Embed Size (px)
Citation preview
THESE DE DOCTORAT DE L’UNIVERSITE
D’AIX-MARSEILLE
Soutenue par
Mme Thi Thu Hang LE
Pour obtenir le grade de Docteur de l’Université d’Aix-Marseille
Spécialité: Biochimie structurale
Type VI secretion system effectors
Soutenue le 22 Février 2017 devant le jury :
Dr. Valerie CAMPANACC (I2BC, Gif-sur-Yvette) Rapporteur
Prof. Gérard LAMBEAU (IPMC, Nice-Sophia Antipolis) Rapporteur
Prof. Sophie BLEVES (IMM, AMU, Marseille) Examinateur
Dr. Coralie BOMPARD (USTH, Lille) Examinateur
Dr. Tâm MIGNOT (LCB, Marseille) Examinateur
Dr. Alain ROUSSEL (AFMB, Marseille) Directeur de thèse
Dr. Christian CAMBILLAU (AFMB, Marseille) Co-directeur
Dr. Stéphane CANAAN (IMM, Marseille) Invité
1
TABLE OF CONTENTS
SUMMARY…………………..…………………..……………....…………………….p 3
INTRODUCTION……………………………..…………………..…………………..p 4
I. Bacterial secretion systems: diversity and functions………..…………………….p 4
1.1. Two-step secretion mechanism …………………………………………….……p 7
1.1.1. Type V secretion system (T5SS) …………………………………………….……p 8
1.1.2. Chaperone– usher (CU) pathway T7SS……………………………………..……p 9
1.1.3. Curli biogenesis system T8SS……………………………………………….……p 10
1.1.4. Type II secretion system (T2SS) ……………………………………………….…p 11
1.1.5. Por Secretion System (PorSS or T9SS) ………………………………….….……p 12
1.2. One step secretion systems…………………………….…………………….……p 13
1.2.1. Type I secretion system (T1SS) ………………………………………….….……p 13
1.2.2. Type III secretion system (T3SS) ……………..…………………………….……p 14
1.2.3. Type IV secretion system (T4SS) ……………..…………………………….……p 15
1.2.4. Type VI Secretion System (T6SS) ……………..…………………………….……p 16
Structural assembly of T6SS……………..…………………………..…………….……p 16
Contruction/delivery effector proteins……………..………………...…………….……p 21
Disassembly……………..………………………………………………………….……p 24
II. T6SS effectors……………..……………………………..…………………….……p 24
2.1. Cell wall targeting……………..………………………….………………….……p 24
2.2. Membrane-targeting effectors……………..……….……………………….……p 31
2.3. Nucleic acid-targeting effectors ……………..…………………………..….……p 35
AIMS OF THE THESIS……………..…………………………………….…….……p 37
RESULTS
CHAPTER 1 ……….……..…………………………………………………………...p 38
Foreword ……….……..………….……………………………………………………p 38
Publication: A phospholipase A1
antibacterial Type VI secretion effector interacts directly
with the C-terminal domain of the VgrG spike protein for delivery ……..…….……….p 41
Chapter 1b supplementary experiments (not published) ………………..….………….p 83
1b.1. Crystallization and data collection of the Tle1-Tli1 complex. ….…..….………….p 83
1b.2 Nanobodies generation against Tle1 (SciW). ………………..….………………….p 84
2
1b.3 Structure of the nanobody nbTle1-15 against Tle1. ………………..….……….….p 85
Conclusion and perspectives of Chapter 1b. ………………..………………………….p 87
CHAPTER 2
Foreword ………………………………………………………………..….………….p 88
Publication (in preparation) Purification, characterization, crystallization and preliminary
X-ray studies of the type VI secretion effector/immunity protein Tle3AIEC
/ Tli3AIEC
from
Adherent-Invasive Escherichia coli ..….…………………………………...….……….p 91
1 - Introduction …………………………………………..……………..….………….p 91
2 – Results …………………………………………..……………..….……………….p 92
2.1 Cloning and protein expression ……………………………….….………….…….p 93
2.2. Protein purification ……………………………….….………….…….……….….p 94
2.3. Phospholipase A1 fluorescent assays and inhibition studies ..….……………...….p 97
2.4. Analytical Gel Filtration Analysis and MALS/QELS/UV/RI
coupled size exclusion chromatography ….….…………………………..…….……….p 99
2.5. Crystallization of the Tle3AIEC
- Tli3AIEC
complex 101..….……………….……….p 101
2.6. Data collection and processing ..….……………………………………….……….p 101
2.7. Analysis of preliminary X-ray diffraction results ..….…………………….……….p 101
2.8. Generation of llama nanobodies against Tle3 ..….………………………..……….p 103
3 Conclusion ..….……………….………………………………………………….…..p 104
GENERAL CONCLUSIONS..….……………….….…………………………….…..p 106
PERSPECTIVES..….……………….….………………………………………….…..p 108
REFERENCES..….……………………………….….…………………………….…..p 109
ANNEXES
- Inhibition of Type VI Secretion by an Anti- TssM Llama Nanobody………..….p120
- Production, crystallization and X-ray diffraction analysis of a complex between a
fragment of the TssM T6SS protein and a camelid nanobody ………..………....p134
3
SUMMARY
The Type VI secretion system (T6SS) is a multi-protein machine that delivers protein
effectors in both prokaryotic and eukaryotic cells, allowing interbacterial competition and
virulence. The mechanism of action of the T6SS requires the contraction of a sheath-like
structure that propels an inner tube capped by a spike toward target cells, allowing the
delivery of protein effectors. In 2006, for the first time the Type VI Secretion System (T6SS)
has been described in two bacterial genus, Vibrio cholerae and Pseudomonas aeruginosa.
Later, this system has been found in various bacteria including Burkholderia mallei,
Burkholderia cenocepacia, Edwardsiella tarda, Serratia marcescens, Escherichia coli,
Agrobacterium tumefaciens, Aeromonas hydrophila, Helicobacter, Campylobacter as well as
other organisms. The genomic analysis suggested that, T6SSs are found in approximately
25% of all sequenced Gram-negative bacteria, making T6SS the most widespread specialized
secretion system.
Here, we analyzed the Entero-aggregative Escherichia coli Sci-1 T6SS toxin effectors.
We identified Tle1, a toxin effector encoded by this cluster and show that Tle1 possesses
phospholipase A1 and A2 activities required for the inter-bacterial competition. Self-protection
of the attacker cell is secured by an outer membrane lipoprotein, Tli1, which binds Tle1 in a
1:1 stoichiometric ratio with nanomolar affinity, and inhibits its phospholipase activity. Tle1
is delivered into the periplasm of the prey cells using the VgrG1 spike protein as carrier.
Further analyses demonstrate that the C-terminal extension domain of VgrG1, including a
transthyretin-like domain, is responsible for the interaction with Tle1 and its subsequent
delivery into target cells. Based on these results, we propose an additional mechanism of
transport of T6SS effectors in which cognate effectors are selected by specific motifs located
in the C-terminal regions of VgrG proteins.
Meanwhile, a pathogenic group of E. coli, called adherent-invasive E. coli (AIEC),
has been extensively implicated in human Crohn’s disease (CD) and is currently one of the
most exciting players in the pathogen story. There are at least two gene clusters in the AIEC
genome that encode T6SS components named AIEC LF82 T6SS1 and AIEC LF82 T6SS2.
The protein 435 from the pathogen AIEC LF82 has been predicted to be a phospholipase of
the Tle3 effector family with PLA1 activity from a T6SS1 gene cluster. Its toxicity can be
neutralized by the cognate immunity protein 434 that is a putative Tli3, by forming Tle3 - Tli3
protein complex. The two separated proteins and their complex were then called Tle3AIEC
,
Tli3AIEC
and Tle3AIEC
- Tli3AIEC
complex proteins, respectively. In order to further investigate
4
the related mechanism of Tle3AIEC
and Tli3AIEC
, we performed expression, purification,
characterization, crystallization of the two proteins and preliminary X-ray crystallographic
studies of the Tle3AIEC
- Tli3AIEC
complex in order to understand how Tle3AIEC
protein
recognizes and binds to its cognate Tli3AIEC
effector and inhibits its activity. X-ray diffraction
data were collected from selenomethionine-derivatize Tle3AIEC SeMet
- Tli3AIEC
crystals to a
resolution of 3.8 Å.
INTRODUCTION
I. Bacterial secretion systems: diversity and functions
Bacterial infections have caused several serious diseases continuing threat to human
health. Thus, to fight against this threat, it is very important to understand how such bacterial
survive and prosper [3]. Meanwhile, bacterial colonize almost everywhere in different
environments on the earth. They rarely live in isolation even when it diversity species is low.
Therefore, in bacterial life cycle, their growth is influenced by interbacterial interaction [4].
Bacterial cells develop collaborative or antogonistic mechanisms to communicate, exchange
information or compete for space and nutrition [5]. By using protein secretion systems,
bacteria deliver specific proteins to the extracellular environment or directly into target cells.
