in Salmonella typhimurium has been
shown to bind tightly to E. coli SecB in vitro (46). No precursors of
secretory proteins were identified in the silver-stained gels of whole cell
lysates.
To study kinetic effects of the secB null mutation, we repeated the
comparative proteome analysis with cells labeled with [35S]methionine.
In these radioactive gels, 37 radiolabeled spots were significantly
changed (p ⬍ 0.01) in the secB null mutant. Interestingly, several of
these spots were not present in the silver-stained gels. Based on pI and
molecular weight they match the precursors of secretory proteins, such
as OmpA, OmpT, and OppA (Fig. 2D).
Isolation and Characterization of Protein Aggregates from the secB
Null Mutant—The heat shock response in the secB null mutant pointed
to a problem of protein folding and aggregation in the cytoplasm of cells
lacking SecB. In addition, the flow cytometry experiments suggested the
presence of extra internal structures, i.e. extra internal membranes
and/or protein aggregates, in the secB null mutant. Indeed, protein
aggregates containing around 0.5% of total cellular protein could be
isolated from the secB null mutant, but were virtually absent in the
control strain (Fig. 3A). The aggregates were dissolved in Laemmli solubilization buffer, and proteins were separated by SDS-PAGE, followed
by identification using nano-LC electrospray tandem mass spectrometry (nano-LC-ESI-MS/MS). Fourteen secretory proteins and five cytoplasmic proteins were identified (Fig. 3A, Table 2, and supplemental
Table 2). The inclusion body protein IbpA, also part of the heat shock
regulon, was among the cytoplasmic proteins identified in the aggregates (Table 2) (41, 47). Immunoblotting of total cell extracts showed
that the levels of the chaperones IbpA/B are indeed strongly up-regulated in the secB null mutant (Fig. 3B). In contrast, immunoblotting
showed that the levels of the periplasmic chaperone Skp and protease
DegP are not changed in the secB null mutant (data not shown). This
indicates that no significant protein misfolding/aggregation occurred in
the periplasm/outer membrane (38, 48).
Defining the Role of E. coli SecB
TABLE 2
Identification of proteins in aggregates isolated from the secB null mutant and the control
Protein bands were excised from Coomassie-stained one-dimensional gels loaded with aggregates isolated from the secB null mutant and the control (Fig. 3A). Proteins were
identified by MALDI-TOF MS and/or nano-LC-ESI-MS/MS as described under “Experimental Procedures.”
b
c
Gene name(s)b
1
2
3
4
4
5
glgB
tolC, mtcB, mukA, refI
glgA
degP, htrA
tolB
yncE
1,4-␣-Glucan branching enzyme
Outer membrane protein TolC
Glycogene synthase
Heat shock protein HtrA
TolB protein
Hypothetical protein YncE
6
6
ompA, tolG, tut, con
yncE
Outer membrane protein A
Hypothetical protein YncE
7
ydgH
Hypothetical protein YdgH
8
9
fkpA
ybgF
FKBP-type peptidyl-prolyl cis-trans isomerase FkpA
Hypothetical protein YbgF
10
ompA, tolG, tut, con
11
lpp, mulI, mlpA
12
13
14
15
15
infC
ompX
dps
ibpA, hslT, htpN
ygiW
Outer membrane protein A, Outer membrane
protein II*
Major outer membrane lipoprotein precursor
(murein-lipoprotein)
Initiation factor 3
Outer membrane protein X
Starvation-inducible DNA-binding protein
16-kDa heat shock protein A
Hypothetical Protein YgiW
16
lpp, mulI, mlpA
16
ycgK
Localizationc
Protein name(s)
Major outer membrane lipoprotein precursor,
murein-lipoprotein
Hypothetical protein YcgK
Cytoplasmic
Outer membrane
Cytoplasmic
Periplasmic
Periplasmic
Unknown, PSORT predicts periplasmic or outer
membrane
Outer membrane
Unknown, PSORT predicts periplasmic or outer
membrane
Unknown, PSORT predicts periplasmic or outer
membrane
Periplasmic
Unknown, PSORT predicts periplasmic or outer
membrane
Outer membrane
Outer membrane lipoprotein
Cytoplasmic
Outer membrane
Cytoplasmic
Cytoplasmic
Unknown, PSORT predicts periplasmic or outer
membrane
Outer membrane lipoprotein
Unknown, PSORT predicts periplasmic or outer
membrane
The numbering corresponds to the bands in the one-dimensional electrophoresis gel shown in Fig. 3A.
The gene names and synonyms. Names in bold are used to label the corresponding bands in the gel shown in Fig. 3A.
Localization according to the SwissProt database. The localization of unknown proteins was predicted using PSORT.
lated from [35S]methionine-labeled cells (Fig. 3C). Only the precursor
form of OmpA could be isolated from secB null mutant aggregates,
which is another indication for their cytoplasmic localization. Interestingly, the OmpA signal disappeared during the chase with non-radioactive methionine, indicating that after extraction from the aggregates,
pro-OmpA is either degraded or remobilized for translocation.
Analysis of the Outer Membrane Proteome of the secB Null Mutant—
Our analysis of labeled whole cell lysates suggested that the secB null
mutation affects the targeting of secretory proteins by slowing down the
delivery of the precursor to the Sec-translocase. Furthermore, secretory
proteins were identified in aggregates isolated from the secB null
mutant. Therefore, we decided to compare the secretomes (periplasm
and outer membrane) of the secB null mutant and the control using
comparative two-dimensional electrophoresis analysis of radiolabeled
and unlabeled cells. Attempts to analyze the periplasmic proteome
failed, because the isolated periplasmic fractions were insufficiently
pure for conclusive comparative proteome analysis. Importantly, we
were successful at quantitatively comparing the steady state and assembly kinetics of the outer membrane proteomes of the secB null mutant
and the control.
The outer membrane proteins are ␤-barrel proteins and they can be
well resolved on two-dimensional electrophoresis gels when using a
mixture of the non-ionic detergent ASB-14 and thiourea (23). The inner
membrane contains ␣-helical proteins, which typically precipitate during isoelectric focusing and are thus not resolved on the two-dimensional electrophoresis gels. Therefore, two-dimensional electrophoresis
gels from the total membrane fraction visualize predominantly outer
membrane proteins and peripheral membrane proteins (23). Membranes isolated from [35S]methionine-labeled cells were used for the
analysis. The gels were first stained with silver and then used for autoradiography. This allowed us to study the steady state composition and
10030 JOURNAL OF BIOLOGICAL CHEMISTRY
the insertion kinetics of the outer membrane proteome in the same set
of gels. To facilitate identification of proteins by MS, we generated preparative gels stained with Coomassie Brilliant Blue in parallel to the
radioactive silver-stained gels. Twenty-five proteins were identified
(Fig. 4A, Table 3, and supplemental Table 3). Fourteen of those are
established integral (␤-barrel) outer membrane proteins, four are
peripheral outer membrane lipoproteins, and one is a potential integral
outer membrane protein. The remaining six proteins are peripheral
membrane proteins interacting with the inner or outer membrane and
integral inner membrane proteins with one predicted transmembrane
segment.
The comparative analysis of the silver-stained gels showed that levels
of iron-siderophore transporter FhuA, the ferrichrome-iron receptor
FhuE, and peptidoglycan-associated lipoprotein (Pal) were decreased
2-fold in the absence of SecB. In contrast, the level of the ferri-enterobactin receptor FepA was doubled, possibly compensating for the
decrease of the FhuA and FhuE levels. These significant changes in
levels of key players in the strictly regulated iron metabolism of the
E. coli cell had apparently no effect on growth of the secB null mutant
under the experimental conditions used. However, it is very well conceivable that under other conditions similar changes could have significant effects. Steady state levels of all the other identified outer membrane proteins were unchanged.
Importantly, the analysis of the two-dimensional electrophoresis
autoradiograms of [35S]methionine-labeled proteins revealed that many
outer membrane proteins, such as BtuB, FhuA, FhuE, FadL, OmpT,
OmpX, and TolC appear considerably slower in the outer membrane of
the secB null mutant than in the outer membranes of the control. This
shows that SecB helps to improve efficiency of secretion, rather than
being strictly essential. In Fig. 4B, this is shown in more detail for two
examples, FhuA and TolC. In the secB null mutant, both FhuA and TolC
VOLUME 281 • NUMBER 15 • APRIL 14, 2006
Downloaded from www.jbc.org at Stockholm Universitetsbibliotek on August 6, 2007
a
Band no.a
Defining the Role of E. coli SecB
can only be detected after a 10-min chase rather than directly after
labeling as in the control strain.
Identification of Novel SecB-dependent Secretory Proteins—SecB
dependence of 12 potential SecB substrates identified in the protein
aggregates or in the two-dimensional electrophoresis gels was directly
monitored using a pulse-labeling approach (Fig. 5 and supplemental
Table 4).
Cells with and without SecB were labeled with [35S]methionine and
the precursor/processed forms of the secretory proteins tested (DegP,
FhuA, FkpA, OmpT, OmpX, OppA, TolB, TolC, YbgF, YcgK, YgiW and
YncE) were subsequently immunoprecipitated and analyzed by SDSPAGE and autoradiography. Strikingly, translocation of all these secretory proteins was hampered in the secB null mutant, as shown by the
accumulation of precursors compared with control. This shows that all
these proteins indeed need SecB for efficient translocation across the
cytoplasmic membrane. In addition, similar pulse-labeling experiments
in the presence of the SecA inhibitor azide showed that translocation of
these 12 proteins is also SecA dependent (results not shown).
DISCUSSION
To characterize the role of the chaperone SecB in E. coli in more
detail, we studied a secB null mutant strain using a comparative proteomics approach complemented with Western blotting and pulsechase experiments. The secB null mutation did not significantly affect
APRIL 14, 2006 • VOLUME 281 • NUMBER 15
growth, but flow cytometry experiments showed that the secB null
mutation did affect cell morphology; cells are slightly bigger and seem to
contain internal structures.
The comparative proteome analysis of the secB null mutant resulted
in three main observations and conclusions: 1) absence of SecB results in
aggregation of secretory proteins and increased levels of cytoplasmic chaperones; 2) SecB is not required for targeting and translocation of secretory
proteins per se, but rather is needed to improve efficiency of the cytoplasmic
targeting process and delivery to the Sec-translocase; and 3) SecB dependence was established for an additional 12 secretory proteins. Below we
explain and discuss these main conclusions and observations in more detail.
Absence of SecB Affects Protein Homeostasis in the Cytoplasm—The
two-dimensional electrophoresis analysis of radiolabeled whole cell
lysates showed that precursors of secretory proteins accumulate in the
secB null mutant. Furthermore, the secB null mutation induced the ␴32regulated heat shock response, which is diagnostic for protein aggregation/misfolding in the cytoplasm.
Indeed, cytoplasmic protein aggregates, which contained mainly
secretory proteins, were for the first time isolated from secB null mutant
cells. The amount of aggregated proteins in secB null mutant cells
(around 0.5% of the total protein) is about half as much as in single
mutant cells that lack cytoplasmic chaperones like DnaK and TF (32, 49,
50). This suggests that DnaK and TF play a different role than SecB in
protein homeostasis.
JOURNAL OF BIOLOGICAL CHEMISTRY
10031
Downloaded from www.jbc.org at Stockholm Universitetsbibliotek on August 6, 2007
FIGURE 4. Analysis of the outer membrane proteome from the E. coli secB null mutant. A, outer membrane proteomes of the E. coli secB null mutant (SecB⫺) and the control strain
(SecB⫹) were visualized by two-dimensional electrophoresis. 350 ␮g of total membrane proteins was used for each gel. Proteins were visualized by silver stain and differences
between secB null mutant and control gels were analyzed using the PDQuest software. At least three independent samples from each strain were used for the analysis. Differences
in protein expression between the secB null mutant and the control were evaluated using the Student’s t test and a 99% level of confidence (see “Experimental Procedures”). Saturated
spots were excluded from the analysis. One representative gel from each strain is shown. Proteins were identified by mass spectrometry from spots excised from Coomassie-stained
gels (Table 3 and supplemental Table 3). Identified proteins are indicated by gene name. Annotated spots were matched from the Coomassie-stained gels to the silver-stained gels
shown in the figure using the PDQuest software. B, zooms of autoradiographs of two-dimensional electrophoresis gels with radiolabeled outer membranes of the secB null mutant
(SecB⫺) and control (SecB⫹). Membranes were isolated from cells harvested directly after labeling or after a 10-min chase as described under “Experimental Procedures.” Note that
the FepA signal is, just like in the silver-stained gels, higher in the secB null mutant than in the control.
VOLUME 281 • NUMBER 15 • APRIL 14, 2006
c
b
a
D-Methionine-binding lipoprotein metQ
Lipoprotein-28
Lipoprotein-34
Outer membrane protein A
Outer membrane protein C
OmpT, Omptin, Protease A
Outer membrane protein X
Organic solvent tolerance protein
Peptidoglycan-associated lipoprotein
Peptidyl-prolyl cis-trans isomerase D
TolC
Nucleoside-specific channel-forming protein tsx
Outer membrane protein assembly factor YaeT
VacJ lipoprotein
Hypothetical protein in bioA 5⬘ region, Putative
lipoprotein YbhC
Probable tonB-dependent receptor YbiL
Hypothetical protein YfgM
metQ
nlpA
nlpB, dapX
ompA, tolG, tut, con
opmC, meoA, par
ompT
ompX
ostA, imp
pal, excC
ppiD
tolC, mtcB, mukA, refI
tsx, nupA
yaeT
vacJ
ybhC
No significant change
No significant change
No significant change
No significant change
No significant change
No significant change
No significant change
No significant change
No significant change
No significant change
0.49
No significant change
No significant change
No significant change
No significant change
No significant change
No significant change
No significant change
No significant change
No significant change
No significant change
2.16
0.46
0.52
No significant change
Ratio silver stain (secB null
mutant/control)c
No significant change
No significant change
Not detected
No significant change
No significant change
No significant change
No significant change
0.34
0.28
Not detected
No significant change
Not detected
0.30
No significant change
No significant change
No significant change
Not detected
No significant change
0.23
No significant change
0.27
1.28
0.15
0.16
No significant change
Ratio 关35S兴Met (secB
null mutant/control)c
The gene names and synonyms. Names in bold are used to label the corresponding spots in the two-dimensional electrophoresis gel images in Fig. 4, A and B.
Localization according to the SwissProt database. The localization of unknown proteins was predicted using PSORT.
-Fold change of significantly (p ⬍ 0.01 in silver-stained gels and p ⬍ 0.05 in 关35S兴methionine-labeled gels) changed spots calculated as the ratio of the average intensity of spots in secB null mutant gels to the average intensity of matched
spots in the control gels. Note that not all identified proteins were detected in 关35S兴methionine-labeled gels.
Potentially outer membrane
PSORT predicts inner membrane
Inner membrane lipoprotein
Outer membrane
Outer membrane
Outer membrane
Outer membrane
Outer membrane
Outer membrane
Cytoplasm (attaches to the inner membrane during
cell division)
Probably attached to membrane by lipid anchor
Inner membrane lipoprotein
Outer membrane lipoprotein
Outer membrane
Outer membrane
Outer membrane
Outer membrane
Outer membrane
Outer membrane lipoprotein
Inner membrane
Outer membrane
Outer membrane
Outer membrane
Outer membrane lipoprotein
Probably attached to the OM with lipid anchor
Localizationb
Downloaded from www.jbc.org at Stockholm Universitetsbibliotek on August 6, 2007
10032 JOURNAL OF BIOLOGICAL CHEMISTRY
ybiL
yfgM
Acriflavine resistance protein A
Vitamin B12 receptor
Colicin I receptor
Long-chain fatty acid transport protein
Ferrienterobactin receptor
Ferrichrome-iron receptor
FhuE receptor
Cell division protein FtsZ
Protein name
acrA, mtcA,lir
btuB, bfe, cer, dcrC
cirA, cir, feuA
fadL, ttr
fepA, fep, feuB
fhuA, tonA
fhuE
ftsZ, sifB, sulB
Gene name(s)a
关35S兴Methionine-labeled membranes from the secB null mutant and the control were analyzed by two-dimensional electrophoresis (Fig. 4, A and B). Silver-stained and 关35S兴methionine-labeled spots were matched and
quantified as described under “Experimental Procedures.” Proteins were identified by MALDI-TOF mass spectrometry from Coomassie-stained protein spots. Details are given in supplementary Table 3.
TABLE 3
Quantification of silver-stained and 关35S兴methionine-labeled two-dimensional electrophoresis outer membrane maps
Defining the Role of E. coli SecB
Defining the Role of E. coli SecB
The pro-OmpA isolated from [35S]methionine-labeled aggregates
disappears in a chase, showing that the protein aggregates are dynamic.
Proteins extracted from aggregates may either be degraded or get a
second chance to be translocated. Based on recent observations that
Ibp/ClpB/DnaK-mediated reactivation of aggregated proteins plays an
important role in viability of the E. coli cell (41, 51), we suggest that the
aggregates in the secB null mutant are actively reactivated for translocation rather than being degraded.
The composition of the aggregates and whole cell lysates of the secB
null mutant do not point to a significant role of SecB in the folding of
cytoplasmic proteins. However, a recent study suggests that SecB can
play a significant role in the folding of cytoplasmic proteins under specialized conditions, e.g. in the absence of both the chaperones DnaK and
TF (13).