Bacterial secretion systems, are molecular nano-machines, play essential roles in
pathogenesis and also in maintaining lines of communication between bacterial cells and their
host [6]. These systems are responsible for the secretion of various substrates including small
molecules, virulence factors, proteins and DNA outside the cell [7, 8]. These substrates, in
term of the response of a bacterium to its environment, were showed to be the important
factors in several physiological processes like adhesion, adaptation, pathogenicity, and
survival. In addition, these substrates were secreted in a variety mean of transport. Depending
on the targeting purpose, they are injected into a target cell, released into the extracellular
space or even only remain associated with the bacterial outer membrane (OM) [8, 9]. In recent
years, advances in structural biology have revealed the diversity of structures and molecular
mechanisms of several bacterial secretion systems. These studies have strongly improved our
knowledge on the complex structures and how secretion machineries deliver their substrates
into the extracellular milieu (EM) or into target cells [8, 9]. Most secretion systems are
relatively complex and each secretion type is identified by the characteristics of the
constituents involved but they share a number of comment features and rules that are
5
hallmarks of all secretion systems. For example, most protein secretion systems require a
component providing energy, usually ATPases to promote movement of substrates across the
whole bacterial cell envelope. It also requires an OM protein that is the ultimate channel
before the secreted protein can access the EM. Other components of the system are involved
in the scaffolding and/or regulation of the macromolecular complex or in the specific
recognition of secreted substrates [8].
So far, the mechanisms of protein secretion have been most extensively investigated in
Gram-negative bacteria [10]. Protein secretion in Gram-negative bacteria is particularly
complex because the complex cell envelope is composed of two hydrophobic IM and OM that
are separated by a hydrophilic space, the periplasm [8-10]. Consequently, to be released in the
extracellular milieu, substrates synthesized in the cytoplasm need to cross both membranes
and the periplasm of donor cells. In contrast, Gram-positive bacteria are generally regarded as
being simpler in structure because they lack a OM. Therefore, the transporters operating in
that membrane are similar to the ones operating in the IM of Gram-negative bacteria.
However, Gram-positive bacteria have a thick cell wall outside their membrane, made of
complex peptidoglycan, and Gram-positive bacteria have developed mechanisms to anchor
proteins or pili to this thick layer [9-11]. As a result, various protein translocation systems
including single- and double-membrane-spanning secretion systems have been existed.
To date, nine different types of protein secretion systems have been unraveled in
Gram-negative bacteria including the six secretion systems (T1SS-T6SS), the chaperone-
usher (CU) pathway (named T7SS) and the curli biogenesis system (named T8SS) and PorSS
(named T9SS) (Figure 1.1) [7, 12-20].
The protein secretion systems characterized to date in Gram-positive bacteria include
the secretion (Sec), twin-arginine translocation (Tat), flagella export apparatus (FEA),
fimbrilin-protein exporter (FPE), hole forming (holin) and WXG100 (Wss) (Figure 1.2) [10].
Some of these systems, such as Sec, Tat and holin also exist in Gram-negative bacteria, where
they mediate export of proteins into the periplasm [10, 21, 22]. Overall, secretion systems in
bacteria are basically divided into two classes consisting of the one-step and the two-step
mechanisms.
6
Figure 1.1. The complement of the secretome involved in colonization process in Gram-negative
bacteria. Among the nine distinct secretion systems, the T1SS, T2SS, T3SS, T4SS, T5SS, T7SS,
T8SS, and T9SS can be involved in colonization process (depicted in red). Extracytoplasmic proteins,
i.e., single proteins and supramolecular protein structures, potentially involved in surface colonization
are depicted in blue. Orange and yellow arrows indicate the routes for proteins targeted to the CM
(cytoplasmic membrane), possessing or lacking an N-terminal SP, respectively. Violet arrows
indicate the routes for exported proteins and red arrows for secreted proteins. CP, cytoplasm; IM,
inner membrane; PP, periplasm; OM, outer membrane; EC, extracellular milieu; SP, signal peptide
[10].
Extra cytoplasmic proteins, i.e., single proteins and supramolecular protein structures, potentially
involved in surface colonization are depicted in blue. Only branches corresponding to the
complement of the secretome involved in bacterial colonization are colored. Orange and yellow
arrows indicate the routes for proteins targeted to the CM possessing or lacking an N-terminal SP.
Violet arrows indicate the routes for exported/secreted proteins (export and secretion are synonymous
in monoderm bacteria). Green arrow indicates proteins integrated into the CM. CP: cytoplasm; CM:
cytoplasmic membrane; CW :cell wall; EC, extracellular milieu; SP: signal peptide [10].
Figure 1.2. The complement of
the secretome involved in
colonization process in Gram-
positive bacteria . Among the 8
distinct secretion systems through
which a secreted protein can be
translocated across the CM, the
Sec, FPE, Tra, and FEA pathways
can potentially be involved in
colonization process in monoderm
bacteria (depicted in red).
7
1.1. Two-step secretion mechanism
Among the secretion systems, the single membrane-spanning secretion systems (the
T5SS, the CU pathway and the curli) and the double membrane-spanning (T2SS and T9SS)
use a two-step secretion mechanism, whereby substrates are first translocated through the IM
by IM-spanning transporters (such as the Sec or the Tat system) and are subsequently
transferred to the OM or secreted into the extracellular space by a dedicated OM-spanning
secretion system.
IM-spanning transporters: The Sec and Tat systems
The Sec and Tat systems are present in the cytoplasmic membranes of most bacteria
and archaea. In archaea and Gram-positive bacteria these systems are responsible for secretion
of proteins across the cytoplasmic membrane, while in Gram-negative bacteria they are
responsible for exporting of proteins into the periplasm [21-23].
Figure 1.3. The Sec and Tat pathways: The Sec system consists of a membrane-embedded,
protein-conducting channel (PCC) comprising of three proteins (SecY, SecE and SecG), which
form the Sec translocase and a peripherally associated, ATP-driven motor protein, SecA. SecD,
SecF and YajC form the translocon-associated complex and YidC is involved in membrane protein
integration and folding. In the Sec pathway, the transport of unfolded proteins occurs either post- or
co-translationally. In case of the former, the fully synthesized preprotein detaches from the
ribosome and is directed to the Sec translocase, with the help of a chaperone, SecB. In co-
translational targeting, the SRP binds to the signal sequence of the secretory protein while it
emerges from the ribosome and the entire ternary complex of SRP/ribosome/nascent secretory
protein chain is targeted to the Sec translocase with the aid of SRP receptor (SR or FtsY). SecA
accepts the unfolded proteins and threads them through the transmembrane channel (PCC). The Tat
system, dedicated for transport of folded proteins, consists of three membrane-integrated subunits,
8
namely TatA, TatB and TatC, that together form receptor and a protein-conducting machinery for
Tat substrates. TatC and TatB form a complex that is involved in recognition of Tat signal
sequences and their insertion into the membrane, while TatA mediates the actual translocation
event. During or shortly after translocation of the proteins, the hydrophobic signal peptide of most
nonlipoproteins is cleaved off by type I signal peptidase, resulting in the release of the mature
protein. SRP: Signal recognition particle [24].
The Sec system is known as the most essential general secretion pathway, which consists of a
protein-conducting channel SecYEG and an ATP-dependent motor protein SecA (Figure 3)
The machinery of the Sec pathway recognizes a hydrophobic N-terminal signal peptide on
proteins destined for secretion and translocates proteins into periplasm by using ATP
hydrolysis and a proton gradient for energy [21]. Whereas, the Tat pathway was later on
identified for protein translocation across the IM using three membrane components, TatA
TatB and TatC (Figure 1.3) [22, 25, 26]. In the Tat pathway, the translocation of the protein is
in a folded state using only a proton gradient as an energy source through the recognization of
a motif rich in basic amino acid residues (S-R-R-x-F-L-K) in the N-terminal region of large
co-factor containing proteins [22, 26-28]. Overall, both systems operate by recognizing
specific signal peptides at the N-terminus of transport substrates, which are cleaved during
translocation through the IM [21, 22]. The key functional difference between the Sec and Tat
systems is that the Sec apparatus translocate unfolded proteins, whereas the Tat pathway
transports proteins that have already folded [21, 22, 27].
1.1.1. Type V secretion system (T5SS)
The T5SS requires the Sec pathway to transfer an unfolded protein through the IM to
the periplasm. This system, belongs to single-membrane-spanning system, is known as the
auto-transporter systems [7]. The common principle of all autotransporters is their
dependency on the Sec machinery for IM transit, and the presence of a -barrel domain that
inserts into the bacterial OM, where it acts as a transporter for the so-called passenger
domain(s) destined for surface localization [7]. The T5SS can be divided into five subclasses
named T5a – T5e (Figure 1.4). Among them, The T5a is known as a monomeric
autotransporter, T5b is identified as two-partner secretion pathway and T5c is described as
trimeric autotransporter. Whereas, the T5d is described as a fused two-partner secretion
pathway, while the T5e can be classified as classical autotransporter but inverted from T5a
The T5SS secretes mainly virulence factors but also participates in cell-to-cell adhesion and
biofilm formation [7]. Like other OM transport systems, the T5SS does not require an ATP
gradient or a proton gradient, as the energy for transport is derived from the folding of the
passenger domain at the pore exit [29].
9
Figure 1.4. Topology models of the different type V secretion systems. Five different type V
secretion systems from type Va to Type Ve. The translocation domain is displayed in brown,
linker/ two-partner secretion-Tps regions in light green, passenger domains in dark green and
periplasmic domains in orange. Polypeptide transport-associated POTRA domains are labelled
(P). For clarification of the topology, N- and C-termini are indicated [7].
1.1.2. Chaperone– usher (CU) pathway T7SS
The CU pathway responsible for pilus biogenesis is the most widespread of the five
pathways that assemble adhesive pili at the surface of Gram-negative bacteria. CU pili
frequently represent various virulence factors, responsible for specific host attachment and/or
the evasion of host responses [13]. In addition to colonization of host surfaces, CU pili are
important for bacterial–bacterial interactions, biofilm formation, and adhesion to abiotic
surfaces CU pili are assembled at the OM by two proteins, a periplasmic chaperone and an
OM, pore-forming protein called the usher [14, 30, 31]. These pili can be divided into two
general architectural classes: the type 1 and P pilus (Figure 1.5) [6]. The type 1 comprises
rigid, rod-like fibers of larger diameter whereas the P pilus comprises thin, flexible fibers or
amorphous structures [14]. In order to be assembled into a type 1 or P pilus, all subunits are
first secreted via the Sec system, through the IM and into the periplasm in an unfolded state.
And then, the subunits are folded and stabilized by a periplasmic chaperone - FimC for type 1
pilus and PapD for P pilus [30, 32].
10
Figure 1.5. General diagram of the CU
mechanism. (A) Diagram of the P pilus
representing all the subunits involved at the
outer membrane. Each subunit is
represented by oval shapes with a distinct
colour and a letter; for example PapG is
labelled G, Pap F is labelled F, and so forth.
(B) Diagram of Type 1 pili with the same
colour scheme as in (A). The pilus
biogenesis is initiated by the export of the
unfolded pilus subunits to the periplasm via
the Sec translocation machinery,
schematically represented in the inner
membrane. The periplasmic chaperone then
captures subunits at the exit of the Sec
translocon and assists in their folding and
targeting to the outer membrane dimeric
usher where one of the pores is activated.
Once at the activated pore, subunits
polymerize in an ordered sequence until pili
completion. The extracellular, outer
membrane, periplasmic space, inner
membrane and cytoplasm are labelled E,
OM, P, IM and C respectively [6]
1.1.3. Curli biogenesis system T8SS
Curli, another type of fimbrial adhesins, are extracellular protein fibres that belong to
the class of functional amyloids, which protect bacteria from hostile environments by
contributing to biofilm formation [33, 34]. This system plays a role in bacterial pathogenesis
by promoting cell adhesion and invasion. Besides, curli also prevent facilitating interactions
between bacteria with the host immune system [33]. The curli subunits including CsgA,
CsgB, CsgC, CsgG, CsgE and CsgF proteins have Sec-dependent signal sequences allowing
their secretion into the periplasm (Figure 1.6) [18].
11
Figure 1.6. Curli formation in E. coli . Curli are
amyloidous protein fibres assembled on the surface
of E. coli cells within the nutrient-depleted zones
of a biofilm and provide structural integrity. The
curli fibre subunit CsgA is exported across the
outer membrane through the CsgG translocator
channel. Once outside the cell, CsgA interacts with
the CsgB nucleator protein, and polymerizes into
amyloidous fibres that extend away from the cell.
The accessory proteins CsgC and CsgE regulate
export by CsgG and CsgF is required for
nucleation of CsgA by CsgB [18].
The lipoprotein CsgG forms a pore like structure in the OM. The major subunit protein
CsgA and the nucleator CsgB are secreted to the cell surface through the CsgG translocator
channel. One outside the cell, CsgA polymerizes on the surface-exposed nucleator CsgB. The
subunit CsgF associates with the OM and is required for cell association of the minor curli
fiber subunit CsgB, whereas, the subunit proteins CsgC and CsgE regulate export by CsgG
and CsgF.
1.1.4. Type II secretion system (T2SS)
The T2SS is the only double membrane-spanning secretion systems using two-step
secretion mechanism [9, 10]. This system, are found in a wide variety of pathogenic and non-
pathogenic Gram-negative bacteria.
The T2SS uses proteins secreted to periplasm via the Sec or Tat system as substrates,
and then translocates these folded proteins across the OM and release in the EM [35]. This
system is a complex structure, composed of four parts: an OM complex, a periplasmic
pseudopilus, an IM platform and a cytoplasmic ATPase (Figure 1.7) [35]. Some of the T2SS
components have been well characterized both structurally and biochemically. However, a
structure of the entire system is yet to be determined [8]. The T2SS has been suggested to use
ATP hydrolysis both to power the assembly of the periplasmic pseudopilus and to push
substrates through the OM channel [8, 15].
12
Figure 1.7. Model of general type II secretion as proposed by Douzi et al. (2012) [36].
Exoproteins (gray circles) are initially translocated over the inner membrane (IM) into the
periplasmic space via the Sec or Tat pathway (not shown). The prepilin peptidase GspO processes
the pseudopilins (GspG–GspK) by cleavage of the leader peptide. The cleaved pseudopilins are
then assembled by ATP hydrolysis of the cytosolic NTPase (GspE) to form the pseudopilus. GspE
is anchored to the inner membrane by GspL and thereby interacts with the IM-embedded GspF. In
active state, the selection of secretion substrates is presumably performed by an interaction of GspC
and GspD components. Then the appropriate exoproteins can enter the T2SS and are pushed into
the extracellular milieu by the assembling pseudopilus. The exoproteins cross the outer membrane
(OM) through the ring-formed channel (secretin) of GspD-subunits. In some species, GspS
stabilizes GspD-subunits and prevents degradation of the secretin [35].
1.1.5. Por Secretion System (PorSS or T9SS)
The PorSS was first identified in the anaerobic Gram-negative bacterium
Porphyromonas gingivalis [37]. Later, this system was also identified in the gliding bacterium
Flavobacterium johnsoniae [38]. The P. gingivalis PorSS is needed for secretion of gingipain
protease virulence factors, while The F. johnsoniae PorSS is needed for secretion of an
extracellular chitinase. In P. gingivalis, the conserved C-terminal domain (CTD) proteins are
transported to the OM by Sec and are secreted onto the cell surface via the PorS [37, 39]. The
CTD proteins are anchored by A-LPS containing anionic polysaccharide repeating units
(Figure 1.8) [19]. However, the precise mechanisms of glycosylation of the CTD proteins and
transportation of LPS to cell surface remain to be determined.
13
Figure 1.8. Model of the T9SS of Porphyromonas gingivalis.. T9SS comprises more than 10
proteins, including PorK, PorL, PorM, PorN, PorP, PorQ, PorT, PorU, PorV, PorW and Sov. Some of
these proteins were expressed using the two-component system PorXY. PorX and PorY are a
response regulator and a sensor kinase, respectively. CTD proteins, such as Kgp [13] and Rgp (R),
are translocated across the IM via Sec machinery and subsequently secreted across the OM through
the T9SS. CTD, C-terminal domains; CP, cytoplasm; IM, inner membrane; OM, outer membrane;
PP, periplasm; T9SS, type IX secretion system [19].
1.2. One step secretion systems
By contrast with the two-step mechanism, the one-step secretion mechanism is based
on the concept that no pool of secreted protein should be identified in the periplasm. Of the
bacterial secretion systems, the double-membrane-spanning secretion systems including
T1SS, T3SS, T4SS and T6SS use a one-step mechanism, such that substrates are transported
directly from the bacterial cytoplasm into the extracellular space or into a target cell.
1.2.1. Type I secretion system (T1SS)
Bacterial T1SSs are widespread among Gram-negative bacteria [8]. This system
mediates the secretion of a large variety of protein substrates such as adhesins, proteases, and
toxins from the cytoplasm into the EM [24, 40, 41]. Moreover, substrates of T1SS have an
impressive range in size from 20kDa up to 900kDa [41]. The T1SS secrete unfolded or
partially folded substrates [8, 41]. Proteins secreted by the T1SSS systems do not contain a
cleavable N-terminal signal peptide. Instead, specific substrates are recognized by a C-
terminal signal sequence, non-cleavable motif and pass both membranes in one step [8, 41].
This system forms a tripartite double-membrane-spanning channel with an IM component
(IMC), the membrane fusion protein (MFP) and a pore-forming OM protein (Figure 1.9) [20].
14
The substrates bind to the IMC and are transferred from the cytoplasm, across the IM into the
periplasmic cavity of the MFP, using the energy generated from ATP hydrolysis [8].
Figure 1.9. Schematic summary of the general
architecture of a T1SS involved in secretion of an
RTX protein, for example HlyA. The inner membrane
component (IMC) belong to ABC transporter (ATP –
biding cassette transporter ) is shown in blue with the
CLD (C39 like domain) highlighted in red, the MFP
(membrane fusion protein) in green and the OMFP ( outer
membrane fusion protein) in orange. Structures of
components of T1SS are also included. The structures of
the ATP-bound dimer of HlyB, the CLD of HlyB and
TolC are shown in cartoon representation. Adopted from
[20].
1.2.2. Type III secretion system (T3SS)
The T3SS (also called injectisome), a fascinating secretion system, is found in various
pathogenic Gram-negative bacteria, forms large supramolecular structures spanning both
membranes. The T3SS is a highly complex molecular machine and is also one of the most
extensively investigated protein secretion system [16, 42, 43].