SecB Improves Secretion Efficiency but Is Not Required for Secretion
per se—The majority of the proteins we identified in the aggregates
isolated from the secB null mutant are secretory proteins, and, correspondingly, the two-dimensional electrophoresis analysis of whole cell
lysates indicated that deletion of secB affects the targeting kinetics of
secretory proteins. The secB null mutation did not cause any changes in
the levels of the Sec-translocase core components SecA, -Y, or -E, indi-
APRIL 14, 2006 • VOLUME 281 • NUMBER 15
4
L. Baars and J. W. de Gier, unpublished results.
JOURNAL OF BIOLOGICAL CHEMISTRY
10033
Downloaded from www.jbc.org at Stockholm Universitetsbibliotek on August 6, 2007
FIGURE 5. Identification of SecB-dependent secretory proteins. Targeting of the
potential SecB substrates, DegP, FhuA, FkpA, OmpT, OmpX, OppA, TolB, TolC, YbgF, YcgK
YgiW, and YncE, was monitored in the secB null mutant (SecB⫺) and the control (SecB⫹)
using the classical pulse approach, combined with specific immunoprecipitations (see
“Experimental Procedures”). To facilitate immunoprecipitations of the potential SecB
substrates, a hemagglutinin tag was attached to the C terminus of each protein. The
SecB-independent secretory protein ␤-lactamase (AmpC) was used as a control to monitor if the hemagglutinin tag could confer SecB dependence. m, mature; p, precursor.
cating that the translocase capacity was not compromised. Furthermore, the secB null mutation did not induce a PspA response, indicating
that the electrochemical potential, which plays an important role in
protein secretion, is not affected by the absence of SecB. Thus the phenotype of the secB null mutation is a direct consequence of the absence
of SecB. This urged us to take a closer look at the secretome of the secB
null mutant.
Our analysis of the outer membrane proteome showed that the
steady state levels of most proteins were unaffected. However, assembly
of a considerable number of outer membrane proteins into the outer
membrane was delayed in the secB null mutant. Recently, an outer
membrane protein complex consisting of the proteins YeaT, NlpB,
YfgL, and YfiO has been shown to act as an insertion machinery for
outer membrane proteins (52, 53). In our outer membrane two-dimensional electrophoresis gels, we identified YeaT and lipoprotein NlpB.
The steady state levels of these components were unaffected in the secB
null strain. The analysis of radiolabeled proteins showed no significant
differences in the levels of YeaT and NlpB. This suggests that the outer
membrane protein insertion capacity is not affected in the secB null
mutant, which is consistent with the absence of cell envelope stress
responses. The characterization of the outer membrane proteome of
the secB null mutant along with the observation that the secB null mutation does not cause any significant protein misfolding/aggregation in
the periplasm/outer membrane indicates that the bottleneck created by
the absence of SecB is at the level of sorting of outer membrane proteins
across the inner membrane rather than their sorting in the cell envelope,
and that SecB is not essential for targeting of these proteins to the
Sec-translocase but rather facilitates their targeting.
The levels of a few processed secretory proteins (FhuA, FhuE, OmpT,
and OppA) go drastically down in the absence of SecB. Notably, none of
these proteins were detected in the aggregates. It is tempting to speculate that in the absence of SecB the precursor forms of these proteins are
more prone to proteolysis, thereby lowering their levels.
Comparative Proteome Analysis as a Platform for the Identification of
SecB Substrates—We used a pulse-labeling approach to directly monitor the SecB dependence of a subset of aggregated secretory proteins, as
well as secretory proteins that were affected in two-dimensional electrophoresis maps of whole cell lysates and the outer membrane of the
secB null mutant. Strikingly, translocation of all tested potential SecB
substrates was indeed SecB-dependent. This clearly demonstrates that
the comparative proteomics approach is an excellent platform for the
identification of SecB-dependent secretory proteins. Thus far, bioinformatic analysis of these novel and previously established SecB substrates
has not lead to the identification of a common denominator,4 stressing
the importance of experimentation in protein targeting research.
Recently, pulse-labeling experiments showed that the SRP pathway is
required for efficient targeting of the murein lipoprotein Lpp to the
Sec-translocase (20). However, the identification of the precursor form
of Lpp in the aggregates in the current study strongly suggests that
targeting of at least a small fraction of Lpp (which is the most highly
expressed protein in E. coli) also depends on SecB. Our observations
thus further strengthen the idea that selected proteins can be targeted
by both the SRP and SecB targeting pathways (20, 35–37).
Conclusions—The analysis of a secB null mutant using comparative
proteomics clearly points to a primary role of the chaperone SecB in
facilitating targeting of secretory proteins to the Sec-translocase, and
has enabled us to more than triple the number of known SecB substrates. This shows for the first time that comparative proteomics is a
Defining the Role of E. coli SecB
very powerful tool to study protein targeting pathways and their substrates in E. coli.
Acknowledgments—Bacterial strains and plasmid pE63 were a kind gift from
Ken-ichi Nishiyama; Ben de Kruijff, Annemieke van Dalen, Arnold Driessen,
Jan Tommassen, Axel Mogk, Bernd Bukau, and Joen Luirink are thanked for
antisera. Gunnar von Heijne, Joen Luirink, and Dirk-Jan Slotboom are
thanked for valuable discussions.
REFERENCES
10034 JOURNAL OF BIOLOGICAL CHEMISTRY
VOLUME 281 • NUMBER 15 • APRIL 14, 2006
Downloaded from www.jbc.org at Stockholm Universitetsbibliotek on August 6, 2007
1. Manting, E. H., and Driessen, A. J. (2000) Mol. Microbiol. 37, 226 –238
2. Mori, H., and Ito, K. (2001) Trends Microbiol. 9, 494 –500
3. Van den Berg, B., Clemons, W. M., Jr., Collinson, I., Modis, Y., Hartmann, E., Harrison, S. C., and Rapoport, T. A. (2004) Nature 427, 36 – 44
4. Hartl, F. U., Lecker, S., Schiebel, E., Hendrick, J. P., and Wickner, W. (1990) Cell 63,
269 –279
5. Wickner, W., Driessen, A. J., and Hartl, F. U. (1991) Annu. Rev. Biochem. 60, 101–124
6. Fekkes, P., den Blaauwen, T., and Driessen, A. J. (1995) Biochemistry 34,
10078 –10085
7. Fekkes, P., van der Does, C., and Driessen, A. J. (1997) EMBO J. 16, 6105– 6113
8. Fekkes, P., and Driessen, A. J. (1999) Microbiol. Mol. Biol. Rev. 63, 161–173
9. Randall, L. L., and Hardy, S. J. (2002) Cell Mol. Life Sci. 59, 1617–1623
10. Dekker, C., de Kruijff, B., and Gros, P. (2003) J. Struct. Biol. 144, 313–319
11. Xu, Z., Knafels, J. D., and Yoshino, K. (2000) Nat. Struct. Biol. 7, 1172–1177
12. Knoblauch, N. T., Rudiger, S., Schonfeld, H. J., Driessen, A. J., Schneider-Mergener, J.,
and Bukau, B. (1999) J. Biol. Chem. 274, 34219 –34225
13. Ullers, R. S., Luirink, J., Harms, N., Schwager, F., Georgopoulos, C., and Genevaux, P.
(2004) Proc. Natl. Acad. Sci. U. S. A. 101, 7583–7588
14. Shimizu, H., Nishiyama, K., and Tokuda, H. (1997) Mol. Microbiol. 26, 1013–1021
15. Kumamoto, C. A., and Beckwith, J. (1985) J. Bacteriol. 163, 267–274
16. Hewitt, C. J., and Nebe-Von-Caron, G. (2004) Adv. Biochem. Eng. Biotechnol. 89, 197–223
17. Osborn, M. J., Gander, J. E., and Parisi, E. (1972) J. Biol. Chem. 247, 3973–3986
18. Osborn, M. J., Gander, J. E., Parisi, E., and Carson, J. (1972) J. Biol. Chem. 247,
3962–3972
19. Froderberg, L., Rohl, T., van Wijk, K. J., and de Gier, J. W. (2001) FEBS Lett. 498, 52–56
20. Froderberg, L., Houben, E. N., Baars, L., Luirink, J., and De Gier, J. W. (2004) J. Biol.
Chem. 279, 31026 –31032
21. Froderberg, L., Houben, E., Samuelson, J. C., Chen, M., Park, S. K., Phillips, G. J.,
Dalbey, R., Luirink, J., and De Gier, J. W. (2003) Mol. Microbiol. 47, 1015–1027
22. VanBogelen, R. A., and Neidhardt, F. C. (1999) Methods Mol. Biol. 112, 21–29
23. Molloy, M. P., Herbert, B. R., Slade, M. B., Rabilloud, T., Nouwens, A. S., Williams,
K. L., and Gooley, A. A. (2000) Eur. J. Biochem. 267, 2871–2881
24. Peltier, J. B., Friso, G., Kalume, D. E., Roepstorff, P., Nilsson, F., Adamska, I., and van
Wijk, K. J. (2000) Plant Cell 12, 319 –341
25. Schagger, H., Aquila, H., and Von Jagow, G. (1988) Anal. Biochem. 173, 201–205
26. Oakley, B. R., Kirsch, D. R., and Morris, N. R. (1980) Anal. Biochem. 105, 361–363
27. Berven, F. S., Karlsen, O. A., Murrell, J. C., and Jensen, H. B. (2003) Electrophoresis 24,
757–761
28. Peltier, J. B., Emanuelsson, O., Kalume, D. E., Ytterberg, J., Friso, G., Rudella, A.,
Liberles, D. A., Soderberg, L., Roepstorff, P., von Heijne, G., and van Wijk, K. J. (2002)
Plant Cell 14, 211–236
29. Zhang, W., and Chait, B. T. (2000) Anal. Chem. 72, 2482–2489
30. Field, H. I., Fenyo, D., and Beavis, R. C. (2002) Proteomics 2, 36 – 47
31. Friso, G., Giacomelli, L., Ytterberg, A. J., Peltier, J. B., Rudella, A., Sun, Q., and Wijk,
K. J. (2004) Plant Cell 16, 478 – 499
32. Tomoyasu, T., Mogk, A., Langen, H., Goloubinoff, P., and Bukau, B. (2001) Mol.
Microbiol. 40, 397– 413
33. Davey, H. M., and Kell, D. B. (1996) Microbiol. Rev. 60, 641– 696
34. Laminet, A. A., Kumamoto, C. A., and Pluckthun, A. (1991) Mol. Microbiol. 5, 117–122
35. Sijbrandi, R., Urbanus, M. L., Ten Hagen-Jongman, C. M., Bernstein, H. D., Oudega,
B., Otto, B. R., and Luirink, J. (2003) J. Biol. Chem. 278, 4654 – 4659
36. Peterson, J. H., Woolhead, C. A., and Bernstein, H. D. (2003) J. Biol. Chem. 278,
46155– 46162
37. Schierle, C. F., Berkmen, M., Huber, D., Kumamoto, C., Boyd, D., and Beckwith, J.
(2003) J. Bacteriol. 185, 5706 –5713
38. Darwin, A. J. (2005) Mol. Microbiol. 57, 621– 628
39. Wild, J., Walter, W. A., Gross, C. A., and Altman, E. (1993) J. Bacteriol. 175,
3992–3997
40. Kerner, M. J., Naylor, D. J., Ishihama, Y., Maier, T., Chang, H. C., Stines, A. P., Georgopoulos, C., Frishman, D., Hayer-Hartl, M., Mann, M., and Hartl, F. U. (2005) Cell
122, 209 –220
41. Mogk, A., Deuerling, E., Vorderwulbecke, S., Vierling, E., and Bukau, B. (2003) Mol.
Microbiol. 50, 585–595
42. Seong, I. S., Oh, J. Y., Lee, J. W., Tanaka, K., and Chung, C. H. (2000) FEBS Lett. 477,
224 –229
43. Seol, J. H., Yoo, S. J., Shin, D. H., Shim, Y. K., Kang, M. S., Goldberg, A. L., and Chung,
C. H. (1997) Eur. J. Biochem. 247, 1143–1150
44. Arsene, F., Tomoyasu, T., and Bukau, B. (2000) Int. J. Food Microbiol. 55, 3–9
45. Rosen, R., and Ron, E. Z. (2002) Mass Spectrom. Rev. 21, 244 –265
46. Smith, V. F., Hardy, S. J., and Randall, L. L. (1997) Protein Sci. 6, 1746 –1755
47. Carrio, M. M., and Villaverde, A. (2003) FEBS Lett. 537, 215–221
48. Duguay, A. R., and Silhavy, T. J. (2004) Biochim. Biophys. Acta 1694, 121–134
49. Deuerling, E., Patzelt, H., Vorderwulbecke, S., Rauch, T., Kramer, G., Schaffitzel, E.,
Mogk, A., Schulze-Specking, A., Langen, H., and Bukau, B. (2003) Mol. Microbiol. 47,
1317–1328
50. Vorderwulbecke, S., Kramer, G., Merz, F., Kurz, T. A., Rauch, T., Zachmann-Brand,
B., Bukau, B., and Deuerling, E. (2004) FEBS Lett. 559, 181–187
51. Weibezahn, J., Tessarz, P., Schlieker, C., Zahn, R., Maglica, Z., Lee, S., Zentgraf, H.,
Weber-Ban, E. U., Dougan, D. A., Tsai, F. T., Mogk, A., and Bukau, B. (2004) Cell 119,
653– 665
52. Wu, T., Malinverni, J., Ruiz, N., Kim, S., Silhavy, T. J., and Kahne, D. (2005) Cell 121,
235–245
53. Ruiz, N., Falcone, B., Kahne, D., and Silhavy, T. J. (2005) Cell 121, 307–317
54. Traxler, B., and Murphy, C. (1996) J. Biol. Chem. 271, 12394 –12400
55. Powers, E. L., and Randall, L. L. (1995) J. Bacteriol. 177, 1906 –1907
Effects of SecE depletion on the inner and outer
membrane proteomes of E. coli
Louise Baars1, Samuel Wagner1, David Wickström1, Mirjam Klepsch1, A. Jimmy
Ytterberg2§, Klaas J. van Wijk2 and Jan-Willem de Gier1*
1
Stockholm University, Center for Biomembrane Research, Department of Biochemistry
and Biophysics, SE-106 91 Stockholm, Sweden.
2
Cornell University, Department of Plant Biology, 332 Emerson Hall, Ithaca, NY 14853,
USA.
§
Present address: University of California Los Angeles, Department of Chemistry and
Biochemistry, Box 951569, Los Angeles, CA 90095-1569, USA.
*Corresponding author:
Jan-Willem de Gier
Tel: +46-8-162420
Fax: +46-8-153679
E-mail: [email protected]
Keywords: E. coli, membrane protein, secretory protein, Sec-translocon, SecE,
proteomics
1
Abstract
It is generally assumed that the Sec-translocon is required for the translocation of most
secretory proteins across, and the insertion of most integral membrane proteins into, the
Escherichia coli inner membrane. To date, protein translocation and insertion has been
studied using focused approaches and a very limited set of model substrates. To study the
Sec-translocon dependence of secretory and inner membrane proteins in a global way, a
comparative sub-proteome analysis of cells depleted of the essential translocon
component SecE, and cells with normal levels of SecE, was carried out. The steady-state
proteomes and the proteome dynamics were evaluated using 1- and 2D gel analysis,
followed
by
mass
spectrometry
based
protein
identification
and
extensive
immunoblotting. The analysis showed that SecE depletion 1) leads to cytosolic
aggregation of secretory proteins, as well as the induction of the cytoplasmic σ32-stress
response, 2) reduced the accumulation of outer membrane proteins, with the exception of
OmpA, Pal and FadL, and 3) had a strong differential effect on the accumulation levels of
inner membrane proteins - steady state levels and insertion of some integral inner
membrane proteins were reduced, while others were not affected or even increased upon
SecE depletion. The inner membrane proteins that were not affected or increased upon
SecE depletion did not contain large translocated domains and/or consisted of only one or
two transmembrane segments. Our study suggests that several secretory and inner
membrane proteins can either make use of Sec-translocon independent pathways or have
superior access to the Sec-translocon.
2
Introduction
The genome of the Gram-negative bacterium Escherichia coli harbors around four
thousand open reading frames (ORFs) (1)(2). Around 25% of these ORFs encode inner
membrane proteins and around 10% encode secretory (i.e., periplasmic and outer
membrane) proteins (3)(4). It is generally assumed that the Sec-translocon is required for
the translocation of most secretory proteins across, and the insertion of most integral
membrane proteins into, the inner membrane (5)(6). The targeting of secretory proteins to
the Sec-translocon is mostly post-translational and can be facilitated by the cytoplasmic
chaperone SecB (7)(8)(9). Inner membrane proteins are targeted to the Sec-translocon via
the SRP-pathway in a co-translational fashion (7)(6). In E. coli, a small number of
proteins are translocated across, or integrated into, the inner membrane via the TwinArginine protein Transport (TAT)-pathway (10)(11)(12).
The core of the Sec-translocon consists of the integral membrane proteins SecY,
SecE, and SecG (13). SecY and SecE, but not SecG, are essential for viability (13). The
crystal structure of the SecYEβ-complex from the Archaeon Methanococcus jannaschii
shows that in E. coli, the ten transmembrane segments of SecY can be divided in two
halves (transmembrane segment 1-5 and 6-10) that are clamped together by the third and
essential transmembrane segment of SecE (14). Recent evidence suggests that although a
SecYEG heterotrimer serves as the protein translocation channel, multiple SecYEG
heterotrimers may cooperate in protein translocation/insertion (15)(16)(17). SecA, an
ATPase that is associated with the Sec-translocon, drives the stepwise translocation of
secretory proteins and large periplasmic loops of inner membrane proteins across the inner
membrane (13). (13)The Sec-translocon associated proteins SecD, SecF, and YajC form a
complex that facilitates protein translocation, but are not required for viability (13). The
SecDFYajC complex is thought to mediate the interplay between the SecYEG-protein
conducting channel and YidC, an essential inner membrane protein which appears to be
involved in the transfer of transmembrane segments from the Sec-translocon into the lipid
bilayer (18)(6)(19)(20). Evidence is accumulating that YidC by itself can also mediate the
insertion of a subset of membrane proteins (6)(20).
The notion that most secretory and inner membrane proteins require the Sectranslocon for translocation and/or insertion is based on studies using focused approaches
and a limited number of model proteins, like the outer membrane protein OmpA and the
inner membrane protein FtsQ (see e.g., (21)(17)). To study the Sec-translocon
3
dependence of secretory and inner membrane proteins in a more global way, we have
performed a comparative sub-proteome analysis of cells depleted of SecE and cells
expressing normal levels of SecE. This approach allowed us to investigate protein mislocalization, aggregation and changes in the composition of the outer and inner
membrane proteomes in cells with strongly reduced Sec-translocon levels (22). Our
analysis showed that upon SecE depletion, secretory proteins aggregate in the cytoplasm
and the cytoplasmic σ32-stress response is induced. This response is activated upon
protein misfolding/aggregation in the cytoplasm (23). Interestingly, the effects of reduced
Sec-translocon levels on the proteomes of the outer and inner membranes were different.
Both steady-state levels and translocation efficiencies of most outer membrane proteins
were reduced. The integral inner membrane proteins showed a differential response to
SecE depletion. The abundance of approximately half of the identified integral inner
membrane proteins was reduced, whereas the abundance of the other inner membrane
proteins was either unaffected or increased. Notably, all inner membrane proteins that
were unaffected or increased upon SecE depletion lack large periplasmic domains and/or
contain only one or two transmembrane segments.
The 'global' analysis of cells with reduced Sec-translocon levels provides several
testable hypotheses and new substrates to further discover guiding principles for protein
translocation and insertion.
Experimental procedures
Strains and culture conditions
In E. coli strain CM124, the chromosomal copy of the gene encoding SecE is inactivated
and placed on a plasmid under control of the promoter of the araBAD operon (24).
CM124 was cultured in standard M9 minimal medium supplemented with thiamine (10
mM), all amino acids (0.7 mg/ml) except for methionine and cysteine, glucose (0.2%
w/v), arabinose (0.2% w/v) and ampicillin (100 μg/ml) at 37ºC in an Innova 4330 (New
Brunswick Scientific) shaker at 180 rpm. Overnight cultures were washed in fresh
medium without arabinose and then diluted to OD600= 0.035 in fresh medium without
arabinose to deplete cells of SecE (‘SecE depleted cells’) or medium containing 0.2 %
arabinose to induce expression of SecE (‘control cells’). Cells were cultured for six
hours. Growth was monitored by measuring the OD600 with a Shimadzu UV-1601
spectrophotometer.
4
SDS-PAGE, 1D Blue Native-PAGE and immunoblot analysis
Immunoblot analysis was used to monitor the protein levels of SecE, SecY, SecG, SecA,
SecD, SecF, YidC, FtsQ, Lep, Fob, Foc, DegP, Skp, OmpA, OmpF, PhoE, IbpA/B, SecB,
Ffh, and PspA in whole cell lysates and/or inner membranes. Whole cells (0.1 OD600
unit), purified inner membranes (5 µg of protein) and aggregates (from 2 OD600 units of
cells) were solubilized in Laemmli solubilization buffer and separated by sodium dodecyl
sulfate (SDS)-PAGE. Proteins were transferred from the polyacrylamide gels to a
polyvinylidene fluoride (PVDF) membrane (Millipore). Membranes were blocked and
decorated with antisera to the components listed above essentially as described before
(25). Proteins were detected with HRP-conjugated secondary antibodies (Bio-Rad) using
the ECL system (according to the instructions of the manufacturer, GE Healthcare) and a
Fuji LAS 1000-Plus CCD camera. Blots were quantified using the Image Gauge 3.4
software (Fuji). Experiments were repeated with three independent samples.
To monitor the abundance of the SecYEG-protein conducting channel, inner
membrane vesicles were subjected to 1D Blue Native (BN)-PAGE (26) followed by
immunoblot analysis using antibodies to SecY, SecE, and SecG. Inner membrane pellets
(20 μg of protein) were solubilized in buffer containing 750 mM 6-aminocaproic acid,
50 mM Bis-Tris-HCl (pH 7.0 at 4°C) and freshly prepared 0.5% (w/v) n-dodecyl-β-Dmaltopyranoside (DDM). After removal of unsolubilized material by centrifugation
(100.000 x g, 30 minutes), Serva Blue G was added to a final concentration of 0.5% (w/v)
and the samples were loaded onto the first dimension gel. The 0.02% Serva Blue G
cathode buffer of the BN-PAGE was exchanged to a 0.002% Serva Blue G cathode
buffer after 1/3 of the run in order to prevent excessive binding of Coomassie dye to the
PVDF membrane in the subsequent transfer step. Ferritin (440 & 880 kDa), aldolase
(158 kDa) and albumin (66 kDa) (GE Healthcare) were used as molecular weight
markers. Proteins were transferred to PVDF membrane, detected by antisera to SecY,
SecE, and SecG and quantified as described above.
Protein translocation assay
Translocation of OmpA was monitored essentially as described previously (27). Cultures
corresponding to 0.4 OD600 unit were labeled with [35S]methionine (60 µCi/ml, 1 Ci=37
GBq) for 30 seconds followed by precipitation in 10% trichloroacetic acid (TCA), either
directly or after a chase with cold methionine (final concentration 0.5 mg/ml) for 3 and
10 minutes. TCA-precipitated samples were washed with acetone, resuspended in 10 mM
5
Tris-HCl (pH 7.5), 2% SDS and immunoprecipitated with antiserum to OmpA. The
OmpA precipitate was subjected to standard SDS-PAGE analysis. Gels were scanned in a
Fuji FLA-3000 phosphorimager and quantified as described above.
Flow cytometry and microscopy
Analysis of SecE depleted and control cells using flow cytometry was carried out using a
FACSCalibur (BD Biosciences) instrument. To assess viability, cells were incubated in
the dark at room temperature with 30 μM of propidium iodide (PI) for 15 minutes (28).
For staining of the inner membrane, cells were cultured at 37ºC for 30 minutes with 2 µM
of the membrane-specific fluorophore FM4-64 (Invitrogen) (29). Cultures were diluted in
ice cold PBS to a final concentration of approximately 106 cells per ml. A low flow rate
was used throughout data collection with an average of 250 events per second. Forward
and side scatter acquisition was used for comparison of cell morphology (9). Data
acquisition was performed using CellQuest software (BD Biosciences) and data were
analyzed with FloJo software (Tree Star).
For microscopy, cells were mounted on a slide and immobilized in 1% low
melting temperature agarose. Microscopy was performed on a Zeiss Axioplan2
fluorescence microscope equipped with an Orca-ER camera (Hamamatsu). Images were
processed with the AxioVision 4.5 software from Zeiss.
Isolation and analysis of protein aggregates
Protein aggregates were extracted from whole cells essentially as described before (30).
Cells corresponding to 75 OD600 units were used for each aggregate extraction. The
protein content of cell lysates and aggregate extracts was determined with the BCA assay
according to the instructions of the manufacturer (Pierce). Aggregates were analyzed by
SDS-PAGE using 24 cm long 8-16% acrylamide gradient gels. Gels were stained with
Coomassie Brilliant Blue R-250 and proteins were identified by mass spectrometry (MS)
as described below. The aggregate fraction was also subjected to in solution digest
followed by nano-liquid chromatography electrospray tandem MS (nanoLC-ESI-MS/MS)
essentially as described before (31).
Isolation of inner and outer membranes
Inner and outer membranes were isolated essentially as described before (32). Membrane
fractions used for immunoblot analysis were prepared from non-radio-labeled cultures.
6
Membrane fractions used for analysis by 2D-gel electrophoresis (outer membranes) or
2D-BN-PAGE (inner membranes) were prepared from a mixture of labeled and unlabeled
cells as outlined in the Supplemental figure 1. Cells corresponding to 1000 OD600 units
were cultured as described above. An aliquot of 10 OD600 units of cells was labeled with
[35S]methionine (60 µCi/ml, 1 Ci=37 GBq) for 1 minute, followed by a chase of 10
minutes with cold methionine (final concentration 5 mg/ml). Labeled cells were
subsequently collected by centrifugation and cell pellets were snap-frozen in liquid
nitrogen. The rest of the cells (990 OD600 units) was harvested by centrifugation and
washed once with buffer K (50 mM triethanolamine (TEA), 250 mM sucrose, 1 mM
ethylenediaminetetraacetic acid (EDTA), 1 mM dithiothreitol (DTT), pH 7.5). The cell
pellets were snap-frozen in liquid nitrogen and stored at -80°C. Before breaking the cells,
labeled and unlabeled cells from the same culture were pooled in a 1:100 ratio. The
resulting mixture was resuspended in 8 ml buffer K supplemented with 0.1 mg/ml
Pefabloc and 5 µg/ml DNAse and lysed by two cycles of French press (18,000 psi). The
lysate was cleared of unbroken cells by centrifugation at 8000 x g for 20 minutes, and the
total membrane fraction was collected by centrifugation at 100.000 x g for 1 hour. The
membrane pellet was resuspended in 1 ml of buffer M (50 mM TEA, 1 mM EDTA,
1 mM DTT, pH 7.5) and loaded on top of a six-step sucrose gradient (from bottom to
top); 0.5 ml 55%, 1.5 ml 50%, 1.5 ml 45%, 2.5 ml 40%, 2.5 ml 35%, 2.5 ml 30% (w/w
sucrose in buffer M). After centrifugation at 210,000 x g for 15 hours, the inner
membrane and outer membrane fractions were collected from the 35% and 45 % sucrose
layers, respectively. The collected fractions were diluted in TEA buffer (50 mM TEA, 1
mM DTT, pH 7.5) to a sucrose concentration below 10%. Membranes were collected by
centrifugation at 170.000 x g for 1 hour and subsequently resuspended in buffer L
(50 mM TEA, 250 mM Sucrose, 1 mM DTT, pH 7.5). The inner membrane fraction was
snap-frozen in liquid nitrogen and the outer membrane fraction was washed in 0.1 mM
sodium carbonate as described before (9). Protein concentrations were determined using
the BCA assay. Samples were stored at -80°C.
2-dimensional gel-electrophoresis (2DE)
Whole cell lysates (1 OD600 unit) and [35S]methionine labeled outer membranes (185 μg
of protein) isolated by density centrifugation were analysed by 2DE using iso-electric
focusing in the first dimension and SDS-PAGE in the second dimension (9). Gels used
for comparative analysis of whole cells lysates were stained with high sensitivity silver
7
stain (33). Gels used for comparative analysis of the outer membrane proteome and all
gels used for MS based identification of proteins were stained with colloidal Coomassie
(34). Most proteins in the outer membrane gels gave rise to multiple spots with the same
molecular mass but different pI. This phenomenon was also observed in the outer
membrane map of E. coli constructed by Molloy et al. (35). Most of these 'trains of spots'
are caused by modifications induced during sample preparation (36), likely due to
stepwise deamidation of the asparagine and glutamine residues, resulting in loss of 1
Dalton and net loss of one positive charge (37).
Analysis of cytoplasmic membrane fractions by 2D BN-PAGE
Comparative 2D Blue Native electrophoresis was performed as described previously
(32). In short, [35S]methionine labeled inner membranes (100 µg of protein) were
solubilized in 0.5% (w/v) DDM and subjected to Blue-Native electrophoresis in the first
dimension and denaturing SDS-PAGE in the second dimension. For calibration, ferritin
(440 & 880 kDa), aldolase (158 kDa) and albumin (66 kDa) (GE Healthcare) were used
as molecular weight markers. Gels were stained with Coomassie Brilliant Blue R-250 (9).
Image analysis and statistics
Stained gels were scanned using a GS-800 densitometer from BioRad. Radio-labeled gels
were scanned in a Fuji FLA-3000 phosphorimager. Spots were detected, matched and
quantified using PDQuest software version 8.0 (BioRad). The analysis of Coomassie
stained and [35S]methionine labeled outer and inner membrane proteins was done on the
same set of gels. In all cases, each analysis set consisted of at least three gels in each
replicate group (i.e., SecE depleted cells and the control). Each gel in a set represented an
independent sample (i.e., from a different bacterial colony, culture and membrane
preparation). Independent samples were subjected to 2DE or 2D BN-PAGE and image
analysis in parallel, i.e., en groupe. Quantities of stained spots were normalized using the
‘total intensity of valid spots’ method to compensate for non-expression related variations
in spot quantities between gels (there were no significant variations in the total spot
quantity between the two groups; SecE depletion and control). Since protein aggregates
can co-sediment with outer membranes during density gradient centrifugation, an
additional normalization step was required for the analysis of the outer membrane gels
(38)(39). First, to distinguish between outer membrane spots and contaminating
aggregate spots, the outer membrane fraction from SecE depleted and control cells were
8
subjected to aggregate extraction as described above. The resulting extracts were
analysed by 2DE and proteins in the aggregates were visualized by staining with colloidal
Coomassie. Spots detected in the gels of the aggregate extract were removed from the
outer membrane analysis set if the intensity of a spot in the aggregate gel was more than
5% of the intensity of the spot detected in the outer membrane gels. The quantities of the
remaining spots were normalized using the ‘total quantity of valid spot’ method to correct
for the contribution of protein aggregates on protein loading. Spots detected by means of
phosphorimaging were normalized using the correction value calculated from the
corresponding Coomassie stained gel to allow correction for errors in protein loading
while retaining differences in labeling efficiency between the control and cells depleted
of SecE. PDQuest was set to detect differences that were found to be statistically
significant using the student t-test and a 95% level of confidence, including qualitative
differences (“on – off responses’’) present in all gels in a group. Saturated spots were
excluded from the analysis.
Mass spectrometry based identification of proteins
Coomassie stained protein spots or bands were excised, washed, digested with modified
trypsin and peptides were extracted manually or automatically (ProPic and Progest,
Genomic Solutions, Ann Arbor, Mi). Peptides were applied to the MALDI target plate as
described previously (40). Mass spectra were obtained automatically by MALDI-TOF
MS in reflectron mode (Voyager-DE-STR; PerSeptive Biosystems, Framingham, MA),
followed by automatic internal calibration using tryptic peptides from autodigestion. The
spectra were analyzed for monoisotopic peptide peaks (m/z range 850-5000) using the
software MoverZ from Genomic Solutions (http://65.219.84.5/moverz.html) with a signal
to noise ratio threshold of 3.0. Matrix and/or auto-proteolytic trypsin fragments were not
removed. Spectral annotations (in particular assignments of mono isotopic masses) were
verified by manual inspection for a large number of measurements. The resulting peptide
mass lists were used to search the SwissProt 45.0 database (release 10/04) for E. coli with
Mascot (v2.0) in automated mode (www.matrixscience.com), using the following search
parameters/criteria: significant protein MOWSE score at P<0.05; no missed cleavages
allowed; variable methionine oxidation; fixed carbamidomethylation of cysteins;
minimum mass accuracy 50 ppm. The search results pages were extracted and analyzed
by an additional in-house filter (Sun and van Wijk, unpublished) applying the following
three criteria for positive identification: i) Minimum MOWSE score ≥50;ii) ≥four
9
matching peptides with an error distribution within ±25 ppm; iii) ≥15% sequence
coverage. False positive rates were less than 1%, as determined by searching with the .pkl
list against the E. coli database (SwissProt 48.1) mixed with a randomized version of the
E. coli database, generated using a Perl script from Matrix science.
Results
Characterization of the SecE depletion strain CM124
The E. coli strain CM124 was used to study protein translocation and insertion upon
depletion of SecE. In CM124, the chromosomal copy of secE is inactivated and a copy of
secE is placed on a plasmid under control of the promoter of the araBAD operon (41).
Cells were cultured aerobically in M9 minimal medium in the presence of arabinose to
induce expression of SecE (these cells will be further referred to as ‘control cells’) and in
the absence of arabinose to deplete cells of SecE. Growth was monitored by measuring
the OD600 (Figure 1A). As expected, growth of CM124 cells cultured in the absence of
arabinose was much slower than in control cultures.
For all experiments, cells were harvested six hours after inoculation. At this timepoint, control cells are in the mid-log phase with an OD600 of 0.8, and SecE depleted cells
reached an OD600 of 0.3-0.4. Inner membranes were isolated from control cells and SecE
depleted cells and the levels of SecE, SecY, SecG, SecA, SecD, SecF, and YidC were
analysed by immunoblotting. The level of SecE in the membrane of depleted cells was
less than 10% of the SecE detected in membranes prepared from control cells. It should
be noted that the level of SecE in the membrane of the control cells was similar to the
level of SecE in the strain CM124 is derived from (data not shown). The levels of SecY
and SecG in SecE depleted membranes were reduced to 50% and 80%, respectively
(Figure 1B). SecY is degraded by the FtsH protease in the absence of SecE (22). Since
the Sec-translocon is composed of SecY, SecE, and SecG in a 1:1:1 ratio (14), this
indicates that the SecE depleted membrane must contain pools of SecY and SecG, which
are not in a SecYEG-complex. The abundance of SecYEG-heterotrimers in SecE
depleted membranes was monitored by BN-PAGE combined with immunoblotting
(Figure 1C). For this purpose, inner membranes prepared from SecE depleted and nondepleted control cells were solubilized in 0.5% DDM (42). Upon depletion of SecE, the
amount of the SecYEG heterotrimer was strongly reduced. Interestingly, a complex that
most likely represents a SecYG heterodimer could be detected in membranes of SecE
depleted cells but not in control membranes.