The principal known function of the T3SS is to deliver effector proteins across the
bacterial and host membranes into the cytosol of host cells, where they may modulate a large
variety of host cell functions, including immune and defense responses [16, 42]. Being
homologous to the bacterial flagellum, T3SS is composed of several components that deviced
into 3 parts: basal body, needle and pore complex (Figure 1.10) [2]. The basal body is a
spaning IM and OM complex which consist of the scaffold proteins (YscCDJ), export
Figure 1.10. Model of the injectisome -Type
III secretion system (T3SS)
Shown is a cartoon depicting the structural
components of the Yersinia injectisome.
Purple, scaffold proteins: YscC, YscD, YscJ;
Orange, export apparatus proteins: YscR,
YscS, YscT, YscU, YscV; Blue, cytoplasmic
components: YscQ (C-ring) and YscN, YscL,
YscK (ATPase complex); Green, YscI (rod)
and YscF (needle); Red, pore complex: LcrV
(needle tip complex) and YopB/YopD
(translocation pore) [2].
15
apparatus (YscRSTUV), ATPase complex (YscNKL) and C ring (YscQ). Polymeration of
YscF extanding the needle from the basal body into the extracellular milieu. Pore complex is
the component that connect the needle tip to the host cell. This complex contains LcrV at the
needle tip and YopB/YopD which is attached to the host cell membrane.
1.2.3. Type IV secretion system (T4SS)
Type IV secretion system (T4SS) is one of important secretion systems that is found in
both Gram-negative and Gram-positive bacteria [9, 12]. In Gram-negative bacteria, this
system span the entire cellular envelope and secrete a wide variety of substrates from single
proteins to protein–protein and protein–DNA complexes [44]. The T4SS can functionally be
classified into three categories including DNA conjugation, secreting effector proteins and
toxins into host cells, mediate DNA release and uptake [45].
Figure 1.11. Secretion pathway and structure of the T4SS system. (a) T4SS system composition
and assembly. The T4S components are represented according to their proposed localization. (b) T4SS
system component structures. The atomic structures are shown in ribbon representation. They include
full-length VirB5 (in orange), the VirB8 periplasmic domain (in light green), full-length VirB11 (in
violet), the VirD4 soluble domain (in magenta) and the T4SS outer-membrane complex comprising
full-length VirB7 (in green) and the VirB9CTD (in yellow) and VirB10CTD (in red). The structure of
the core complex determined using cryo-EM and single particle analysis is rendered as a cut-out
volume (in grey). The core complex is composed of the full-length VirB7, VirB9 and VirB10 proteins.
The structures of VirB3, VirB4 and VirB6 are unknown [44].
Thus, this system has the unique ability among secretion systems to mediate the
translocation of DNA and also operate bacteria to inject effectors into host cells during
infection [12, 44, 45]. The T4SS constituent proteins can be divided into three classes, based
on the role of the individual proteins: the ATPases (VirB4, VirB11 and VirD4), the pilus
16
proteins (VirB2 and VirB5) and the translocation channel proteins (VirB7, VirB8, VirB9 and
VirB10) (Figure 1.11) [44].
1.2.4. Type VI Secretion System (T6SS)
The type VI secretion system is a double membrane-spanning machine using by many
pathogens as the weapon to inject multiple toxin proteins not only into prey prokaryotes but
also into eukaryotic cells, in a contact-dependent and in a one-step manner, during bacterial
infections [4, 46-48]. Since, a high-resolution structure of the whole T6SS complex is not yet
available, the T6SS is believed to resemble a bacteriophage tail-like structure. Encoded
usually within clustered genes, T6SSs are built by the assembly of at least 13 conserved
proteins essential for function named TssA–M (although several have common alternative
names) (Table 1) [49]. Similar to Myophages, the contraction of T6SS leads to secretion of
effector proteins, and no secretion signals have yet been identified. The mode of action of
T6SS could be divided into three main steps: assembly/extended state, substrate
delivery/contruction and disassembly [49] (Figure 1.12)
Structural assembly of T6SS:
The biogenesis of T6SS is first assembly of the membrane and baseplate complexes at
the site of secretion and is followed the assembly of the tube and the tail sheath by the
elongation of the cytoplasmic tubular structure built by Hcp hexamers stacking on each other
coupled to the polymerization of the TssBC sheath.
Figure 1.12. Schematic Representation of the Structure and Mechanism of the Type VI
Secretion System (T6SS). (A) The extended or ‘primed to fire’ machinery is assembled from
cytoplasmic and membrane components. The membrane complex, which may initiate T6SS assembly
17
at the inner membrane, contains TssJ, TssL, and TssM, represented in yellow, red and orange
respectively. A putative baseplate-like structure, formed by TssAEFGK and represented in brown,
sits at the cytoplasmic face of the inner membrane. Upon VgrG, within the baseplate, an elongated
tubular structure of Hcp hexamers (light blue) is built and extends into the cytoplasm, encompassed
in a TssBC sheath (blue). (B) The second step, ‘firing’, corresponds to sheath contraction and propels
the inner tube towards the target cell. PAAR and VgrG, represented in pink and purple triangles
respectively, form the puncturing device responsible for membrane perforation prior to effector
delivery. (C) Once effectors (grey stars) are delivered into the target cell, the contracted sheath is
disassembled by ClpV (green hexamers). Abbreviations: IM, inner membrane; OM, outer membrane;
PG, peptidoglycan [17].
Resembling a bacteriophage tail-like structure, T6SS proteins are deviced into three
main parts, encoring membrane complex, puncturing devices and a predicted baseplate-like
structure (Figure 1.12). This nano-mechinary is stably anchored to the cell envelope of the
attacker by a membrane complex that is assembled by the sequential addition of the subunits,
outer membrane lipoprotein TssJ, TssM, inner membrane proteins TssL (and inner membrane
anchored proteins TagL was found in some systems) (Figure 1.13) [50]. These proteins are
connected by interactions between TssJ and TssM, between TssM and TssL (and between
TssL and TagL) [50, 51]. Structure of the fully assembled TssJLM complex was determined
by negative-stain electron microscopy at 11.6Å resolution. This 5-fold symmetry complex
extends in the periplasm forming a double-ring structure containing the C-terminal domain of
TssM and TssJ that is anchored in the outer membrane [5].
TssJ is a lipoprotein anchored in the outer membrane and exposed in the periplasm. It
is necessary for T6S-dependent secretion of the Hcp protein and for biofilm formation [52].
Crystal structure of the enteroaggregative Escherichia coli Sci1 TssJ lipoprotein exhibits a
transthyretin fold (a sandwich of four sheets) with an additional -helical domain and a
protruding loop that is important for interaction with TssM. Indeed, depleted mutant of this
protruding loop results in losing the interaction between TssJ and TssM in vitro or in vivo
[50].
TssM is a complex protein that anchored in the inner membrane. Containing three
transmembrane domains and a large periplasmic domain, it interacts with TssJ at its C-
terminal region [50]. Studies in the plant pathogen Agrobacterium tumefaciens provided the
first biochemical evidences for TssM exhibiting ATPase activity to power the secretion of the
T6SS hallmark protein, hemolysin-coregulated protein (Hcp). These findings, therefore,
strongly argue that TssM functions as a T6SS energizer to recruit Hcp into the TssM-TssL
inner membrane complex prior to Hcp secretion across the outer membrane [51].
18
Table I. Conserved T6SS components.
The Type VI Secretion System (T6SS) Component Nomenclature (adopted from [17])
Historically, the naming of conserved T6SS components has been confused, with different
names used for different systems. A unified ‘Tss’ nomenclature [53] has now been widely
adopted for the core components; nevertheless, several of them are still more often called by
their common names. The table I below summarises the Tss and most commonly
encountered alternative names of core and selected accessory T6SS proteins, together with
COG descriptors and any phage or secretion system homologues. The names used in this
review are highlighted in bold.
Figure. 1.13. Type VI secretion gene organization (Adopted from [49]). Theminimal set of T6SS
19
genes required for the assembly of a functional T6SS is shown. Genes are partitioned as genes
involved in the assembly of complex (tssJ, tssK, tssL and tssM) or of the tail complex (tssA, tssB,
tssC, hcp, tssE, tssF, tssG, clpV and vgrG). The localization of the products of these genes is
indicated (cyto, cytoplasm; IM, inner membrane; OM, outer membrane). For membrane proteins, the
location of the transmembrane segments (as defined by topology studies) is also shown. For T6SS
subunits implicated in the assembly of the tail complex, the localization in the tail structure is also
indicated (using the phage nomenclature: tube, sheath and baseplate). Questionmarks indicate genes
for which no function and localization has been inferred based on experimental or phylogenetic
evidence (but are thought to be part of the phage tail-related complex based on gene organization
conservation). The clpV gene, although not related to any phage protein, is classified in the tail
complex based on its interaction with the TssBC proteins and its role in the disassembly of the
contracted TssBC sheath.
TssL is anchored in the inner membrane by a transmembrane -helix. Its N-terminal
domain is exposed in the cytoplasm and interacts with TssM N -terminal region [54]. TssL, in
most casess, carries a C-terminal OmpA-like domain exposed to the periplasm which
functions as a peptidoglycan binding domain, otherwide there is an accessory membrane
protein-TagL that interact with TssL will replace the function of peptidoglycan binding
domain [55]. In enteroaggregative Escherichia coli, TagL was showed to be a polytopic inner
membrane protein with three trans-membrane segments and is required for T6SS function
[56].