10
The amount of SecA was almost doubled in SecE depleted membranes (Figure
1B)(43). The levels of SecD, SecF, and YidC were reduced to approximately 50% upon
SecE depletion. In addition, the steady state levels of the well studied model proteins
FtsQ,
Lep,
Fob,
and
Foc
were
monitored
by
immunoblotting
(Figures
1B)(21)(44)(45)(46)(47). As expected, the accumulation levels of the Sec-translocon
dependent inner membrane proteins FtsQ and Lep were strongly reduced upon SecE
depletion, whereas the level of the Sec-translocon independent protein Foc was only
slightly increased. To our surprise, the level of Fob was somewhat increased. This is
contradictory to previous work proposing that translocation of Fob is dependent on the
Sec-translocon (45).
Upon SecE depletion, translocation of OmpA was delayed but not abolished
(Figure 1D). Notably, the intensity of the total OmpA signal was clearly increased in
SecE depleted cells compared to control cells, indicating that OmpA expression was
induced. We do not have a ready explanation for this observation.
Flow cytometry and microscopy
The integrity of the inner membrane of SecE depleted and control cells was monitored
using propidium iodide (PI) staining combined with flow cytometry (28). 9.0% (+/2.0%) of SecE depleted cells stained fluorescently red with PI, compared to 1.0% (+/0.3%) of the control cells, indicating that SecE depletion did not have a major impact on
the integrity of the inner membrane. Furthermore, we detected a small increase of both
the forward scatter and side scatter of cells depleted for SecE (Figure 2A). This indicates
that SecE depleted cells are slightly bigger than control cells and most likely contain
extra internal structures (i.e., extra membranes and/or protein aggregates). Light
microscopy showed that SecE depleted cells were slightly elongated compared to the
control cells (Figure 2A). The inner membranes of SecE depleted cells and control cells
were stained with the fluorescent dye FM4-64 and cells were analyzed by flow cytometry
(29). The fluorescence of SecE depleted cells was enhanced approximately four times
compared to the control cells (Figure 2B). This is in keeping with the observation that
SecE depletion induces the formation of endoplasmic membranes (48).
11
SecE depletion leads to accumulation of secretory proteins in the cytoplasm and the
induction of the σ32-stress response
Whole cell lysates of SecE depleted and control cells were compared by 2DE and
immunoblot analysis. The comparative 2DE analysis was based on four biological
replicates. Proteins were separated by denaturing immobilized pH gradient (IPG) strips
(pH 4-7) in the first dimension and by Tricine-SDS-PAGE in the second dimension. Gels
were stained with silver or colloidal Coomassie and spot volumes were compared using
PDQuest. This analysis demonstrated that the volumes of 28 spots were significantly
(P<0.05) changed in the lysates of SecE depleted cells compared to the control; the
intensity increased in 13 spots and decreased in 15 spots. The affected spots were excised
and used for protein identification by matrix assisted laser desorption/ionization mass
spectrometry (MALDI-TOF MS) and peptide mass finger printing (PMF) (Figure 3A,
Table 1). Spot statistics and MS data are provided in Supplemental table 1. The effects of
SecE depletion on protein accumulation levels are shown as fold changes (SecE
depletion/control) in Figure 3B.
Accumulation levels of a number of secretory proteins (β-lactamase, DppA, FliY,
LivJ, PotD, OmpC, OmpT, OppA, RbsB, TolB, UshA, YbiS, YehZ, YggE, YodA, and
ZnuA) were reduced in SecE depleted cells. Based on the pI and molecular weight, most
of these spots corresponded to the mature forms of the proteins (Table 1). MetQ, OmpC,
OmpA, and RbsB were identified in spots that were stronger in lysates of SecE depleted
cells. The OmpC spot corresponded to a degraded form of the protein while the increased
RbsB spot most likely corresponded to the precursor form. The OmpA spot likely
corresponded to the precursor form of OmpA, since we observed a peptide mass
matching with a predicted tryptic peptide within the signal sequence.
To study the effect of SecE depletion in more detail, the accumulation levels of
the periplasmic proteins DegP and Skp and the outer membrane proteins OmpA, OmpF,
and, PhoE were monitored by immunoblotting (Figure 3C). Upon SecE depletion, the
precursor form of all these proteins was detected, indicating accumulation in the
cytoplasm due to hampered translocation. In the case of Skp, DegP, and PhoE, this was
accompanied by a decrease of the mature form of the proteins. Interestingly, the total
levels of DegP and Skp were not significantly affected upon SecE depletion. This
suggests that no extracytoplasmic stress responses are activated upon SecE depletion
(49). The accumulation levels of the mature form of OmpA and OmpF were unaffected
by the SecE depletion.
12
Upon SecE depletion, accumulation levels of the σ32-inducible, cytoplasmic
chaperones DnaK, GroEL, GroES and ClpB were increased (Figure 3B). The upregulation of DnaK and GroEL was confirmed by Western-blotting (results not shown).
Since inclusion body proteins IbpA/B, SecA, SecB, Ffh, and phage shock protein A
(PspA) were not identified in the 2D gels, we monitored their accumulation levels by
immunoblotting (Figure 3D). The level of the heat shock chaperones IbpA/B, also part of
the σ32-regulon, was increased. Inclusion body proteins associate with protein aggregates
and facilitate the extraction of proteins from aggregates by ClpB and DnaK (50). In
agreement with the membrane blotting experiments (see above), the total level of SecA
was increased in SecE depleted cells, consistent with insufficient Sec-translocon capacity
(43). Accumulation levels of SecB and Ffh, components involved in the targeting of
secretory and inner membrane proteins to the Sec-translocon, respectively, were both
unaffected upon SecE depletion. This indicates that the protein targeting capacity is not
affected upon SecE depletion. The level of PspA was monitored since the electrochemical
potential plays an important role in protein translocation and the expression of PspA is upregulated when it is affected. Just like in several other translocation and insertion-mutant
strains, a considerable PspA response was detected in SecE depleted cells (51).
Taken together, the up-regulation of SecA and the accumulation of the
unprocessed forms of secretory proteins indicate that protein translocation across the
cytoplasmic membrane is strongly hampered. Furthermore, the accumulation levels of the
σ32-regulated chaperones DnaK, GroEL/ES, ClpB, and IbpA/B are all increased upon
SecE depletion, suggesting that reduced Sec-translocon levels lead to protein missfolding/aggregation in the cytoplasm.
Accumulation of cytoplasmic protein aggregates in SecE depleted cells
Protein aggregates were extracted from whole cells depleted of SecE. The aggregates
from SecE depleted cells contained 2.6% of the total cellular protein compared to 0.3% in
the control. The protein composition of the aggregates was analysed by 1D gel
electrophoresis combined with MALDI TOF MS PMF (Figure 4, Supplemental table 2)
and by nanoLC-ESI-MS/MS of solubilized aggregates digested with trypsin
(Supplemental table 2). In total, 61 proteins were identified in aggregates isolated from
cells depleted of SecE; 19 secretory proteins, 5 inner membrane proteins, 36 cytoplasmic
proteins and one protein with a localization that could not be unambiguously predicted
(Table 2). Among the identified cytoplasmic proteins were the chaperones IbpA, DnaK
13
and DnaJ. The MS/MS analysis revealed that at least four of the secretory proteins,
OmpA, Lpp, β-lactamase and SlyB, contained an uncleaved signal sequence (results not
shown), indicating that these proteins aggregate in the cytoplasm rather than the
periplasm. The identified inner membrane proteins ElaB and YqjD contain one predicted
transmembrane segment, while YhjK contains two. The Penicillin-binding protein 5
(DacA), Penicillin-binding protein 6 (DacC) are probably attached to the inner membrane
via a C-terminal amphiphilic α-helix (52). It is possible that the number of identified
inner membrane proteins is somewhat under-represented due to experimental problems
associated with MS based identification of α-helical membrane proteins (53).
Intensities of the bands in the gel shown in Figure 4 were quantified to get an idea
of the relative abundance of different classes of proteins in the aggregates isolated from
SecE depleted cells. 70% represented secretory proteins and 18% represented
cytoplasmic proteins. The cytoplasmic chaperones DnaK and IbpA together constituted
2% of the total band intensity. The protein content of the remaining bands is unknown.
Effect of SecE depletion on the outer and inner membrane proteomes
To study the effect of SecE depletion on the insertion and composition of the inner and
the outer membrane proteomes, cells were labeled with [35S]methionine. Outer and inner
membranes were subsequently isolated using a combination of French-press and sucrose
gradient centrifugation (Supplemental figure 1). The outer membrane proteome was
analysed by 2DE using IEF in the first dimension and SDS-PAGE in the second
dimension (9). The inner membrane proteome was analysed with backed 2D BN-PAGE
that allows relative quantification (32). Gels were stained with colloidal Coomassie and
[35S]methionine labeled proteins were detected by phosphorimaging. Spot intensities
were quantified and compared using PDQuest. Each analysis set contained at least three
biological replicates and the threshold for acceptance was 95% significance determined
by the student-t test. Spots were excised and used for protein identification by MALDITOF MS and PMF.
Outer membrane proteome – MS analysis of the Coomassie stained spots in the 2D gels
of the outer membrane proteome resulted in the identification of 39 different proteins
from 51 spots (Figure 5A, Supplemental table 3). 40 of these spots could be matched to
spots detected by phosphorimaging (Figure 5B). We found that the outer membrane
fraction of SecE depleted cells was contaminated with aggregates that can co-sediment
14
with outer membranes during density gradient centrifugation (38)(39). The spots
corresponding to aggregated proteins were identified and removed from the analysis set
as described in the ‘Experimental procedures’ (Supplemental table 3, Supplemental
figure 2).
The bar diagram in Figure 5C shows the average fold-change of the spot
intensities (SecE depletion/control) for proteins detected by Coomassie (black) or
phosphorimaging (grey). Statistically significant (P<0.05) fold-changes are indicated by
bold numbers in Table 3 and Supplemental table 3. The steady-state levels and insertion
efficiencies of most outer membrane proteins were reduced upon SecE depletion. Three
outer membrane proteins were unaffected or slightly increased upon SecE depletion;
OmpA, Pal, and FadL. After the 10 minutes chase, the level of [35S]methionine labeled
OmpA detected in the outer membrane of SecE depleted cells was not significantly
affected. Furthermore, the steady-state level of OmpA was increased by approximately
20% (Figure 5A, Table 3, Supplemental table 3). This was in agreement with the pulse
chase analysis shown in Figure 1D, which demonstrates that translocation of OmpA was
slowed down but not abolished upon SecE depletion.
Inner membrane proteome - MS analysis identified proteins in 85 spots of the Coomassie
stained 2D BN-PAGE gels. 28 additional spots could be annotated with the help of our
previously published reference map (Figure 7A, Table 4, Supplemental table 4)(32). 56
proteins were integral membrane proteins and 27 were proteins located at the cytoplasmic
side of the inner membrane, mostly as part of membrane localized complexes. In
addition, five secretory proteins were identified in the 2D BN-PAGE gels.
The bar diagram in Figure 6C shows the effect of the SecE depletion as foldchanges calculated from the average spot intensities (SecE depletion/control) of stained
and [35S]methionine labeled proteins. Statistically significant (P<0.05) fold-changes are
indicated by bold numbers in Table 4 and Supplemental table 4. Since several proteins
were identified in more than one spot, we also calculated the effect of the SecE depletion
on the total level of each identified integral membrane protein (Supplemental table 5).
The total levels of approximately 30 integral membrane proteins were reduced in the
membrane of SecE depleted cells. Notably, the steady levels of all components of the
FtsH-HflKC protease complex were significantly reduced (Table 4, Supplemental table
4). It is possible that this could affect the stability of the inner membrane proteome,
although it should be noted that FtsH-HflKC mediated proteolysis has only been shown
15
for a few membrane proteins (54). The cytochrome bo3 terminal oxidase subunits CyoA
and CyoB were reduced by 75% and 85%, respectively. The biogenesis of CyoA, which
is a lipid modified integral membrane protein, has recently been shown to depend on both
YidC and the Sec-translocon (55)(56)(57).
Interestingly, the accumulation levels of a surprisingly large number of integral
membrane proteins (approximately 30) were not significantly affected or even increased
by SecE depletion (Supplemental table 5). Among the significantly increased proteins
were; Aas, AtpF, MscS, MgtA, NarI, YbbK, YhcB, YhjG, and YajC. We tested if the
effect of SecE depletion could be correlated to different membrane protein properties.
Our analysis showed that protein abundance, topology, hydrophobicity, and the energy
required for membrane integration of the first transmembrane domain (ΔGapp) (58) do not
correlate with the effect on total protein levels upon SecE depletion (data not shown).
However, when the fold-changes (Coomassie staining) were plotted against the number
of amino acids in the largest translocated domain of each protein, we found that almost
all proteins with large periplasmic domains are sensitive to SecE depletion (Figure 7A,
Supplemental table 5). In contrast, almost all proteins that were positively affected by
SecE depletion do not contain any large periplasmic domains. A closer look at the
proteins that do not follow this trend (Aas, YhjG, YbbK, and YhcB) revealed that they
consist of only one or two transmembrane segments. This prompted us to perform a
combined analysis of the effect of number of transmembrane segments and the size of
periplasmic domains. Based on the plot shown in figure 7A, we divided the proteins into
two groups; proteins with large translocated domains (≥60 amino acids) and proteins with
small translocated domains (≤60 amino acids). The 60 amino acid cut-off for Secdependence is in agreement with previous studies (59). The effect of SecE depletion on
these two groups was plotted against the number of transmembrane segments (Figure 7B,
Supplemental table 5). This clearly demonstrated that proteins that do not have large
periplasmic domains are overrepresented among the proteins that are either unaffected
(fold change ≥0.75≤1.25) or positively affected (fold change ≥1.25) by SecE depletion.
The few exceptions are proteins that contain only one or two transmembrane segments.
Collectively, our analysis suggests that many proteins that do not contain large
periplasmic domains and/or contain one or two transmembrane segment(s) can insert
efficiently into the membrane when the levels of the Sec-translocon are reduced.
16
Discussion
So far, the Sec-translocon dependence of protein translocation and insertion in E. coli has
been studied using focused approaches and a very limited set of model substrates. To
characterize the Sec-translocon dependence of secretory and inner membrane proteins in
a global way, we have performed a comparative sub-proteome analysis of E. coli cells
depleted of the Sec-translocon component SecE. Depletion of SecE resulted in 90% (10fold) reduction in SecE and a 50%-70% (2-fold) reduction in the SecY,G translocon
components. 1D BN-PAGE combined with immunoblotting showed that the level of the
SecYEG-complex
was
strongly
reduced
upon
SecE-depletion.
Furthermore,
accumulation of SecA was increased 1.7 fold. This indicates that translocation of the
translocation monitor SecM is hampered due to insufficient Sec-translocon capacity,
leading to increased expression of SecA (43). SecE depletion did not affect the
accumulation levels of SecB or Ffh, which are both involved in protein targeting.
Our analysis of the subproteomes of cells with strongly reduced Sec-translocon
levels resulted in three main observations and conclusions. Reduced Sec-translocon
levels 1) resulted in the accumulation of secretory proteins in the cytoplasm, the
formation of protein aggregates and a σ32-response, 2) negatively affected levels of all
constituents of the outer membrane proteome, with the exception of OmpA, Pal, and
FadL, and 3) had differential effects on inner membrane proteins - steady state levels and
insertion of some integral inner membrane proteins were reduced, while others were not
affected or even increased in the membranes of SecE depleted cells. The proteins that
were not affected or increased upon SecE depletion did not contain large translocated
domains and/or consisted of only one or two transmembrane segments. Below these main
observations and conclusions are explained and discussed in more detail.
Reduced Sec-translocon levels induce the formation of protein aggregates
Upon SecE depletion, secretory proteins accumulate in the cytoplasm, leading to
aggregate formation and the induction of a σ32-response. It was estimated that secretory
proteins made up 70% of the total protein in the aggregates. The MS/MS analysis
revealed that at least three secretory proteins, OmpA, Lpp, and β-lactamase, contained an
uncleaved signal sequence, pointing to aggregation in the cytoplasm rather than the
periplasm. Sequence analysis of all the aggregated secretory proteins with the
aggregation propensity prediction program Tango showed that their signal sequences are
more aggregation prone than the mature part of these proteins ((60), (data not shown)).
17
This suggests that secretory proteins are more prone to aggregation in the cytoplasm than
in the periplasm. The accumulation of secretory proteins upon SecE depletion induced a
σ32-response, leading to increased levels of the cytoplasmic chaperones IbpA/B, DnaK,
GroEL and ClpB. These chaperones protect proteins from aggregation and are also
involved in the disaggregation, refolding and degradation of aggregated proteins
(61)(62)(63). IbpA/B, DnaK, GroEL and the Lon protease were among the proteins that
were identified in the aggregate fraction. Thus, aggregated secretory proteins may either
be actively reactivated for translocation or degraded. Recently, we have shown that in an
E. coli secB null mutant in the cytoplasm aggregated OmpA is extracted from aggregates
(9). Although our analysis did not provide any evidence for aggregate formation in the
periplasm we cannot exclude this.
Only a few inner membrane proteins were identified in the aggregates. This may
be explained by the efficient degradation by SsrA mRNA dependent tagging of stalled
nascent chains of co-translationally targeted membrane proteins, and subsequent turnover
by proteases (64)(65). Sec-translocon independent membrane insertion mechanisms
could also explain the low abundance of inner membrane proteins in the aggregates (see
below).
SecE depletion reduces the insertion and steady-state levels of most outer membrane
proteins
A direct correlation between the effects on the outer membrane proteome and Sectranslocon dependence is difficult to make since key players involved in outer membrane
protein biogenesis – like YaeT and Skp (66) – are affected by SecE depletion.
Nevertheless, the analysis of the outer membrane proteome indicates that translocation of
proteins across the inner membrane is hamped but not blocked upon SecE depletion. The
components of the outer membrane proteome were differentially affected by the
depletion of SecE. OmpA, FadL, and Pal were unaffected or increased in the outer
membrane upon SecE depletion, while most other outer membrane proteins were reduced
to different extents. One explanation is superior accessof these proteins tothe Sectranslocon. Signal peptide based selective modulation of protein translocation occurs in
the ER during stress (67). If such a mechanism for modulation of translocation efficiency
exists in E. coli, it should become apparent upon lowering Sec-translocon levels. We
were not able to identify any signal sequence characteristics (e.g., hydrophobicity, charge
18
distribution, length) that correlated with the differential effects on the constituents of the
outer membrane proteome upon depletion of SecE (data not shown).
Using a small number of model proteins it has been shown that DnaK can keep
outer membrane proteins, but not periplasmic proteins, in a prolonged export competent
state upon depletion of SecA (68). This suggests that affinity towards DnaK and other
chaperones could also affect the translocation efficiency of secretory proteins during
SecE depletion. However, it should be noted that both periplasmic and outer membrane
proteins were identified in aggregates from SecE depleted cells (Figure 4, Table 2). We
were not able to extend our analysis to include the periplasmic proteome, since it was not
possible to isolate sufficiently pure periplasmic fractions from SecE depleted cells. Thus,
it is not clear if the outer membrane proteome is in fact less affected than the periplasmic
proteome or if the chaperone mediated protection of secretory proteins is independent of
the final destination of the protein.
It is conceivable, that proteins that under normal conditions use the Sectranslocon can cross the membrane via alternative pathways upon SecE depletion. It has
been shown that there are secretory proteins that are promiscuous; i.e., can use both the
Sec- and TAT-protein translocation pathways (12). In this respect it should be noted that
the TAT-pathway is still operationalupon SecE depletion (69). Interestingly, our Blue
Native analysis revealed that SecA dimers accumulated at the inner membrane of SecE
depleted cells. Impaired Sec-translocon function results in increased expression of SecA,
mediated by the secretion monitor SecM (43). SecA is the peripheral subunit of the Sectranslocon and responsible for the ATP dependent translocation of secretory proteins and
large periplasmic domains of integral membrane proteins. The increased levels of SecA
may enhance translocation efficiency when the pressure on the translocon is particularly
high. Recently, it has been shown that the Sec-translocon catalyzes the monomerization
of the SecA dimer (70). This could explain the accumulation of SecA dimers at the
membrane observed upon SecE depletion. It has also been proposed that the SecA dimer
by itself can act as an alternative translocase for secretory proteins (71). If this is indeed
the case, it could mean that the SecA dimer functions as a backup translocon when Sectranslocon capacity is not sufficient.
SecE depletion has differential effects on the inner membrane proteome
Depletion of SecE resulted in reduced steady state levels and integration of
approximately 30 integral membrane proteins, while another 30 were not significantly
19
affected or even increased. The immunoblot and 2D BN-PAGE analysis showed that
FtsQ, Lep, and the cytochrome bo3 subunit CyoA were all strongly reduced in the inner
membrane of SecE depleted cells. These proteins have been shown to require both the
Sec-translocon and YidC for proper assembly into the inner membrane (21)(44)(56)(55).
A surprisingly large number of inner membrane proteins were either unaffected or
even increased in the membrane of SecE depleted cells. Among the unaffected proteins
was the Foc subunit of the Fo sector of the ATP synthase. This is in keeping with previous
studies showing that Foc is inserted into the inner membrane in a Sec-translocon
independent but YidC dependent fashion (45)(46)(47). The efficient insertion of Foc
demonstrates that the YidC pathway was operational, although YidC levels were 50%
reduced upon SecE depletion. Interestingly, the level of Foa and Fob, also components of
the Fo sector of the ATP synthase, were slightly increased upon SecE depletion (Table 4
and Figure 1C, respectively). It has been proposed that insertion of both Foa and Fob is
dependent on the Sec-translocon (45). However, it should be noted that integration was
studied in cells depleted of SecDF rather than SecE/Y (45). Furthermore, Fob integration
was examined using a Fob variant with a N-terminal T7 tag, which may affect the
biogenesis requirements of Fob.
The analysis of the inner membrane proteome of SecE depleted and control cells
allowed us to search for common features among proteins that were reduced, unaffected
or increased in the membrane of SecE depleted cells. We found no correlations between
the effect of SecE depletion and properties of the first transmembrane segment (e.g.,
hydrophobicity, ΔGapp for insertion (58)) as could be expected if the differences were due
to different affinities towards the residual translocons. However, we found that the inner
membrane proteins that were either unaffected or increased upon depletion of SecE, do
not contain any large periplasmic domains and/or consist of only one or two
transmembrane segments (Figure 7, Supplemental table 5). Interestingly, all the proteins
that so far have been shown to integrate via the Sec-translocon independent/YidC
dependent pathway (M13, Pf3, Foc, MscL and a C-terminally truncated ProW variant)
share similar features (20). Thus, it is tempting to speculate that the proteins that are
unaffected or increased in the membrane of SecE depleted cells are potential substrates of
the YidC only pathway. However, it should be noted that membrane integration of the E.
coli inner membrane protein KdpD, which consists of four transmembrane segments and
exceptionally big cytoplasmic N- and C-terminal domains, is not affected by either SecE
or YidC depletion (72). Based on these observations it has been proposed that an inner
20
membrane assembly pathway, which is independent of both the Sec-translocon and YidC,
may exist in E. coli (72). It is also possible that some integral membrane proteins, just
like some secretory proteins, are promiscuous; i.e., use the insertion pathway that is
available. Clearly, the observation that the levels of such a large number of inner
membrane proteins are not affected or go up upon SecE depletion is intriguing and
warrants further investigations. For instance, it would be interesting to use a proteomics
approach to analyze membrane protein biogenesis in YidC depletion and SecE/YidC
double depletion backgrounds.
In conclusion, substantial protein translocation and insertion activity was still observed in
SecE depleted cells. This suggests that the significance of Sec-translocon independent
translocation/insertion and pathway promiscuity in outer and inner membrane protein
biogenesis have been underestimated. Our study provides several testable hypotheses and
new substrates to further discover guiding principles for protein translocation and
insertion in the model organism E. coli.
Acknowledgements
Claudia Wagner is thanked for assistance with the flow cytometry experiments. Dirk-Jan
Scheffers and Joen Luirink are thanked for critically reading the manuscript. This
research was supported by grants from the Swedish Research Council, the Carl Tryggers
Stiftelse, the Marianne and Marcus Wallenberg Foundation and the SSF supported Center
for Biomembrane Research to JWdG, and a grant from The Swedish Foundation for
International Cooperation in Research and Higher Education (STINT) to JWdG and
KJvW. Proteomics infrastructure was supported by a grant from NYSTAR to KJvW.
21
References
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
Blattner, F. R., Plunkett, G., Bloch, C. A., Perna, N. T., Burland, V., Riley, M.,
Collado-Vides, J., Glasner, J. D., Rode, C. K., Mayhew, G. F., Gregor, J., Davis,
N. W., Kirkpatrick, H. A., Goeden, M. A., Rose, D. J., Mau, B., and Shao, Y.
(1997) Science 277, 1453-1462
Baba, T., Ara, T., Hasegawa, M., Takai, Y., Okumura, Y., Baba, M., Datsenko, K.
A., Tomita, M., Wanner, B. L., and Mori, H. (2006) Mol Syst Biol 2, 2006 0008
Daley, D. O., Rapp, M., Granseth, E., Melen, K., Drew, D., and von Heijne, G.
(2005) Science 308, 1321-1323
Rey, S., Acab, M., Gardy, J. L., Laird, M. R., deFays, K., Lambert, C., and
Brinkman, F. S. (2005) Nucleic Acids Res 33, D164-168
Osborne, A. R., Rapoport, T. A., and van den Berg, B. (2005) Annu Rev Cell Dev
Biol 21, 529-550
Luirink, J., von Heijne, G., Houben, E., and de Gier, J. W. (2005) Annu Rev
Microbiol 59, 329-355
Driessen, A. J. M., Manting, E. H., and van der Does, C. (2001) Nature Structural
Biology 8, 492-498
Randall, L. L., and Hardy, S. J. (2002) Cell Mol Life Sci 59, 1617-1623
Baars, L., Ytterberg, A. J., Drew, D., Wagner, S., Thilo, C., van Wijk, K. J., and
de Gier, J. W. (2006) J Biol Chem 281, 10024-10034
Hatzixanthis, K., Palmer, T., and Sargent, F. (2003) Mol Microbiol 49, 1377-1390
Lee, P. A., Tullman-Ercek, D., and Georgiou, G. (2006) Annu Rev Microbiol 60,
373-395
Tullman-Ercek, D., DeLisa, M. P., Kawarasaki, Y., Iranpour, P., Ribnicky, B.,
Palmer, T., and Georgiou, G. (2007) J Biol Chem 282, 8309-8316
Manting, E. H., and Driessen, A. J. (2000) Mol Microbiol 37, 226-238
Van den Berg, B., Clemons, W. M., Jr., Collinson, I., Modis, Y., Hartmann, E.,
Harrison, S. C., and Rapoport, T. A. (2004) Nature 427, 36-44
Mitra, K., Schaffitzel, C., Shaikh, T., Tama, F., Jenni, S., Brooks, C. L., 3rd, Ban,
N., and Frank, J. (2005) Nature 438, 318-324
Mitra, K., Frank, J., and Driessen, A. (2006) Nat Struct Mol Biol 13, 957-964
Osborne, A. R., and Rapoport, T. A. (2007) Cell 129, 97-110
Nouwen, N., and Driessen, A. J. (2002) Mol Microbiol 44, 1397-1405
Xie, K., Kiefer, D., Nagler, G., Dalbey, R. E., and Kuhn, A. (2006) Biochemistry
45, 13401-13408
Kiefer, D., and Kuhn, A. (2007) Int Rev Cytol 259, 113-138
Facey, S. J., and Kuhn, A. (2004) Biochim Biophys Acta 1694, 55-66
Akiyama, Y., Kihara, A., Tokuda, H., and Ito, K. (1996) J Biol Chem 271, 3119631201
Arsene, F., Tomoyasu, T., and Bukau, B. (2000) Int J Food Microbiol 55, 3-9
Traxler, B., and Murphy, C. (1996) J Biol Chem 271, 12394-12400
Froderberg, L., Rohl, T., van Wijk, K. J., and de Gier, J. W. (2001) FEBS Lett
498, 52-56
22
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
49.
Schagger, H., and von Jagow, G. (1991) Anal Biochem 199, 223-231
Froderberg, L., Houben, E., Samuelson, J. C., Chen, M. Y., Park, S. K., Phillips,
G. J., Dalbey, R., Luirink, J., and de Gier, J. W. L. (2003) Mol Microbiol 47,
1015-1027
Hewitt, C. J., and Nebe-Von-Caron, G. (2004) Adv Biochem Eng Biotechnol 89,
197-223
Fishov, I., and Woldringh, C. L. (1999) Mol Microbiol 32, 1166-1172
Tomoyasu, T., Mogk, A., Langen, H., Goloubinoff, P., and Bukau, B. (2001) Mol
Microbiol 40, 397-413
Peltier, J. B., Ytterberg, A. J., Sun, Q., and van Wijk, K. J. (2004) J Biol Chem
279, 49367-49383
Wagner, S., Baars, L., Ytterberg, A. J., Klussmeier, A., Wagner, C. S., Nord, O.,
Nygren, P. A., van Wijk, K. J., and de Gier, J. W. (2007) Mol Cell Proteomics
Oakley, B. R., Kirsch, D. R., and Morris, N. R. (1980) Anal Biochem 105, 361363
Neuhoff, V., Arold, N., Taube, D., and Ehrhardt, W. (1988) Electrophoresis 9,
255-262
Molloy, M. P., Herbert, B. R., Slade, M. B., Rabilloud, T., Nouwens, A. S.,
Williams, K. L., and Gooley, A. A. (2000) Eur J Biochem 267, 2871-2881.
Berven, F. S., Karlsen, O. A., Murrell, J. C., and Jensen, H. B. (2003)
Electrophoresis 24, 757-761
Zabrouskov, V., Han, X., Welker, E., Zhai, H., Lin, C., van Wijk, K. J., Scheraga,
H. A., and McLafferty, F. W. (2006) Biochemistry 45, 987-992
Laskowska, E., Bohdanowicz, J., Kuczynska-Wisnik, D., Matuszewska, E.,
Kedzierska, S., and Taylor, A. (2004) Microbiology 150, 247-259
Marani, P., Wagner, S., Baars, L., Genevaux, P., de Gier, J. W., Nilsson, I.,
Casadio, R., and von Heijne, G. (2006) Protein Sci 15, 884-889
Peltier, J. B., Emanuelsson, O., Kalume, D. E., Ytterberg, J., Friso, G., Rudella,
A., Liberles, D. A., Soderberg, L., Roepstorff, P., von Heijne, G., and van Wijk,
K. J. (2002) Plant Cell 14, 211-236
Traxler, B., and Murphy, C. (1996) J Biol Chem 271, 12394-12400
Bessonneau, P., Besson, V., Collinson, I., and Duong, F. (2002) Embo J 21, 9951003
Nakatogawa, H., Murakami, A., and Ito, K. (2004) Curr Opin Microbiol 7, 145150
Dalbey, R. E., and Chen, M. (2004) Biochim Biophys Acta 1694, 37-53
Yi, L., Celebi, N., Chen, M., and Dalbey, R. E. (2004) J Biol Chem 279, 3926039267
van Bloois, E., Jan Haan, G., de Gier, J. W., Oudega, B., and Luirink, J. (2004)
FEBS Lett 576, 97-100
van der Laan, M., Bechtluft, P., Kol, S., Nouwen, N., and Driessen, A. J. (2004) J
Cell Biol 165, 213-222
Herskovits, A. A., Shimoni, E., Minsky, A., and Bibi, E. (2002) Journal of Cell
Biology 159, 403-410
Ruiz, N., and Silhavy, T. J. (2005) Curr Opin Microbiol 8, 122-126
23
50.
51.
52.
53.
54.
55.
56.
57.
58.
59.
60.
61.
62.
63.
64.
65.
66.
67.
68.
69.
70.
71.
72.
Mogk, A., Schlieker, C., Friedrich, K. L., Schonfeld, H. J., Vierling, E., and
Bukau, B. (2003) J Biol Chem 278, 31033-31042
Darwin, A. J. (2005) Mol Microbiol 57, 621-628
Gittins, J. R., Phoenix, D. A., and Pratt, J. M. (1994) FEMS Microbiol Rev 13, 112
Wu, C. C., and Yates, J. R. (2003) Nature Biotechnology 21, 262-267
Ito, K., and Akiyama, Y. (2005) Annu Rev Microbiol 59, 211-231
van Bloois, E., Haan, G. J., de Gier, J. W., Oudega, B., and Luirink, J. (2006) J
Biol Chem 281, 10002-10009
du Plessis, D. J., Nouwen, N., and Driessen, A. J. (2006) J Biol Chem 281, 1224812252
Celebi, N., Yi, L., Facey, S. J., Kuhn, A., and Dalbey, R. E. (2006) J Mol Biol
357, 1428-1436
Hessa, T., Kim, H., Bihlmaier, K., Lundin, C., Boekel, J., Andersson, H., Nilsson,
I., White, S. H., and von Heijne, G. (2005) Nature 433, 377-381
Andersson, H., and von Heijne, G. (1993) Embo J 12, 683-691
Fernandez-Escamilla, A. M., Rousseau, F., Schymkowitz, J., and Serrano, L.
(2004) Nat Biotechnol 22, 1302-1306
Carrio, M. M., and Villaverde, A. (2003) FEBS Lett 537, 215-221
Carrio, M. M., and Villaverde, A. (2005) J Bacteriol 187, 3599-3601
Weibezahn, J., Tessarz, P., Schlieker, C., Zahn, R., Maglica, Z., Lee, S., Zentgraf,
H., Weber-Ban, E. U., Dougan, D. A., Tsai, F. T., Mogk, A., and Bukau, B.
(2004) Cell 119, 653-665
Withey, J. H., and Friedman, D. I. (2003) Annu Rev Microbiol 57, 101-123
Choy, J. S., Aung, L. L., and Karzai, A. W. (2007) J Bacteriol
Ruiz, N., Kahne, D., and Silhavy, T. J. (2006) Nat Rev Microbiol 4, 57-66
Kang, S. W., Rane, N. S., Kim, S. J., Garrison, J. L., Taunton, J., and Hegde, R. S.
(2006) Cell 127, 999-1013
Qi, H. Y., Hyndman, J. B., and Bernstein, H. D. (2002) Journal of Biological
Chemistry 277, 51077-51083
Cristobal, S., Scotti, P., Luirink, J., von Heijne, G., and de Gier, J. W. L. (1999) J
Biol Chem 274, 20068-20070
Alami, M., Dalal, K., Lelj-Garolla, B., Sligar, S. G., and Duong, F. (2007) Embo J
26, 1995-2004
Chen, Y., Tai, P. C., and Sui, S. F. (2007) J Struct Biol
Facey, S. J., and Kuhn, A. (2003) European Journal of Biochemistry 270, 17241734
24
FIGURE LEGENDS
Figure 1. Effect of SecE depletion on growth, steady state levels of Sec-components,
model inner membrane proteins, and OmpA translocation. A. Effect of SecE
depletion on cell growth. Growth of CM124 cultured in the presence (control) and
absence (SecE depletion) of 0.2% arabinose was monitored by measuring the OD600. B.