Puncturing device is composed of Hcp protein – forming inner tube, VgrG protein -
expected to sit at the top of the Hcp tube and TssBC protein – surrounding sheath of the inner
tube. Hcp is the protein structurally homology to gp19 the tail-tube protein of phage T4 [57].
The inner tube was supposed to be formed by the stacking of hexamer Hcp ring [58]. Indeed,
Hcp is assembled in a head-to-tail manner was demonstrated in vivo [59].
Trimeric VgrG protein share a high structural homology to the gp27/gp5 proteins that
forms the tail-spike of bacteriophage T4 [57]. Evidently, structures C-terminal VgrG fragment
of E. coli was showed, indicating a possible evolutional relationship between C-terminal
VgrG fragment and the central spike protein (gp138) of phi92 phage [60]. VgrG is expected to
sit at the top of the Hcp tube, as evidenced by directly interaction of Hcp with VgrG-1 from
Agrobacterium tumefaciens on co-purification from Escherichia coli [61].
TssB and TssC are the two proteins wrap around the Hcp forming tubular sheaths
structures. Upon contraction of these TssBC sheath, T6SS killing machine fires toxins into
target cells. Time-lapse fluorescence light microscopy reveals that sheaths cycle between
assembly, quick contraction, disassembly and re-assembly. In addition, whole-cell electron
cryotomography further shows that the sheaths appear as long tubular structures in either
20
extended or contracted conformations that are connected to the inner membrane by a distinct
basal structure [62].
Cryo-electron tomography studies demonstrated that the sheaths assembling in the
cytoplasm and exist in a contracted form or in a thinner extended form [63]. In 2015,
structures of contracted sheaths were showed by high-resolution cryo-EM at < 4 Å revealing
an overall helical structure with six folds axial symmetry [1, 64] (Figure 1.14). However, the
precise mechanism of sheath contraction is yet to be described.
PAAR proteins were often found downstream of VgrG genes. It binds to the C-
terminal of VgrG protein forming a conical tip of puncturing device complex and also acting
as a site for effector recruitment [65, 66]. Deletion of all PAAR proteins in Acinetobacter
baylyi resulted in almost complete loss of Hcp secretion and killing activity showing that
PAAR proteins are essential for T6SS-mediated secretion and target cell killing [65].
Recently, Cianfanelli has discovered that two different PAAR-containing Rhs proteins can
functionally pair with the same VgrG protein. This study defines an essential yet flexible role
for PAAR proteins in the T6SS and highlights the existence of distinct versions of the
machinery with differential effector specificity and efficiency of target cell delivery [17].
The 5 proteins including TssAEFGK have been proposed to be components of the T6SS
cytoplasmic baseplate complex [49, 67]. Such a complex would be cytoplasmic but anchored
to the inner membrane by association with the membrane complex [62]. Among them, the
TssK protein was exposed in the cytoplasmic but inner membrane-associated trimeric protein
Figure 1.14. Cryo-EM Structure of
the T6SS Sheath (Adopted from
[1].
21
[68, 69]. It shown to co-precipitate with TssBC sheaths through direct interaction with TssC
[62]. In addition, TssK interacts with the cytoplasmic domain of the TssL, and Hcp in E. coli
[68], and with a TssFG complex in S. marcescens [69] suggest that it may have a role in
connecting the T6SS tail and the TssJLM membrane complexes [68]. The cytoplasmic protein
TssE also exhibits detectable similarity to a phage protein (gp25 of T4) and could form part of
a cytoplasmic baseplate-like structure [70]. It is also required for TssBC assembly [62]. Other
core components, TssA, F and G, predicted in the absence of any localisation or functional
information to be cytoplasmic, might also form part of a basal complex, perhaps related to the
phage baseplate.
Contruction/delivery effector proteins
The recent findings gained insight into the process of effector delivery and described
how T6SS loaded and fired effectors into target cells. Similar to the action of contractile
bacteriophage tails, the T6SS delivers effector proteins directly into target cells using a
dynamic ‘firing’ mechanism [17]. However, different effectors are delivered in different
pathways. When the tail fibers make contact with target cells, elongation of the tail structure
occurred, then effectors can be loaded inside the inner Hcp tube or attached to the Hcp or
VgrG or PAAR proteins [57]. Contraction of the tail sheath propels the inner Hcp tube and
spike toward the target, allowing penetration and delivery of the effectors [71]. The effectors
are diverse base on what they are targeted with different activities such as phospholipases,
peptidoglycan hydrolases, nucleases, and membrane pore-forming proteins [3, 4] (Figure
1.15).
In fact, many genes encoding effector are found in close proximately to the Hcp, VgrG
or PAAR genes, suggesting that the effector delivery is associated with the neighbouring core
component [47, 72]. In addition, genomic analyses showed that sequence of genes encoding
specialized effectors contained part of sequence of the hcp, vgrG or paar genes suggesting that
the specialized effector proteins fused to Hcp or VgrG or PAAR proteins [17].
The mechanisms of T6SS delivering effectors in target cells are generally classified
into two pathways that are based on ‘cargo’ effectors - noncovalent interaction with one of the
Hcp, VgrG or PAAR proteins, while ‘specialized’ effectors - covalently fused to one of these
core components [17, 73, 74]. The specialized effectors typically represent as one of multiple
homologues of that core component and often present diverse toxic activities than cargo
effectors [74]. In both cases, effectors are associated with the components of the Hcp-VgrG-
PAAR structure.
22
Figure 1.15. Anti-bacterial Type VI secreted effectors: cellular targets and self-protection
mechanism. Schematic depiction of different classes of effector toxins and their sites of action once
delivered into a target cell. Because this is a resistant (sibling) cell, specific immunity proteins bind to
their cognate effectors in order to neutralize their activity. Dark circles with stars represent effectors,
whilst lighter circles represent corresponding immunity proteins. Examples of atomic structures of
effector: immunity complexes for four classes of effector are shown in the insets (PDB entries:
Tle4PA/Tli4PA, 4R1D; VgrG3VC/TsiV3, 4NOO; Tge1PA/Tgi1PA, 4N88 and Tae4.1SM/Tai4aSM,
4BI8). Black circles represent anti-bacterial effectors whose function is currently unknown. OM,
outer membrane; PG, peptidoglycan cell wall; IM, inner membrane. Adopted from [3].
The first evidence of effector delivery associated with Hcp came from Edwardsiella
tarda, where a direct interaction between the effector EvpP and Hcp was identified [75].
Then, Blondel et al., showed in 2009 that in Salmonella enterica, a gene encoding an orphan
Hcp-like protein fused with the Hcp/COG3157 domain linked to a C-terminal extension,
suggesting that this complex might also be capable of functioning as specialized effectors[76].
Recently, Silverman et al 2013 first showed that Hcp of P. aeruginosa acts as a chaperone and
receptor of substrates. The Hcp-effector bound to the inner surface of the hexameric Hcp ring,
allowing Hcp to direct their secretion and that this binding is required for their stability in the
cytoplasm [77]. The effectors of Hi-T6SS including Tse1, Tse2, Tse3 showed this
characteristic [46, 78, 79]. These Hcp-associated effectors bind the internal cavity of the
hexameric Hcp ring because of the ~40 Å internal pore of the Hcp hexamer [77, 80, 81].
Furthermore, Hcp can define substrate specificity and promote the intracellular stability of
these cargo effectors [73, 77].
23
Genomic evidence suggested that vgrG genes are often found adjacent to putative
effector genes [47, 82]. Besides, VgrG proteins are able to interact with specific cargo
effectors. Koskiniemi et al. demonstrated in 2013 that the Rhs DNases effectors of Dickeya
dadantii are linked to Hcp and VgrG [83]. In V. cholera, the VgrG-1 and VgrG-3 are
described as specialized effectors based on the covalent linkage of the C-terminal effector
domain to the VgrG core domain, while the TseL and VasX are proposed to non-covalently
bind surface features of the VgrG and are thus referred to as cargo effectors [81, 84-86].
Interestingly, Whitney et al. demonstrated in 2014 that the protein effectors of the H1-T6SS
of P. aeruginosa interact with either VgrG or Hcp. These authors found that Tse2 and Tse4,
(the Hcp-stabilized effectors) had function independently of specific VgrG proteins, whereas
Tse5 and Tse6 (the PAAR domain-containing effectors) did not require Hcp for stability and
displayed a strict functional requirement for their cognate VgrG proteins [73].
A subsequent study of Unterweger et al., in 2015 demonstrated that in V. cholera,
translocation of the effector TseL depends on VgrG-1 via effector-specific accessory protein
Tap-1, since the genes vgrG-1 and tap-1 are located immediately upstream of tseL [87].
Recently, Liang et al. reported that T6SS effector chaperone (TEC) proteins shared a highly
conserved domain (DUF4123) are required for effector delivery through binding to VgrG and
effector proteins. The author further demonstrated that TseC secretion requires its cognate
TEC protein and an associated VgrG protein in Aeromonas hydrophila [88].
Many effector domains fused to PAAR proteins that can be involved in specific
effector delivery and have been described in different organisms [65, 89, 90]. Shneider et al.
demonstrated in 2013 that PAAR proteins are essential for T6SS-mediated secretion and
target cell killing by Vibrio cholerae and Acinetobacter baylyi and the VgrG–PAAR spike
complex is decorated with multiple effectors [65]. T6SS-associated Rhs proteins contain
PAAR domains in their N-terminal regions, responsible for binding to VgrG, long Rhs repeat
regions and variable C-terminal effector domains [3, 83]. Studies on Agrobacterium
tumefaciens, by Ma et al. in 2014, showed that the Tde1 effector contains only a recognizable
C-terminal toxin_43 domain could be delivered in the pathway for cargo effectors [48].