Quantification of the steady-state levels of SecE, SecY, SecG, SecA, SecD, SecF, YidC
as well as the model substrates FtsQ, Lep, Fob, and Foc in the inner membrane of SecE
depleted and control cells. Inner membranes from SecE depleted and control cells were
subjected to SDS-PAGE followed by immunoblot analysis with antibodies to the
components listed above. The bar diagram indicates the average fold-change of the
intensities of the bands upon SecE depletion compared to the control. The quantification
is based on three independent samples. C. Analysis of the integrity and abundance of the
SecYEG-complex. Inner membranes from SecE depleted and control cells were subjected
to Blue-Native PAGE analysis followed by detection of SecE, SecY, and SecG by
immunoblotting. The SecYEG trimer is indicated by < and the putative SecYG complex
is indicated by <<. D. Effect of the depletion of SecE on the translocation of the major
outer membrane protein OmpA. SecE depleted and control cells were labeled with
[35S]methionine for 30 seconds and after adding cold methionine, chased for 3 and 10
minutes. OmpA was immunoprecipitated, subjected to standard SDS-PAGE analysis and
labeled material was detected by phosphorimaging. The bars in the diagram indicate the
percentage of the precursor and mature form of OmpA detected in the SecE depleted cells
as compared to the mature OmpA detected in the control cells.
Figure 2. Flow cytometric properties of control and SecE depleted cells. SecE
depleted and control cells were analysed by flow cytometry. A. Size of the population
(forward scatter, FSC) plotted versus granularity (side scatter, SSC) for SecE depleted
and control cells. Insets show microscopy pictures of a representative cell for the SecE
depleted and control cultures. Cell length is indicated with scale bars. B. Histograms
representing the fluorescence of cultures stained with the membrane-specific fluorophore
FM4-64.
25
Figure 3. Analysis of whole cell lysates of SecE depleted and control cells by 2DE
and immunoblotting. A. Comparative 2DE analysis of total lysates from SecE depleted
and control cells. Proteins from 1 OD600 unit of solubilized cells were separated by 2DE.
Proteins were visualized by silver staining and differences between SecE depleted and
control cells were analyzed using PDQuest. 28 spots were significantly (P<0.05) affected
by SecE depletion. Proteins were identified by MALDI-TOF MS PMF from spots
excised from gels stained with Coomassie (Table 1, Supplemental table 1). If several
proteins were identified in the same spot, the first gene name listed corresponds to the
one with the highest Mascot MOWSE score. Primary gene names were taken from
SwissProt (www.expasy.org). Annotated spots were matched onto the silver stained gels
shown using PDQuest. B. Bar graph showing the fold-changes of spots that are
significantly (P<0.05) affected by SecE depletion. Fold-changes were calculated as the
average spot intensities in SecE depleted samples/average spot intensities in control
samples. C. Quantification of the precursor (p) and mature (m) forms of secretory
proteins in SecE depleted and control cells by immunoblotting. Whole cells were
subjected to SDS-PAGE followed by immunoblot analysis with antibodies to two
periplasmic proteins (DegP and Skp) and three outer membrane proteins (OmpA, OmpF,
and PhoE). The bars in the diagram indicate the percentage of the precursor and mature
form of the proteins detected in the SecE depleted cells as compared to the mature form
detected in the control cells. The quantification is based on three independent samples. D.
Quantification of the levels of IbpA/B, SecA, Ffh, SecB, and PspA in whole cells. SecE
depleted and control cells were subjected to SDS-PAGE followed by immunoblot
analysis with antibodies to the components listed above. The bar graph shows the foldchanges calculated as the average band intensities detected in SecE depleted cells/average
band intensities detected in control cells. The quantification is based on three independent
samples.
Figure 4. Characterization of aggregates isolated from SecE depleted cells.
Aggregates isolated from SecE depleted and control cells were analyzed by SDS-PAGE.
Proteins were stained with colloidal Coomassie, and subsequently identified by MALDITOF MS PMF (Table 2, Supplemental table 2). If several proteins were identified in the
same band, the first gene name listed corresponds to the protein with the highest Mascot
MOWSE score.
26
Figure 5. 2DE analysis of the outer membrane proteome from SecE depleted and
control cells. Cells were labeled with [35S]methionine for 1 minute followed by a chase of
10 minutes with cold methionine. The outer membrane fractions were isolated by density
centrifugation from a mixture of labeled and non-labeled cells as out-lined in
Supplemental figure 1 (see ‘Experimental procedures’ for details). The outer membrane
fractions were used for separation by 2DE. Proteins were identified by MALDI-TOF MS
PMF from spots excised from with Coomassie stained gels (Table 3, Supplemental table
3). The outer membrane fraction of SecE depleted cells was contaminated with aggregates
that co-sediment with the outer membrane during density centrifugation (Supplemental
figure 2). The spots corresponding to the proteins in these aggregates were identified and
removed from the analysis set as described in the ‘Image analysis’ section of the
‘Experimental procedures’. Differences in the outer membrane proteomes of the SecE
depleted and control cells were analyzed using PDQuest. Significantly affected (P<0.05)
proteins are indicated by bold numbers in Table 3 and Supplemental table 3. A.
Representative 2D gels showing proteins in the outer membrane fraction stained with
colloidal Coomassie (protein steady-state levels) B. Representative 2D gels showing
proteins in the outer membrane fraction detected by phosphorimaging (protein insertion).
C. Bar graph showing the fold-changes (average spot intensity from SecE depleted
samples/average spot intensity of control sample) of proteins visualized by Coomassie
(black) and phosphorimaging (grey). A fold-change of 100 indicates that a spot was only
detected in SecE depleted samples, a fold-change of 0.01 indicates that a spot was only
detected in the control samples. Numbers refer to spot positions on the gels in Figure 5A
and B.
Figure 6. 2D BN-PAGE analysis of the inner membrane proteome from SecE
depleted and control cells. Cells were labeled with [35S]methionine for 1 minute
followed by a chase of 10 minutes with cold methionine. The inner membrane fractions
were isolated by density centrifugation from a mixture of labeled and non-labeled cells as
outlined in Supplemental figure 1 (see ‘Experimental procedures’ for details). The inner
membrane fractions were analysed by 2D BN-PAGE. Proteins were identified by
MALDI-TOF MS PMF (Table 4, Supplemental table 4) from spots excised from
Coomassie stained gels. If several proteins were identified in one spot, the first gene name
listed corresponds to the protein with the highest Mascot MOWSE score. Primary gene
names were taken from SwissProt (www.expasy.org). Differences in the inner membrane
27
proteomes of SecE depleted and control cells were analyzed using PDQuest. Significantly
affected (P<0.05) proteins are indicated by bold numbers in Table 4 and Supplemental
table 4. A. Representative 2D BN-PAGE gels with proteins detected by staining with
colloidal Coomassie (protein steady-state levels). B. Representative 2D BN-PAGE gels
with proteins detected by phosphorimaging (protein insertion). C. Bar graph showing the
fold changes (average spot intensities from SecE depleted samples/average spot
intensities of control samples) of proteins detected by Coomassie staining (black) and
phosphorimaging (grey). A fold-change of 100 indicates a spot that was only detected in
SecE depleted samples, a fold-change of 0.01 indicates that it was only detected in the
control samples. Numbers refer to spot positions on the gels in Figure 6A and B.
Figure 7. Correlations between properties of inner membrane proteins and the
effect on their steady-state levels and insertion upon SecE depletion. A. Fold-changes
of steady-state levels (Coomassie) and insertion (phosporimaging) plotted against the
number of amino acids in the largest translocated domain of each protein (Supplemental
table 5). Almost all proteins that were positively affected by SecE depletion do not
contain any large periplasmic domains. Exceptions to this trend are indicated in the plots
with their gene names and number of predicted transmembrane segments. B. Proteins
were divided into two groups; proteins with large translocated domains (≥60 amino acids)
and proteins with small translocated domains (≤60 amino acids). The fold-changes upon
SecE depletion on these two groups were plotted against the number of transmembrane
segments (Figure 7B, Supplemental table 5).
28
Table 1. Comparative 2D-gel analysis and mass spectrometry identification of proteins from total lysates of SecE
depleted and control cells.
Spots visualized by silver staining (Figure 3A) were quantified and compared using PDQuest. Spot quantities were
normalized using the 'total quantity of valid spot' method. Values of 0.01 and 100 correspond to spots that are missing or
turned on in the SecE depletion, respectively. All spots that were significantly (P<0.05) changed upon SecE depletion
were excised from gels stained with colloidal Coomassie and proteins were identified by MALDI-TOF MS PMF.
(a)
(b)
(c)
(d)
predicted
MW (kDa)
(precursor
/mature)
(e)
(f)
(g)
(h)
(i)
1
ybbN
Protein ybbN
c
31.8
4.5
31.8
4.5
2.23
spot Nr. gene name
protein name
local.
predicted pI
(precursor
/mature)
observed
MW (kDa)
fold change
observed pI (SecE depl.
/control)
2
groL
60 kDa chaperonin
c
57.3
4.85
55.3
4.68
100
3
dnaK
Chaperone protein dnaK
c
69.1
4.83
68.9
4.83
1.74
4
groL
60 kDa chaperonin
c
57.3
4.85
60.2
4.83
1.96
5
potD
Spermidine/putrescine-binding periplasmic protein
p
38.9/36.5
5.24/4.86
35.87
4.76
0.01
6
metQ
D-methionine-binding lipoprotein metQ
im/om lp
38.9/36.5
5.24/4.86
27.25
4.73
2.96
6
rpsB
30S ribosomal protein S2
c
26.7
6.61
27.25
4.73
2.96
6
grpE
Protein grpE
c
21.8
4.68
27.25
4.73
2.96
7
metQ
D-methionine-binding lipoprotein metQ
im/om lp
38.9/36.5
5.24/4.86
27.4
4.88
4.32
8
no id.
27.98
5.11
100
9
fliY
Cystine-binding periplasmic protein
p
29.3/26.1
6.22/5.29
25.58
5.18
0.13
10
livJ
Leu/Ile/Val-binding protein
p
39.1/36.8
5.54/5.28
38.7
5.22
0.20
11
hdhA
7-alpha-hydroxysteroid dehydrogenase
c
26.8
5.22
23.39
5.21
0.37
12
clpB
Chaperone clpB
c
95.6
5.37
79.12
5.29
2.96
13
ushA
Protein ushA
p
60.8/58.2
5.47/5.4
59.65
5.31
0.20
14
cysK
Cysteine synthase A
amb
34.5
5.83
35.47
5.34
0.01
14
ompT
Protease 7
om
35.6/33.5
5.76/5.38
35.47
5.34
0.01
14
trxB
Thioredoxin reductase
c
34.6
5.3
35.47
5.34
0.01
15
no id.
26.48
5.35
7.68
16
bla
Beta-lactamase TEM
p
31.5/28.9
5.69/5.46
29.8
5.39
0.46
17
no id.
24.4
5.39
12.90
18
znuA
High-affinity zinc uptake system protein znuA
p
33.8/31.1
5.61/5.44
33.29
5.52
0.07
19
yehZ
Hypothetical protein yehZ
p
32.6/30.2
5.82/5.56
31.19
5.48
0.01
20
ybiS
Protein ybiS
p
33.42/30.86
5.99/5.6
31.19
5.57
0.34
20
yggE
Hypothetical protein yggE
p
26.6/24.5
6.1/5.60
31.19
5.57
0.34
21
yodA
Metal-binding protein yodA
p
24.8/22.3
5.91/5.66
23.15
5.6
0.01
22
dppA
Periplasmic dipeptide transport protein
p
60.3/57.4
6.21/5.75
55.21
5.74
0.36
23
ompC
Outer membrane protein C
om
40.4/38.3
4.58/4.48
32.71
5.76
100
24
oppA
Periplasmic oligopeptide-binding protein
p
60.97/58.5
6.05/5.85
54.94
6.02
0.13
0.23
25
tolB
Protein tolB
p
46.0/43.6
6.98/6.14
41.31
6.09
26
ompA
Outer membrane protein A
om
37.2/35.2
5.99/5.60
37.3
6.07
100
27
rbsB
D-ribose-binding periplasmic protein
p
31.0/28.5
6.85/5.99
30.67
6.03
100
28
rbsB
D-ribose-binding periplasmic protein
p
31.0/28.5
6.85/5.99
28.46
6.02
0.17
(a)
The numbering corresponds to the spots in the 2D-gel images in Figure 3A.
(b)
Gene name from the Swiss Prot database for E. coli . 'no id.' indicates that no protein was identified in the spot.
(c)
Protein name from the Swiss Prot database for E. coli . 'no id.' indicates that no protein was identified in the spot.
(d)
Localization based on the information given in the Swiss Prot for E. coli . Unknown localizations were predicted by PSORT. For integral membrane
proteins, the number of transmembrane segments are indicated. Abbreviations: amb., ambiguous localization; c., cytoplasmic; im lp., inner
membrane lipoprotein; local., localization; om., outer membrane; om lp., outer membrane lipoprotein; p., periplasmic.
(e)
Protein sizes (in kDa) predicted from amino acid sequences. Two sizes are given for secretory proteins, the first size corresponds to the precursor
form, and the second size corresponds to the mature form of the protein.
(f)
pI predicted from amino acid sequence. Two values are given for secretory proteins, the first value corresponds to the precursor form, and the
second value corresponds to the mature form of the protein.
(g)
Size of proteins calculated from the spot position on the 2D-gels used for the analysis.
(h)
pI of proteins calculated from the spot position on the 2D-gels used for the analysis.
(i)
Fold-change, i.e., the ratio of the average intensity of significantly (P<0.05) affected spots in the gels of the SecE depletion to the average intensity of
matched spots in the control gels.
29
Table 2. Mass spectrometry identification of proteins in aggregates isolated from lysates of SecE depleted and control
cells.
Protein aggregates were extracted from lysates of SecE depleted and control cells. The protein content of the aggregates was
analysed by 1DE followed by MALDI-TOF MS PMF (Figure 4, Supplemental table 2) or nanoLC-MS/MS analysis of
solubilized aggregates digested with trypsin (Supplemental table 2).
band nr.
hit rank
gene name
protein name
local./ TMs
predicted MW
(kDa) (precursor
/mature)
(a)
(b)
(c)
(d)
(e)
(f)
2
adhE
Aldehyde-alcohol dehydrogenase
c
96.1
17
atpA
ATP synthase subunit alpha
c, im ass
55.4
bla
Beta-lactamase TEM
p
31.6/28.9
22, 30, 33
1
6
38
25
clpB
Chaperone clpB
c
95.7
crp
Catabolite gene activator
c
23.6
28
cysK
Cysteine synthase A
amb
34.4
14
cysN
Sulfate adenylyltransferase subunit 1
c
52.6
25
dacA
Penicillin-binding protein 5
1 *
40.3/38.3
23
dacC
Penicillin-binding protein 6
1 *
43.6/40.8
19
degP
Protease do
p
49.4/46.8
23
dnaJ
Chaperone protein dnaJ
c
41.1
dnaK
Chaperone protein dnaK
c
69.1
13
dps
DNA protection during starvation protein
c
18.6
14
elaB
Protein elaB
1
11.3
4, 5
6
fusA
Elongation factor G
c
77.6
18
12
glgA
Glycogen synthase
c
53.0
7, 36
17
84.3
9
42
glgB
1,4-alpha-glucan branching enzyme
c
11
glmS
Glucosamine--fructose-6-phosphate aminotransferase
c
66.9
16
guaB
Inosine-5'-monophosphate dehydrogenase
c
52.0
66.0
11, 36
46, 47
htpG
Chaperone protein htpG
c
21
hupA
DNA-binding protein HU-alpha
c
9.5
16
ibpA
Small heat shock protein ibpA
c
15.8
23
iscS
Cysteine desulfurase
c
45.2
lpp
Major outer membrane lipoprotein
om lp
8.3/6.4
lysS
Lysyl-tRNA synthetase
c
57.6
metE
5-methyltetrahydropteroyltriglutamate--homocysteine
methyltransferase
c
84.7
26
mreB
Rod shape-determining protein mreB
c
37.1
29
nlpD
Lipoprotein nlpD
im lp
40.2/37.5
49
9
12
4, 5
15
27, 33, 34
3
ompA
Outer membrane protein A
om
37.2/35.2
24, 25
5
ompC
Outer membrane protein C
om
40.3/38.3
23, 24
18
ompF
Outer membrane protein F
om
39.3/37.1
ompT
Protease 7
om
35.5/33.5
24
41
ompX
Outer membrane protein X
om
18.7/16.2
13
ptsI
Phosphoenolpyruvate-protein phosphotransferase
c
63.6
35
rplC
50S ribosomal protein L3
c
22.2
37
rplD
50S ribosomal protein L4
c
22.1
rplE
50S ribosomal protein L5
c
20.5
39
rplF
50S ribosomal protein L6
c
18.9
44
rplI
50S ribosomal protein L9
c
15.8
40
8
24
1
rpoC
DNA-directed RNA polymerase beta' chain
c
155.1
10, 31
rpsA
30S ribosomal protein S1
c
61.2
32
rpsB
30S ribosomal protein S2
c
26.7
19, 21
rpsC
30S ribosomal protein S3
c
25.8
43
rpsE
30S ribosomal protein S5
c
17.5
40
rpsG
30S ribosomal protein S7
c
20.0
30
48
26
3
rpsJ
30S ribosomal protein S10
c
11.7
secA
Preprotein translocase subunit secA
c
102.0
17.7/15.7
43, 44
10
skp
Chaperone protein skp
p
45
4
slyB
Outer membrane lipoprotein slyB
om lp
15.6/13.8
spb
Sulfate-binding protein
p
18.7/16.2
25
20
7
15
2
p
36.6/34.7
om
54.0/51.5
tufA
Elongation factor Tu
c
43.3
GTP-binding protein typA/BipA
c
67.4
yajG
Uncharacterized lipoprotein yajG
om lp
21.0/19.0
yfiO
Lipoprotein yfiO
om lp
23.9/21.7
11
ygiW
Protein ygiW
p
14.0/12.0
27
yhjK
Protein yhjK
2
73.1
28
yncE
Hypothetical protein yncE
sec
38.6/35.3
yqjD
Hypothetical protein yqjD
1
11.1
yrbC
Protein yrbC
sec
23.9/21.7
20
35
22
35
Protein tolB
Outer membrane protein tolC
typA
8
25
tolB
tolC
(a)
(b)
(c)
The numbering corresponds to the bands in the 1D-gel shown in Figure 4.
The ranking is based on the Mascot MOWSE score of proteins identified by in-solution digestion/ nanoLC-ESI-MS/MS.
Protein name extracted from the Swiss Prot database for E. coli .
(d)
Gene name extracted from the Swiss Prot database for E. coli .
(e)
Localization based on the information given in the Swiss Prot database for E. coli . Unknown localizations were predicted by PSORT. For integral
membrane proteins, the number of transmembrane segments are indicated. '*' indicates that the membrane inserted segment may work as an
anchor rather than a true transmembrane segment. Abbreviations: amb., ambiguous localization; c., cytoplasmic; im ass., inner membrane
associated; im lp., inner membrane lipoprotein; om., outer membrane; om lp., outer membrane lipoprotein; p., periplasmic; sec., secretory; TMs.,
trans-membrane segments.
(f)
Protein sizes (in kDa) predicted from amino acid sequence. Two sizes are given for secretory proteins, the first size corresponds to the precursor
form, and the second size corresponds to the mature form of the protein.
31
Table 3. Comparative 2D-gel analysis and mass spectrometry identification of proteins in the outer membrane
fraction of SecE depleted and control cells.
The spots in 2D-gels of the outer membrane fraction from SecE depleted and control cells were excised from gels
stained with colloidal Coomassie (Figure 5A). Proteins were identified by MALDI-TOF MS PMF. Spots visualized by
Coomassie staining and phosphorimaging (Figure 5A and B) were quantified and compared using PDQuest. Quantities
of Coomassie stained spots were normalized using the 'total quantity of valid spot' tool, excluding spots detected in the
2D gels of protein aggregates extracted from the outer membrane fraction (Supplemental figure 1). Quantities of spots
visualized by phosphorimaging were normalized using the normalization factor calculated for the corresponding
Coomassie stained gel. Values of 0.01 and 100 correspond to spots that are missing or turned on in the SecE depletion,
respectively. Significant fold-changes (P<0.05) are indicated with bold numbers. Graphs of fold-changes are shown in
Figure 5C.
spot nr.
gene
name
local.
/TMs
protein name
predicted
predicted pI observed
MW (kDa)
(precursor
MW
(precursor
/mature)
(kDa)
/mature)
coomassie phosphor.
fold
fold
observed
change
change
pI
(SecE
(SecE
depl.
depl.
/control)
/control)
(h)
(i)
(i)
(a)
(b)
(c)
(d)
(e)
(f)
(g)
1
mdtE
Multidrug resistance protein mdtE
im lp
41.3/38.9
5.73/5.12
53.22
4.14
1.06
n.d.
2
dcrB
Protein dcrB
amb
19.8/17.8
5.09/4.91
17.7
4.29
0.22
0.70
3
yfgL
Lipoprotein YfgL
om lp
41.9/39.9
4.72/4.61
37.75
4.35
0.83
0.76
4
fhuE
FhuE receptor
om
81.2/77.4
4.75/4.72
77.17
4.65
0.25
0.56
5
hemX
Putative uroporphyrinogen-III C-methyltransferase
1
42.9
4.68
45.08
4.59
1.93
1.93
0.01
7
imp
LPS-assembly protein
om
89.7/87.1
4.94/4.85
91.39
4.87
0.50
8
yaeT
Outer membrane protein assembly factor yaeT
om
90.6/88.4
4.93/4.87
88.4
4.87
0.33
0.62
12
fadL
Long-chain fatty acid transport protein
om
48.5/45.9
5.09/4.91
44.86
4.84
1.03
1.17
1.01
14
nlpB
Lipoprotein 34
om lp
36.9/34.4
5.34/4.96
34.68
4.74
0.52
15
metQ
D-methionine-binding lipoprotein metQ
om/im lp
29.5/27.2
5.13/4.93
26.94
4.76
0.07
n.d.
16
tsx
Nucleoside-specific channel-forming protein tsx
om
33.6/31.4
5.07/4.87
27.47
4.86
0.61
1.25
n.d.
17
ybaY
Hypothetical lipoprotein ybaY
om/im lp
19.5/17.7
7.88/6.31
23.21
4.78
0.44
19
mdtE
Multidrug resistance protein mdtE
im lp
41.2/38.9
5.73/5.12
40.1
4.97
0.31
n.d.
20
tsx
Nucleoside-specific channel-forming protein tsx
om
33.6/31.4
5.07/4.87
27.54
4.96
0.47
0.50
21
mipA
MltA-interacting protein
om
27.8/25.7
5.50/5.03
25.74
4.95
0.26
0.64
23
ompX
Outer membrane protein X
om
18.6/16.4
6.56/5.3
16.12
4.91
0.49
1.13
25
cirA
Colicin I receptor
om
74.1/71.2
5.11/5.03
75.78
5.05
0.11
0.77
26
yiaF
Hypothetical protein yiaF
amb
30.43
9.35
23.57
5.01
0.21
0.91
27
fhuA
Ferrichrome-iron receptor
om
82.4/78.7
5.47/5.13
79.29
5.16
0.32
0.77
28
btuB
Vitamin B12 transporter btuB
om
68.4/66.3
5.23/5.10
66.13
5.13
0.48
1.02
30
nlpA
Lipoprotein 28
im lp
29.4/27.1
5.77/5.29
26.64
5.12
0.23
0.01
31
mipA
MltA-interacting protein
om
27.8/25.7
5.50/5.03
25.52
5.11
0.49
1.02
33
pal
Peptidoglycan-associated lipoprotein
om lp
16.9/16.6
6.29/5.59
17.65
5.08
1.02
1.14
36
fepA
Ferrienterobactin receptor
om
82.1/79.8
5.39/5.23
81.97
5.23
0.07
0.43
37
yfiO
Lipoprotein yfiO
om lp
27.9/25.8
6.16/5.48
24.87
5.2
0.59
0.62
38
ompX
Outer membrane protein X
om
18.6/16.4
6.56/5.3
15.7
5.25
0.64
0.72
39
pal
Peptidoglycan-associated lipoprotein
om lp
16.9/16.6
6.29/5.59
17.43
5.38
1.19
1.12
44
yfiO
Lipoprotein yfiO
om lp
27.9/25.8
6.16/5.48
25.39
5.53
0.69
n.d.
45
ompW
Outer membrane protein W
om
22.9/20.9
6.03/5.58
21.5
5.54
0.31
0.10
48
fusA
Elongation factor G
c
77.6
5.24
83.22
5.69
0.09
n.d.
49
yicH
Hypothetical protein yicH
sec
62.3/58.7
5.67/5.38
63.8
5.85
0.15
n.d.
50
ompA
Outer membrane protein A
om
37.2/35.2
5.99/5.60
31.74
6
1.22
1.12
53
ompX
Outer membrane protein X
om
18.6/16.4
6.56/5.3
16.49
6.19
0.37
0.46
(a)
The numbers corresponds to the spot numbers in the 2D-gels shown in Figure 5A and B.
(b)
Gene name extracted from the Swiss Prot database for E. coli .
(c)
Protein name from the Swiss Prot database for E. coli . 'no id.' indicates that no protein was identified in the spot.
(d)
Localization based on the information given in the Swiss Prot data base for E.coli . Unknown localizations were predicted by PSORT
(http://psort.hgc.jp/form.html). Abbreviations: amb., ambiguous localization; c., cytoplasmic; im lp., inner membrane lipoprotein; local., localization;
om., outer membrane; om lp., outer membrane lipoprotein; sec., secretory; TMs., transmembrane segments.
32
(e)
Protein sizes (in kDa) predicted from amino acid sequences. Two sizes are given for secretory proteins, the first size corresponds to the precursor
form including the signal sequence, and the second size corresponds to the mature form of the protein after signal sequence processing.
(f)
(g)
pI predicted from amino acid sequence. Two values are given for secretory proteins, the first value corresponds to the precursor form, and the
second value corresponds to the mature form of the protein.
Size of proteins calculated from the spot position on the 2D-gels shown in Figure 5.
(h)
pI of proteins calculated from the spot position on the 2D-gels shown in Figure 5.
(i)
Fold-change, i.e., the ratio of the average spot intensity. Fold-change of significantly (P<0.05) affected spots are indicated by bold numbers. 'n.d.'
indicates that the spot was not detected. Abbreviations: phosphor., phosphorimaging
33
Table 4. 2D Blue Native PAGE analysis of the inner membrane proteome of SecE depleted and control cells.
Spots in 2D BN PAGE gels of inner membranes (Figure 6A) were excised from gels stained with colloidal Coomassie.
Proteins were identified by MALDI-TOF MS PMF. Proteins belonging to the same complex have the same gray fill color.
Spots visualized by Coomassie staining and/or phosphorimaging (Figure 6A and B) were quantified and compared using
PDQuest. 0.01 and 100 correspond to spots that are missing or turned on in the SecE depletion, respectively. Significant foldchanges (P<0.05) are indicated with bold numbers.
spot nr.
gene
name
protein name
local. /TMs
predicted
MW (kDa)
(precursor
/mature)
observed
MW (kDa)
observed
native MW
(kDa)
Coomassie
fold change
(SecE dep
/control)
phosphor.
fold change
(SecE depl
/control)
(a)
(b)
(c)
(d)
(e)
(f)
(g)
(h)
(h)
1
yjeP
Hypothetical mscS family protein yjeP
10
124.0
103.7
1000
0.52
n.d.
2
kefA
Potassium efflux system kefA
11
127.2
100.7
1000
0.23
n.d.
3
ftsH
Cell division protease ftsH
2
70.7
69.3
1000
0.24
0.56
4
hflK
Protein hflK
1
45.5
45.4
1000
0.31
n.d.
5
hflC
Protein hflC
1
37.7
37.4
1000
0.43
n.d.
6
ybbK
Hypothetical protein ybbK
1
33.7
32.2
1000
1.36
n.d.
6
corA
Magnesium transport protein corA
2
36.6
32.2
1000
1.36
n.d.
7
nuoC
NADH-quinone oxidoreductase subunit C/D
c, im ass
68.7
67.1
966
0.70
n.d.
8
creD
Inner membrane protein creD
6
49.8
42.7
733
0.20
n.d.
8
hemY
Protein hemY
2
45.2
42.7
733
0.20
n.d.
9
groL
60 kDa chaperonin
c
57.2
64.0
828
3.72
1.30
10
atpA
ATP synthase subunit alpha
c, im ass
55.2
55.4
638
1.36
0.91
11
atpD
ATP synthase subunit beta
c, im ass
50.2
53.0
637
1.50
1.44
12
atpG
ATP synthase gamma chain
c, im ass
31.6
31.4
614
1.19
1.49
15
no id.
22.5
609
2.02
0.93
13
atpH
ATP synthase delta chain
c, im ass
19.3
20.8
606
1.11
n.d.
14
atpF
ATP synthase B chain
1
17.3
18.9
599
1.47
n.d.
16
fadE
Acyl-coenzyme A dehydrogenase
2
89.2
85.1
600
1.28
n.d.
16
plsB
Glycerol-3-phosphate acyltransferase
c, im ass
91.3
85.1
600
1.28
n.d.
17
wzzE
Lipopolysaccharide biosynthesis protein wzzE
2
39.6
38.6
603
0.90
0.72
17
ybdG
Hypothetical protein ybdG
1
36.4
38.6
603
0.90
0.72
18
nuoC
NADH-quinone oxidoreductase subunit C/D
c, im ass
68.7
68.1
543
0.86
n.d.
19
nuoB
NADH-quinone oxidoreductase subunit B
c, im ass
25.1
25.5
510
0.59
n.d.
20
nuoI
NADH-quinone oxidoreductase subunit I
c, im ass
20.5
23.5
509
0.84
n.d.
21
wzzB
Chain length determinant protein
2
36.5
35.1
531
0.55
0.07
22
narG
Respiratory nitrate reductase 1 alpha chain
c, im ass
140.4
113.1
463
1.92
n.d.
23
narI #
Respiratory nitrate reductase 1 gamma chain
5
25.5
21.7
437
2.58
0.79
24
acrB
lavine resistance protein B
12
113.6
92.7
478
0.23
0.01
25
sdhA
Succinate dehydrogenase flavoprotein subunit
c, im ass
64.4
68.6
440
0.92
n.d.
26
sdhD
Succinate dehydrogenase hydrophobic membrane anchor
subunit
4
12.9
12.0
427
0.94
n.d.
27
sdhB
Succinate dehydrogenase iron-sulfur subunit
c, im ass
26.8
27.2
438
1.44
1.14
27
manZ
Mannose permease IID component
1
31.3
27.2
438
1.44
1.14
28
manX
PTS system mannose-specific EIIAB component
c, im ass
34.9
36.8
439
2.40
1.26
29
no id.
30.5
436
1.02
0.78
34
30
no id.
31
yrbD
32
no id.
33
nuoC
NADH-quinone oxidoreductase subunit C/D
c, im ass
34
atpA
ATP synthase subunit alpha
c, im ass
Hypothetical protein yrbD
sec
13.7
434
1.00
n.d
87.5
426
0.36
n.d.
30.5
406
0.52
0.47
68.7
69.3
363
0.81
n.d.
55.2
56.4
366
1.02
0.83
19.6/16.5
35
atpD
ATP synthase subunit beta
c, im ass
50.2
54.0
364
1.33
1.26
36
atpG
ATP synthase gamma chain
c, im ass
31.6
32.2
348
1.68
1.73
37
atpF
ATP synthase B chain
1
17.3
19.5
301
1.27
n.d.
38
no id.
88.8
327
1.10
n.d.
39
mscS
24.4
302
0.77
0.94
40
no id.
93.9
228
0.61
n.d.
Small-conductance mechanosensitive channel
3
30.9
41
aas
AAS bifunctional protein
2
80.7
78.7
248
4.38
1.59
41
yhjG
Hypothetical protein yhjG
2
75.1
78.7
248
4.38
1.59
42
ppiD
Peptidyl-prolyl cis-trans isomerase D
1
68.2
71.6
264
0.22
0.01
43
msbA
Lipid A export ATP-binding/permease protein msbA
5
64.5
59.7
238
0.64
0.68
44
hemX
Putative uroporphyrinogen-III C-methyltransferase
1
43.0
49.0
265
0.84
0.83
45
ydjN
Hypothetical symporter ydjN
9
48.7
35.1
272
0.98
0.69
46
exbB
Biopolymer transport exbB protein
3
26.3
24.7
243
0.31
0.46
47
no id.
17.4
231
0.01
0.01
48
secA
Preprotein translocase subunit secA
c, im ass
102.0
108.1
209
50.77
100.00
49
cyoB
Ubiquinol oxidase subunit 1
15
74.4
50.7
203
0.15
0.81
50
cyoA
Ubiquinol oxidase subunit 2
2
34.9
32.6
201
0.25
0.79
51
yhbG
Probable ABC transporter ATP-binding protein yhbG
c, im ass
26.7
27.5
201
0.35
n.d.
52
ygiM
Hypothetical protein ygiM
1
23.1
22.6
218
0.31
n.d.
52
tolQ
Protein tolQ
3
25.6
22.6
218
0.31
n.d.
53
atpF
ATP synthase B chain
1
17.3
20.1
200
0.11
0.88
54
ppiD
Peptidyl-prolyl cis-trans isomerase D
1
68.2
71.4
176
0.51
0.59
55
no id.
49.4
159
100.00
1.84
56
no id.
49.2
160
0.50
1.70
57
no id.
58
dacC
58
mdtE
Multidrug resistance protein mdtE
im, lp
41.2
42.6
158
0.87
0.81
58
cysA
Sulfate/thiosulfate import ATP-binding protein cysA
c, im ass
41.1
42.6
158
0.87
0.81
Penicillin-binding protein 6
1
43.6/40.8
44.3
159
0.84
0.83
42.6
158
0.87
0.81
59
cysA
Sulfate/thiosulfate import ATP-binding protein cysA
c, im ass
41.1
43.1
177
2.27
1.37
59
dacC
Penicillin-binding protein 6
c, im ass
41.1
43.1
177
2.27
1.37
59
hemY
Protein hemY
2
45.2
43.1
177
60
dacA
Penicillin-binding protein 5
1
44.4/41.3
41.4
2.27
1.37
1.40
n.d.
60
dppF
Dipeptide transport ATP-binding protein dppF
c, im ass
37.6
41.4
160
1.40
n.d.
61
dppD
Dipeptide transport ATP-binding protein dppD
c, im ass
35.8
37.3
160
1.70
n.d.
62
no id.
36.0
193
4.71
n.d.
63
metQ
D-methionine-binding lipoprotein metQ
im/om lp
29.4/27.2
29.0
166
0.27
0.48
63
nlpA
Lipoprotein 28
im/om lp
29.4/27.1
29.0
166
0.27
0.48
64
no id.
25.2
160
14.15
1.96
65
yfgM
Protein yfgM
1
22.2
23.7
165
0.41
0.66
66
nuoG
NADH-quinone oxidoreductase subunit F
c, im ass
100.2
102.5
159
1.30
n.d.