Whereas, the sequence of DUF4150 domain of Tde2 revealed significant conservation aligned
with PAAR motif-containing proteins suggested that DUF4150 could act as a PAAR-like
protein 2014 [48]. Thus, the DUF4150 motif of Tde2 effector may be required to adapt or
connect the protein at the tip of a VgrG spike to allow for delivery.
Disassembly
24
The clpV/TssH gene (encoding cytosolic protein AAA+ ATPase), not related to any
phage protein, is classified in the tail complex based on its interaction with the TssBC
proteins and its role in the disassembly of the contracted sheath and recycle TssBC sheath
[49].
II. T6SS effectors
A common feature of effectors targeting bacterial cell is that they prevent
autointoxication by producing specific immunity proteins that bind to the cognate toxin [74].
The classification of these effectors into three groups is based on targeted components of prey
cells: cell wall, membrane and nucleic acids [4].
2.1. Cell wall targeting
In Gram-negative bacteria, the cell wall is the first steady protecting barrier of the cell,
in which the peptidoglycan is a very important component responsible for that. As a result,
and due to its importance, the enzymes degrading peptidoglycan are accounting for the largest
part of T6SS effectors [91]. In order to destroy the bacterial peptidoglycan layer, effectors
show the ability to cleave it at different positions, based on different mechanisms such as
those of amidase (Tae), glycoside (Tge) or muramidase activities.
Tae genes are found adjacent to ORFs that encode immunity proteins [4, 92]. Tai
proteins are exported to the periplasm of the owner strain where they interact with Tae
effector to neutralize their amidase activity [74]. During secretion, T6SS effectors of the
secreting cell do not have access to the preriplasm suggesting that the Tai immunity protein
have a protecting function toward sibling cells toxin injection, rather than from the same cell
itself [46].
Displaying different bond cleavage specificities, Tae effector groups were subdivided
into four sub-groups from in which the towins were named Tae1 to Tae4. Tae1 and Tae4 act
as DL-endopeptidases and are associated with the cleavage of D-Glu–mDAP, whereas Tae2
and Tae3 act as DD-endopeptidases cleaving the cross-bridge between mDAP and D-Ala
[74].
The two effector Tse1 and Tse3 from P. aeruginosa belong to the Tae group and have
peptidoglycan degrading activity; they were therefore renamed Tae1PA
and Tge1PA
,
respectively. They were the two of three first antibacterial T6SS effectors to be biochemically
25
characterized [46]. P.aeruginosa Tse1 (or Tae1PA
) is one of the most thoroughly characterized
T6SS Tae effector [4].
Figure 2.1. Crystal structure of Pseudomonas. aeruginosa Tse1 and Tse1/Tsi1 complex.
(A) Crystal structure of P.aeruginosa Tse1. The central antiparallel β sheet [8] is surrounded by six α-
helices (blue). Cys30, located at the beginning of helixB, is shown as a stick model[93]. (B) Crystal
structure of the Pseudomonas aeruginosa Tse1/Tsi1 complex using the color scheme of figure 1 for
Tse1. Tsi1 is formed by three antiparallel β-sheets (orange) arranged as a partial b-propeller, and a
short C-terminal α -helix (green). The N-terminal β-sheet consists of the β-strands β1, β2, β3, β5, and
β6. The central β-sheet is made of the bstrands b1, b8, b9, b10, and β11. The C-terminal β-sheet is
formed by the β-strands β12, β13, and β14. Residues and disulfide bonds [94] in Tse1 (DSB2) as well
as in Tsi1 (DSB1 and 3) are depicted as sticks.
The available X-ray structure of Tae1PA
shows that, being similar to housekeeping
amidases, it adopts the NlpC/P60 or CHAP (cysteine/histidine dependent
amidohydrolase/peptidase) fold. It is a α /β protein that harbours a conserved Cys-His dyad
[74, 93, 95]. Notably, Tae1PA
is unique among the N1pC/P60 peptidases family members
since it has a single domain and lacks the additional domains that serve as cellular localization
modules present in other N1pC/P60 peptidases [93] (Figure 2.1A).
Its immunity-protein Tsi1 (or Tai1PA
) is produced and transported into the periplasm
to protect from fortuitous sibling cells attacks [93]. Tsi1/ Tai1PA
is formed of three twisted
antiparallel β -sheets resembling a fragmented β-propeller domain followed by a short C-
terminal α-helix. It forms a stable complex with a 1:1 stoichiometry and a Kd of ~2–3 nM
with Tse1 via a contact surface which occludes the active site of Tse1 (Figure 2.1B) [95 , 96].
Noteworthy, structural comparisons between Tse1 alone and in complex with Tsi1 did not
reveal any major conformational changes in the overall structure nor in the catalytic site [93].
Besides Tae1PA
, other amidases were found to be secreted by T6SS, such as the
Burkholderia thailandensis Tae2BT
, Ralstonia pickettii Tae3RP
and several Tae4 from different
speices of S. marcescens, Salmonella typhimurium and Enterobacter cloacae [86, 91, 97-99].
26
The X-ray structures of Tae3, Tai3 and their complex from Ralstonia pickettii has
been determined [100].
Consisting of two subdomains, Tae3, unlike Tae1PA
, possess a typical NlpC/P60 fold
[101]. It consists of a N-terminal sub-domain composed of three α-helices and a C-terminal
sub-domain that has three β-strands packed against a short α-helix. Tae3 harbours a conserved
catalytic triad with a Cys-His-Xaa motif, located in a cleft between the two sub-domains
(Figure 2.2B) [86]. Interestingly, in Tae3 a lid loop covers the catalytic triad resulting in a
close confomation, while in other NlpC/P60 domain proteins the lid was found in an open
conformation [86].
Figure 2.2. Overall structure of the Tae3–Tai3 complex, Tae3 and Tai3. (A) Overall structure
of two Tae3 (full-length) molecules interacting with four Tai3 (23–151) molecules to form a
heterohexamer (PDB codes 4HZ9 and 4HZB). Tae3 molecules are represented in green and the
co-ordinate Tai3 molecules are coloured cyan and violet respectively. (B) Ribbon representation
of the Tae3 monomer (PDB code 4HZB). The secondary structure elements are labelled. The
secondary structure units of Tae3 are arranged in the order α1-α2-α3-β1-β2-α4-β3. The catalytic
residues Cys23, His81 and Ser92 are shown in stick representation. (C) Overall structure of Tai3
shown in a rainbow colour scheme (PDB code 4HZB) [86].
Therefore, in order to neutralize the toxin activity of Tae3, Tai3 (consiting of four
helices and five strands (Figure 2.2C)), inserts a loop (-loop) into the catalytic groove of
Tae3 [86]. The Tae3–Tai3 complex exists as a stable heterohexamer in solution. Its
stoichiometry is rather unusual as two Tae3 molecules are associated two Tai3 homodimers in
a [Tae3]2:[Tai3]4 asssembly (Figure 2.2A) [74].
The crystal structures of Tae4 from E. cloacae (Ec Tae4) and the Tae4-Tai4 complex
from different speices of S. marcescens, Salmonella typhimurium and Enterobacter cloacae
have been determined.
The overall structure of Tae4 exhibits the now common fold of the NlpC/P60 domain,
27
with its two sub-domains: the N-terminal subdomain is formed by five -helices and a -
hairpin, and the C-terminal subdomain is composed of an antiparallel -sheet flanked by a
short -helix [97-99]. Notably, dissimilar to Tae1, a winding loop stablized by a disulfide
bond bears the third catalytic Asp residue that, with the Cys and His residues, form the
conserved catalytic triad (Figure 2.3A). The Tai4 immunity proteins is composed of five -
helices forming a homodimer complex with two Tae4 (Figure 2.3B) [100]. Interestingly, the
crystal structure of Salmonella typhimurium St-Tae4 in complex with the Enterobacter
cloacae Ec-Tai4 provided clear structural evidence to support the previous in vivo
observations of cross-immunity within T6SS families.
Figure 2.3. Overall structure of the
Tae4 and of the StTae4-EcTai4
complex. (A) Overall structure of Tae4
with helices in pink, the sheets in dark
green, and the loops in light green. The
residues from Gly-145 to Gly-148 are
without interpretable electron density in
the crystal and are connected by dashed
lines. The disulfide bond formed
between Cys-137 and Cys-141 is shown
in orange. Two subdomains (N terminus
and C terminus) are connected by loop
L7. Two loops L10 and L11 are involved
in the catalysis site.
(B) Overall structure of StTae4-EcTai4
complex. The heterotetramer is
composed of an EcTai4 homodimer
[named subunit I (cyan) and subunit II
(orange), binding two StTae4 molecules
in green.
It also provides a basis for understanding the important role of cross-immunity in
polymicrobial environments [100].