67
nuoF
NADH-quinone oxidoreductase subunit G
c, im ass
49.3
51.1
150
0.95
n.d.
35
68
mgtA
Magnesium-transporting ATPase, P-type 1
10
99.5
98.6
149
1.34
n.d.
69
copA
Copper-transporting P-type ATPase
8
87.7
93.4
158
0.90
1.44
69
mrcA
Penicillin-binding protein 1A
1
93.6
93.4
158
0.90
1.44
70
frdA
Fumarate reductase flavoprotein subunit
c, im ass
65.8
74.2
145
1.24
n.d.
71
secD
Protein-export membrane protein secD
6
66.6
65.5
151
0.47
0.60
72
cydC
Transport ATP-binding protein cydC
6
62.9
57.5
150
1.86
1.79
72
cydd
Transport ATP-binding protein cydD
6
65.1
57.5
150
1.86
1.79
73
cydA
Cytochrome d ubiquinol oxidase subunit 1
7
58.2
48.1
137
0.19
n.d.
74
metN
Methionine import ATP-binding protein metN
c
37.8
40.4
147
1.41
n.d.
75
degS
Protease degS
p
37.6/34.6
37.7
151
0.45
n.d.
76
metQ
D-methionine-binding lipoprotein metQ
im/om lp
29.4/27.2
29.1
135
0.61
0.67
77
rnfG
Electron transport complex protein rnfG
1
21.9
22.6
143
0.10
0.01
78
no id.
21.4
135
1.03
1.14
79
atpF
ATP synthase B chain
1
17.3
20.1
149
2.07
1.25
80
glnP
Glutamine transport system permease protein glnP
5
24.4
19.0
128
0.83
0.92
81
yhcB
Putative cytochrome d ubiquinol oxidase subunit III
1
15.2
14.5
144
2.41
1.52
82
yhcB
Putative cytochrome d ubiquinol oxidase subunit III
1
15.2
14.6
133
1.05
0.90
83
ydiJ
Hypothetical protein ydiJ
amb
113.2
103.2
115
2.27
n.d.
84
plsB
Glycerol-3-phosphate acyltransferase
c, im ass
91.3
94.4
116
0.91
1.31
85
gcd
Quinoprotein glucose dehydrogenase
5
86.7
89.7
117
0.41
0.66
86
ppiD
Peptidyl-prolyl cis-trans isomerase D
1
68.2
73.1
126
0.68
0.99
87
nuoC
NADH-quinone oxidoreductase subunit C/D
c, im ass
68.7
72.2
119
0.96
n.d.
88
oxaA
Inner membrane protein oxaA
6
61.5
60.6
125
0.31
0.42
89
ybhG
Membrane protein ybhG
1
36.4/34.4
37.9
130
0.56
1.17
90
nuoB
NADH-quinone oxidoreductase subunit B
c, im ass
25.1
26.5
119
1.17
n.d.
91
yajC
Membrane protein yajC
1
11.9
11.7
111
0.65
0.51
92
secD
Protein-export membrane protein secD
6
66.6
67.5
106
0.21
0.33
92
yicH
Hypothetical protein yicH
amb
62.3
67.5
106
0.21
0.33
93
no id.
52.8
115
n.d
1.42
94
dadA
D-amino acid dehydrogenase small subunit
c, im ass
47.6
52.2
105
1.04
1.07
94
zipA
Cell division protein zipA
1
36.5
52.2
105
1.04
1.07
95
ndh
NADH dehydrogenase
amb
47.2
49.3
111
1.63
1.36
96
proP
Proline/betaine transporter
12
54.8
41.7
111
0.62
0.85
97
metQ
D-methionine-binding lipoprotein metQ
im/om lp
29.4/27.2
29.1
110
0.96
0.94
98
no id.
96.4
92
0.62
n.d.
99
yijP
Membrane protein yijP
5
66.6
62.9
96
0.45
n.d.
100
ydgA
Protein ydgA
p, im ass
54.7/52.8
59.3
96
0.42
0.66
101
emrA
Multidrug resistance protein A
1
42.7
45.5
89
0.86
0.72
102
dacA
Penicillin-binding protein 5
1
44.4/41.3
43.6
92
0.53
0.49
102
dacC
Penicillin-binding protein 6
1
43.6/40.8
43.6
92
0.53
0.49
103
pyrD
Dihydroorotate dehydrogenase
c, im ass
36.8
39.4
83
0.37
0.10
104
gadC
Probable glutamate/gamma-aminobutyrate antiporter
12
55.1
38.3
98
0.27
0.70
105
glpT
Glycerol-3-phosphate transporter
12
47.2
50.3
87
0.54
0.42
106
ompA
Outer membrane protein A
om
37.2
31.8
87
0.93
0.87
106
metQ
D-methionine-binding lipoprotein metQ
im/om lp
29.4/27.2
31.8
87
0.93
0.87
36
107
pgsA
CDP-diacylglycerol--glycerol-3-phosphate 3phosphatidyltransferase
4
20.6
17.2
86
1.83
1.73
108
dnaK
Chaperone protein dnaK
c
69.0
83.7
77
5.89
1.72
109
dld
D-lactate dehydrogenase
c, im ass
64.5
73.3
71
0.72
1.15
110
cysK
Cysteine synthase A
c
34.4
38.1
63
1.10
0.86
111
lepB
Signal peptidase I
2
36.0
35.2
69
0.50
0.95
112
metQ
D-methionine-binding lipoprotein metQ
im/om lp
29.4/27.2
29.5
70
0.39
n.d.
112
rpsB
30S ribosomal protein S2
c
26.6
29.5
70
0.39
n.d.
113
no id.
27.2
57
12.16
1.73
114
no id.
23.0
63
2.34
0.96
115
no id.
21.5
58
0.30
0.42
116
no id.
19.1
72
0.61
n.d.
117
yajC
11.7
71
1.44
1.43
Membrane protein yajC
1
11.9
(a)
The numbering corresponds to the spots in the 2D BN PAGE gels shown in Figure 6A.
(b)
Gene name from the Swiss Prot database for E. coli . '#' indicates that the criteria for MS identification was not fulfilled, but annotation fits with expected
size of complex and monomer.
Protein name from the Swiss Prot database for E. coli .
(c)
(d)
(e)
Localization based on the information given in the Swiss Prot database for E. coli. Unknown localizations were predicted by PSORT. For integral
membrane proteins, the number of transmembrane segments is indicated. '*' indicates that the membrane inserted segment may work as an anchor
rather than a true transmembrane segment. Abbreviations: amb., ambiguous localization; c., cytoplasmic; im ass., inner membrane associated; im lp.,
inner membrane lipoprotein; local., localization; om., outer membrane; om lp., outer membrane lipoprotein; sec., secretory; TMs., transmembrane
segments.
Protein sizes (in kDa) predicted from amino acid sequences. Two sizes are given for secretory proteins, the first size corresponds to the precursor form,
and the second size corresponds to the mature form of the protein.
(f)
Size of protein calculated from the spot position on the 2D BN PAGE gels used for the analysis.
(g)
Native mass (in kDa) based on the position in the 2D BN PAGE gel in Figure 6A.
(h)
The fold-change, i.e., the ratio of the average intensity of spots in gels of the SecE depleted cells to the average intensity of matched spots in the control
gels. 'n.d.' indicates that the spot was not detected. Abbreviations: phosphor., phosphorimaging.
37
Figure 1
A.
C.
2.5
control
α SecY
α SecG
SecE depletion
2
OD600
α SecE
Native MW
(kDa)
160
1.5
1
<
0.5
0
0
2
4
6
8
10
12
14
<<
time (h)
66
B.
l
de
ro
cE
Se
l
de
ro
cE
nt
co
Se
nt
de
l
pl
pl
pl
α SecY
co
cE
ro
nt
Se
co
α SecE
D.
α SecG
200
precursor
mature
% of control
α SecA
α SecD
α SecF
α YidC
α FtsQ
150
100
50
0
0
α Lep
3
10
chase (min)
α Fo b
pre-OmpA
OmpA
α Fo c
cE
Se
pl
de
pl
pl
de
l
l
ro
E
nt
co
c
Se
de
38
ro
cE
l
ro
nt
co
2.0
nt
1.0
1.5
fold-change
(SecE depl/control)
Se
0.5
co
pl
de
l
ro
cE
Se
nt
co
0
Figure 2
B.
A.
10
3
10
2
10
1
10
0
SecE depl
2.13 μm
10
0
10
1
10
2
Forward Scatter
10
3
10
4
10
4
10
3
10
2
10
1
10
0
100
2.86 μm
10
0
10
% of Maximum
4
Side Scatter
Side Scatter
control
10
1
10
2
Forward Scatter
39
10
3
10
4
control
SecE depl
80
60
40
20
0
10
0
10
1
10
2
FM4-64
10
3
10
4
Figure 3
A.
B.
control
Mw (kDa)
12
100
75
3
4
13
10
5
14
7
9
11
17 21
27
28
Mw (kDa)
12
100
75
3
4
24
25
26
18
19 20
16
1
6
22
SecE depletion
13
22
2
50
24
25
26
37
1
6 7
25
18
20 23
16
8
15
9
17
11
21
2 groL
8 no id.
23 ompC
50
26 ompA
37
17 no id.
27 rbsB
27
28
15 no id.
25
7 metQ
6 metQ, rpsB, grpE
20
20
12 clpB
15
15
1 ybbN
4 groL
10
10
3 dnaK
16 bla
11 hdhA
pI
4.6
5.0
5.4
6.0
pI
4.6
5.0
5.4
6.0
22 dppA
20 ybiS, yggE
C.
25 tolB
D.
13 ushA
10 livJ
αPhoE
p
m
p
m
αSkp
p
m
αSecB
αOmpA
p
m
αFfh
αOmpF
p
m
αPspA
αDegP
precursor
mature
αIbpB
28 rbsB
αSecA
24 oppA
9 fliY
18 znuA
5 potD
14 cysK, ompT, trxB
21 yodA
pl
de
cE
Se
l
ro
nt
co
pl
de
cE
Se
l
ro
nt
co
0 50 100 150 200 250
% of control
19 yehZ
40
0.1
1
10
fold-change
(SecE depl/control)
0.001 0.01 0.1
1
10 100 1000
fold-change
(SecE depl/control)
Figure 4
Mw
(kDa)
1
150
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
100
75
50
37
25
20
41
42
43
44
45
46
47
48
15
10
49
co
nt
Se
ro
l
41
cE
de
pl
Figure 5
A.
Coomassie
control
78
1
5
3
1
51
50
44
33
2
23
24
38
35
4.2
6
28
49
51
6.6
4.2
20
53
38
15
46
34
37
25
45
35
5.8
54
41 44
39
23
24
47
5.0
52
50
33
15
50
43
22
53
100
75
48
29
15 16 20
30
21 31 37
17
26
32
2
39
42
40
27 36
Mw (kDa)
14
20
46
34
3
25
45
18
12
19
13
5
37
50
15 16 20
30
21 31 37
26
17
32
22
7
8
9 25
10
11
4
75
49
14
pI
100
48
36
25 27
9
10
28
11
29
18
12
19
4
SecE depletion
Mw (kDa)
47
5.0
5.8
6.6
B.
phosphorimaging
control
4
5
78
27 36
9 25
10
28
100
75
48
50
3
14
16 20
30
21 31 37
26
33
23
16 20 30
21 31 37
26
25
45
22
2
3
37
50
20
39
2
15
46
4.2
5.0
5.8
6.6
4.2
75
51
50
52
23
0.1
5.0
3 yfgL
44 yfiO
38 ompX
16 tsx
37 yfiO
14 nlpB
29 tolC
7 ostA
23 ompX
31 mipA
28 btuB
20 tsx
17 ybaY
53 ompX
8 yaeT
27 fhuA
19 mdtE
45 ompW
21 mipA
4 fhuE
30 nlpA
2 dcrB
26 yiaF
49 yicH
25 cirA
48 fusA
36 fepA
15 metQ
50 ompA
39 pal
1 mdtE
12 fadL
33 pal
1
0.01
0.001
42
37
54
25
20
39
53
38
34
100
10
50
41
15
46
5.8
C.
fold-change (SecE depl/control)
pI
100
45
33
24
Mw (kDa)
48
22
53
38
42
78
9 25 27 36
10
28
29
18
5
12
6
13
14
4
51
29
18
12
SecE depletion
Mw (kDa)
Coomassie
phosphorimaging
6.6
Figure 6
A.
Coomassie
control
66
68
69
24
38
31
16
83
7
18 25
9
70
75
94
56 67
95
73
57
101
59 58
102
60 74
96
103 110
89
104
61
75
62
105
111
50
54
71
43
72
34
35
10
11
49
44
4
8
5
17
21
6
12
28
45
2932
36
19
20
97
90
37
47
114
115
53
43
107
34
35
17
28
21
37
6
12
55 67
44 49 56
73
59 5758
60 74
61
75
62
45
29 32
36
15
63
51
64
20
39 46
20
13
14
98
52
23
75
109
92
99
100
94
95
89 96
76
50
101
102
103
104
110
105 111
106
112
97
90
37
113
25
65
114
115
77
78
79 80
53
37
100
108
50
27
19
88
72
8
116
15
81
82
30
20
116
107
15
81
82
30
71
10
11
5
83
84
85
70 86
87
54
42
9
4
66
68
69
40
41
38
33
18 25
7
25
77
78
79
80
23
3
106
112
113
76
63
51
39
64
46
65
52
27
15
13
14
50
16
Mw (kDa)
48
22
24
31
1
2
108
86
87
109
92
99
88
100
42
33
100
98
84
85
41
3
SecE depletion
Mw (kDa)
22
21
26
91
117
26
10
91
117
10
Native mass
(kDa)
880
440
160
Native mass
(kDa)
66
880
440
160
66
B.
phosphorimaging
control
69
42
43
10
11
34
35
72
59
17
21
12
56
57
58
80
28
45
36
29
76
63
27
64
65
46
15
23
53
47
75
3
43
50
37
17
114
115
81
56
12
29
50
32
76
63
27
20
39
46
23
53
440
160
37
106
112
97
113
25
65
15
114
115
77
78
79
20
80
107
15
82
117
66
Native mass 880
(kDa)
43
15
81
82
91
10
Native mass 880
(kDa)
50
101
102
96
104 103
110
105
111
64
107
91
94
95
57
58
45
36
100
93
55
89
28
21
25
77
78
79
80
44
88
72
49
59
106
112
113
97
34
35
75
109
92
71
10
11
108
86
54
9
95 101
96 102
104 103
110
105
111
84
85
41
100
50
32
39
109
93 94
55
49
44
88
100
69
92
71
48
108
86
54
9
Mw (kDa)
100
84
85
41
3
SecE depletion
Mw (kDa)
24
117
10
440
160
66
Figure 6
C.
45 ydjN
87 nuoC
97 metQ
67 nuoF
106 ompA, metQ
25 sdhA
84 plsB
17 wzzE, ybdG
69 mrcA
58 dacC, mdtE, cysA
101 emrA
18 nuoC
44 hemX
20 nuoI
80 glnP
33 nuoC
39 mscS
109 dld
7 nuoC
86 ppiD
91 yajC
43 msbA
96 proP
76 metQ
19 nuoB
89 ybhG
21 wzzB
105 glpT
102 dacA, dacC
1 yjeP
54 ppiD
111 lepB
71 secD
99 yijP
75 degS
5 hflC
100 ydgA
65 yfgM
85 gcd
112 metQ, rpsB
103 pyrD
31 yrbD
51 yhbG
88 oxaA
46 exbB
52 ygim, tolQ
4 hflK
104 gadC
63 metQ, nlpA
50 cyoA
3 ftsH
24 acrB
2 kefA
42 ppiD
92 secD
92 yicH
8 creD, hemY
73 cydA
49 cyoB
53 atpF
77 rnfG
Coomassie
phosphorimaging
48 secA
108 dnaK
41 aas, yhjG
9 groL
23 narI #
81 yhcB
28 manX
83 ydij
59 cysA, dacC, hemY
79 atpF
22 narG
72 cydC, cydD
107 pgsA
61 dppD
36 atpG
95 ndh
11 atpD
14 atpF
27 sdhB, manZ
117 yajC
74 metN
60 dacA, dppF
10 atpA
6 ybbK, corA
68 mgtA, copA
35 atpD
66 nuoG
16 fadE, plsB
37 atpF
70 frdA
12 atpG
90 nuoB
13 atpH
110 cysK
82 yhcB
94 dadA, zipA
34 atpA
26 sdhD
0.1
1
10
100
1000
fold-change (SecE depletion/control)
0.00
0.01
0.10
1
fold-change (SecE depletion/control)
44
10
Figure 7
A.
Coomassie
phosphorimaging
5
3
41; yhjG, 2TMs
fold-change (SecE depl/control)
fold-change (SecE depl/control)
41; aas, 2TMs
4
3
81; yhcB, 1TM
2
6; ybbK, 1TM
16; fadE, 2TMs
1
0
2
41; aas, 2TMs
41; yhjG, 2TMs
69; mrcA, 1TM
81; yhcB, 1TM
6; ybbK, 1TM
1
0
0
100 200 300 400 500 600 700 800 900
0 100 200 300 400 500 600 700 800 900
translocated domain
(nr. of amino acids)
translocated domain
(nr. of amino acids)
B.
all periplasmic domains shorter than 60 aa
at least one periplasmic domain longer than 60 aa
Coomassie
phosphorimaging
3
fold-change (SecE depl/control)
fold-change (SecE depl/control)
5
4
3
2
1
2
1
0
0
0
2
4
6
8
10
12
14
0
16
nr. of TMs
2
4
6
8
10
nr. of TMs
45
12
14
16