Tge (glycoside hydrolase effector) has a glucosaminidases or muramidases function. It
performs cleveage of the β1,4 bonds between GlcNAc and MurNAc or between MurNAc and
GlcNAc of the glycan backbone of peptidoglycan. They have been further classed into three
28
sub-groups from Tge1 to Tge3 [46, 102]. Tse3, the first Tge1 studied, belongs to the large
superfamily of T6SS Tge proteins, has a muramidase function. While Tse1 targets the peptide
bonds contained within the peptidoglycan, Tse3 acts on the sugar backbone of the β1, 4 bonds
between muramic acid and N-acetylglucosamine [46]. The Tse3 effector displays a tilted Y-
shape in the complex with Tsi3 [103]. Mainly composed of -helices and short loops, the
Tse3 molecule consist of a small N-terminal domain and a C-terminal catalytic domain with
two major lobes (Lobe1, Lobe2) forming a V shape (Figure 2.4). In agreement with the
known bacteriolytic muramidase activity of Tse3, its catalytic domain adopts a goose-type
lysozyme-like structure with an open, accessible catalytic groove bearing the negatively
charged catalytic key residue Glu-250 and three Ca2+
cations (Figure 4) [103, 104].
Unlike Tsi1 with an all-β fold [95] or Tsi2 with an all- fold [79], the cognate immunity
protein Tsi3 adopts an / β architecture and assembles ten β-strands forming two antiparalles
β- -helix composed of only nine residues[103]. Tsi3 binds tightly
on the top of the Y-shaped Tse3 forming a stable binary complex with Tse3 with a 1:1
stoichiometry [103]. Tsi3 inhibits Tse3 by inserting deeply three loops into its substrate-
binding groove (Figure 2.4) [103].
Figure 2.4. Overall structure of Tse3-Tsi3-N complex. Ribbon and surface representations of the
Tse3-Tsi3-N complex structure. Three loops of Tsi3 protruded to Tse3 and fully occupied the
enzymatic groove (yellow, Tsi3 -sheet; green, Tsi3 -helix; cyan, Tse3 C-terminal domain (residues
147–408); blue, Tse3 N-terminal domain; purple, linker loop (residues127–146); orange, 7-helix of
Tse3); Calcium ions are indicated by arrows [103].
The antibacterial effector, from Pseudomonas protegens (Tge2PP
) belongs to the Tge2 family
member and was shown to be secreted by the T6SS machinery [4]. But, even using a wide
range of activity conditions, it was impossible to detect significant degradation of purified
peptidoglycan sacculi by Tge2PP
. This finding suggested that the effector may require a
29
special periplasmic cofactor. The X-ray crystal structure of Tge2PP
show that it adopts an /β
architecture of lysozyme, resembling the family members of glycoside hydrolases 73, with a
substrate-binding groove located between the large and small lobes. The large lobe is made up
of the short 310 helix, and the helices 1-2 and 4-6, whereas the small lobe consists of 3
and a β-hairpin formed by β1- β2. Glu69 lies deep within the active site and Tyr147 forms a
hydrogen bond with Glu69, in a similar way that in the aforementioned peptidoglycan
glucosaminidase and G-type lysozyme enzymes. (Figure 2.5A,B) [102].
Tgi2PP
structure is different from those of the other immunity proteins of known structure. It
consists of a short N-terminal 310 helix and a central five-stranded β-sheet flanked by three -
helices (Figure 2.5C), topology similar to that of the periplasmic E. coli colicin M immunity
protein [105]. Crystals of the Tge2PP
-Tgi2PP
complex contained a single 1:1 complex in the
asymmetric unit. Tgi2PP
interact with Tge2PP
with a dissociation constant of 0.26 nM,
indicating a very strong binding. This occurs by the insertion of the β-sheet core of Tgi2PP
into
the catalytic groove of Tge2PP
, occluding thus its active site [102].
Figure 2.5. Overall structure of Tge2PP
in complex with Tgi2PP
. Tge2PP
adopts a lysozyme-like
fold. Ribbon (A) and surface (B) representations of Tge2PP
shown at two orthogonal orientations.
Secondary structure elements and the catalytic acid, Glu69, are labeled. (C) structure of the Tge2PP
-
Tgi2PP
complex. Tge2PP
and Tgi2PP
are shown as surface and ribbon representati ons, respectively.
Secondary structure elements of Tgi2PP
are labeled, and Glu69 of Tge2PP
is colored yellow.
Tge3 effectors members have yet to be structurally and functionaly characteried but
bioinformatic analysis suggests that Tge3 effectors carry a conserved catalytic Glu–Asp–Thr
triad similarly to lysozymes of phage T4 [102].
Since VgrG3 protein of V. cholera contains a domain with a predicted muramidase fold, the
presence of a cognate immunity gene suggests that the VgrG3 protein is itself a T6SS effector
[86]. Given that the VgrG3 effector domain is unrelated to Tge family proteins and is likely to
be exported by a distinct mechanism, it may represent a distinct family of muramidase
30
effectors [4]. While all previously characterized VgrG effector domains mediate
microorganism–host interactions [84, 106]. VgrG3 is the first example of a VgrG effector
domain that participates in interbacterial T6SS attacks peptidoglycan layer [81]. Recently, the
C-terminal extension of the VgrG3 protein of V. cholerae has been found to be a
peptidoglycan hydrolase and to cause cell lysis when expressed in the periplasm [81, 107]. Its
crystal structure showed that its C-terminal domain (VgrG3CCD
) adopts a chitosanase fold that
belong to GH family [107] containing nine -helices and a three-stranded β-sheet. VgrG3CCD
is composed of two domains containing nine -helices and a three-stranded β-sheet connected
by a bent 5 helix. VgrG3CCD
possess an Glu/Asp catalytic dyad similar to that of
characterized chitosanases. The VgrG3CCD
immunity protein, TsiV3, is a monomer that
contains a three helical domain stablized by two disulfid bonds (Figure 2.5 b) [107]. TsiV3
strongly interacts with VgrG3CCD
with a Kd of approximately 5 nM. The VgrG3CCD
–TsiV3
complex is a heterotetramer ([VgrG3CCD
]2[TsiV3]2) with an overall architecture similar to
that of the Tae4–Tai4 complex [90] in which three-helical bundle of TsiV3 are inserted into
the deep groove of VgrG3CCD
(Figure 2.6 a).
Figure 2.6 Structure of the VgrG3CCD
–TsiV3 complex. (a) Overall structure of the VgrG3CCD
–
TsiV3 complex. The heterotetramer complex in the asymmetric unit is composed of one TsiV3
homodimer (subunit I, yellow; subunit II, orange) and two VgrG3CCD
molecules (cyan and blue). The
twofold symmetry axis is marked as a black oval. (b) A deep groove between the two domains of
VgrG3CCD
is formed and TsiV3 is inserted into the groove of VgrG3CCD
. The catalytic residues
Glu827 and Asp842 are shown as sticks in deep blue [107].
2.2. Membrane-targeting effectors
31
The bacterial phospholipidic membrane is also an important target for type VI secretion
system-delivered effectors. These effector groups were found to have (phospho) lipase
activities. They have been called Tle effectors and were classified into five sub-groups, Tle1 –
Tle5, based on their sequence and phylogenetic distribution [47]. Tle1–Tle4 families exhibit
hydrolases family fold that is commonly found in lipases, and
esterases. Tle5, however, exhibits a dual HxKxxxxD motif that is common in phospholipase
D enzymes [47]. Eight Tle proteins have been characterized so far including Tle1 (the
Burkholderia thailandensis Tle1
BT [47], P. aeruginosa Tle1
PA [108] and Enteroaggregative
Escherichia coli Tle1EAEC
[109] effectors), Tle2 (Vibrio cholerae Tle2VC [47, 86]), Tle4
(Pseudomonas aeruginosa Tle4 [110]) or Tle5 (P. aeruginosa Tle5aPA
[47, 111], and Tle5bPA
[47, 111, 112], Klebsiella pneumoniae Tle5bKP
[113]) groups but not Tle3.
Recently, the first crystal structural of a T6SS phospholipase effector Tle1 from P. aeruginosa
(Tle1PA
) was solved. Tle1PA
possessed PLA2 activity, consistent with that from Burkholderia
thailandensis [47]. The overall structure of Tle1PA
can be divided into two distinct parts which
has not previously been observed in known lipase structures, the phospholipase catalytic
module and the putative membrane-anchoring module (Figure 2.7 A). The catalytic module
has a canonical /-hydrolase fold that functions as the D1 phospholipase domain. It
composes of seven central parallel -sheets and two small twisted -sheets, flanked by 13 -
helices on both sides (Figure 2.7 B). Its active site adopts an extended and closed binding
pocket that contains the highly conserved catalytic triad Ser235–Asp279– His377. And the
mutation of any of these three residues in catalytic triad abolishes its toxicity. The putative
membrane anchoring module hangs over the phospholipase catalytic module. It perform a
remarkable open conformation consists of three independent domains: D2, D3 and D4 (Figure
2.7 C). Although these three domains were found homologues to several periplasmic or
membrane proteins, its overall conformations are novel and different from known structures.
Noteworthy, unlike D2 and D4 domains, D3 is not integrated into the bilayer, as observed in
the molecular-dynamics simulations, but is directly facing and proximal to the bilayer, which
may establish direct access to extract the phospholipids to the catalytic triad in the catalytic
pocket. Importantly, helices 12, 20 and the loop 20–12 in D3 are formation components
of the catalytic pocket in Tle1 showing the vital role of D3 domain. Indeed, deletion of D3
can obviously reduce the toxicity of Tle1 because of destruction of the catalytic pocket. As a
consequence, D3 functions as a lid/cap domain which not only protect the catalytic site but
also play an essential role in defining selectivity in the hydrolysis of phospholipids by Tle1
32
[108].
Figure 2.7. Structural characteristics of Tle1. (A) Overall view of the Tle1 structure colour-coded as a ramp
from blue (N-terminus) through cyan, green, yellow and orange to red (C-terminus) and composed of two
distinct parts: the phospholipase catalytic module and the membrane-anchoring module. (B), (C) Cartoon
representation showing the domain architecture of Tle1 with topology numbers labelled. The phospholipase
catalytic module consists of the catalytic domain D1 (cyan) with two inserted helices 1 (blue) and 7 (green)
located on the opposite side. The membrane-anchoring module with an open conformation is composed of
domains D2 (wheat), D3 (orange) and D4 [8].
The Tle4-Tli4 complex (PA1510–PA1509) is a newly identified effector-immunity (E–I) pair
from P. aeruginosa. Tle4 (PA1510) is a putative phospholipase effector and is transported by
H2-T6SS (is one of the three distinct T6SS haemolysin coregulated protein (Hcp) secretion
islands have been identified in P. aeruginosa [47, 112] to the periplasm of the neighbouring
cells where it destroys the cell membrane of prey cells. Tle4 is composed of a canonical /-
hydrolase fold domain with a conserved catalytic triad of Ser256, His535 and Asp497 and an
unusual cap domain. It gathers 18 -helices, 15 -strands and eight 310-helices and displays a
lipase-like fold The complex cap domain contains two lid regions (lid1 and lid2 with two
overlapping loops/the dual-door system and one hydrophobic helix) displaying a closed
conformation that buries the catalytic triad in a deep funnel, whereas, in other bacterial
lipases, the lids mainly contain a unique mobile helix (Fig 2.8 A, B).
Interestingly, the Tle4-catalytic domain bears a T-X-S-X-G motif instead of G-X-S-X-G as
observed in other Tle1–Tle4 effector families. Composed of 17 -strands and four helices,
Tli4 displays a two-domain conformation in which these two domains display a similar fold
and topology not only to eachother but also with another T6SS immunity protein Tsi3 [86,
103, 104]. As a consequence, this raises the question whether this fold is widespread in T6SS
immunity proteins [86, 110]. The structure of the Tle4–Tli4 complex was determined and
contains one heterodimer complex molecule per asymmetric unit (Fig 2.8 C) and the molar
33
ratio of Tle4 and Tli4 is in line with the 1:1 stoichiometry observed in solution. Interestingly,
two domains of Tli4 interact with the cap domain of Tle4 creating a grasp-inactivation
mechanism that prevents Tle4 effector protein binding to its substrate. This is a very different
inhibition mechanism among insertion-inactivation mechanism described in T6SS E–I pairs,
as interaction is not directy targeting the active site, but prevents opening of the lid covering
it.
Figure 2.8. Overall structure of Tle4. (A) The structure of Tle4 is shown as a cartoon. The /-
hydrolase fold domain is depicted in marine, the large subdomain of the cap domain is depicted in hot
pink and the small subdomain of the cap domain is depicted in orange. Secondary-structure elements
referred to in the text are labelled. The calcium ion-binding motif is shown, in which the coordinated
residues are shown as ball-and-stick models, and water and calcium molecules are shown as spheres in
cyan and magenta, respectively. The 2Fo-Fc electron-density map for the calcium ion is contoured at 2
sigma. (B) Cap domain of Tle4 in a closed conformation. Tle4 is shown as a cartoon. In the left panel
the /-hydrolase fold domain is shown in blue, the cap domain is shown in hot pink and lid1 and lid2
in the cap domain are depicted in orange and yellow, respectively. The right panel shows detail of the
lid1 and lid2 region, and the residues in these regions are depicted as sticks. The catalytic triad is also
shown as sticks. (C) The overall structure of the Tle4–Tli4 complex. Cartoon diagram depicting the
binary complex, in which Tle4 is shown in marine and Tli4 is shown in salmon.
In 2013, Dong et al. first identified the V. cholerae Tle2VC
(known as TseL) which was shown
to be delivered by the T6SS to competing bacteria, cause lysis and is required to escape
amoeba [86]. Tle2VC
possessed PLA1 activity whereas Tle1BT
and Tle1PA
act as PLA2 [47,
34
108].
Alteri 2016 Tle5-type effectors from P. aeruginosa (PldA and PldB) are phospholipase D
enzymes, which act as effectors against both bacterial and eukaryotic cells [47, 112].
PldA/Tle5aPA
is an eukaryotic-like phospholipase D3 transits the haemolysin co-regulated
protein secretion island II T6SS (H2-T6SS) pathway [47, 111]. In vivo studies on specificity
of Tle5aPA
showed that, by degrading phosphatidylethanolamine – the major component of
bacterial membranes allow P. aeruginosa against a vast array of competitors [47]. Whereas
PldB/ Tle5bPA
is a secreted effector of the H3 Type VI secretion system (H3-T6SS) which is
linked tightly to its three cognate immunity proteins. Interestingly, PldB targets not only the
periplasm of prokaryotic cells and fuctions as an antibacterial activity but also facilitates
intracellular invasion of host eukaryotic cells by activation of the PI3K/Akt pathway,
revealing it to be a transkingdom effector. Recently, an additional member of the Tle5 family
from Klebsiella pneumoniae (Tle5KP
) was characterized. The expression of Tle5KP
in E. coli
results in the accumulation of phosphatidyl glycerol. Moreover, it also acts on eukaryotic
membranes and is essential for K. pneumoniae virulence in a mouse model of infection [113].
All the findings suggest a potentially widespread T6SS-mediated mechanism in which a
single phospholipase effector has capacity of targeting both prokaryotic cells and eukaryotic
hosts [112]. In addition, several T6SS effectors display similarity with pore-forming colicins
such as the V. cholerae VasX, the B. thailandensis BTH_I2691 and the P. aeruginosa
PA14_69520 proteins [90, 91, 114]. These effectors, therefore, are predicted to target the
membranes in a different way compared with T6SS phospholipase effectors.
2.3. Nucleic acid-targeting effectors
Beside the vast T6SS effectors targeting the cell envelope, nucleases and predicted nucleases
are also represented in the T6SS interbacterial effector repertoire [4]. In agreement with the
cytoplasmic localization of the cognate immunity proteins, several effectors identified from
T6SS secretomes were showed to be toxic when expressed in the cytoplasm of recipient cells
[78, 83].
Tse2 from P. aeruginosa is a cytoplasmic-acting effector that acts as a potent inhibitor of
target cell proliferation and is proposed to act as a ribonuclease [79]. It is toxic when
produced into the cytosol of both bacterial and eukaryotic cells, suggesting that the target is
conserved across kingdoms [74, 78]. In agreement with cognate protein localization, the Tsi2
immunity protein interacts with Tse2 in the cytoplasm of the donor cell, and the Tse2–Tsi2
complex is likely dissociated before or during translocation [79].
35
Recently, there are several studies on a subset of recombination rearrangement hotspot
proteins (Rhs proteins) which are known as T6SS nuclease effectors [83, 115]. In Dickeya
dadantii, the two protein RhsA and RhsB, carried C-terminal endonuclease domains, are
delivered by the T6SS in a VgrG-dependent process. Moreover, both RhsA and RhsB proteins
contain PAAR repeat regions, thus, predicts that these T6SS nuclease effectors should bind to
the tip of VgrG trimers through their PAAR repeat domains and in this way be targeted for
secretion and translocation into prey cells [65]. When expressed in Escherichia coli these
endonuclease domains result in chromosomal and plasmid DNA degradation, growth
inhibition and a loss of nucleic acid staining [83]. In Pseudomonas aeruginosa, VgrG1 of the
H1-T6SS is also responsible for delivering an RhsP1 effector protein to target cells [90].
Therefore, Rhs domain proteins function as antibacterial toxins in contact-dependent
inhibition systems, which suggests that Rhs proteins have evolved to be delivered by both the
T6SS and two-partner secretion systems [116]. Noteworthy, two new Rhs antibacterial
effectors delivered by T6SS have been identified from S. marcescens, which one protein was
shown to act as a DNase toxin, while the other contains a novel, cytoplasmic acting toxin
domain. Importantly, this class can play a primary role in competition between closely related
bacteria, and identify a new accessory factor needed for their delivery [3].
In Agrobacterium tumefaciens, a new family of T6SS effectors displayed DNase activity
(named Tde) - dependent on a conserved HxxD motif and its cognate immunities (named Tdi)
has been described. The antibacterial activity of this nuclease provides a competitive
advantage for A. tumefaciens in plant co-infection assays [48] was further demonstrated.
Notably, there are many other T6SS-dependent effectors, either experimentally identified or
predicted based on genomic context, whose function is not known or readily predictable [73,
82, 91, 117-119]. This implies significant diversity of effectors and the promise of novel
cellular targets yet to be identified
36
AIMS OF THE THESIS
General Aim
The aim of this thesis is to better understand bacterial-targeting T6SS through the
identification, characterization and investiagation on the functionality and structure of the
T6SS putative effectors/immuny proteins in EAEC and AIEC.
Specific Aims
1. Identification, characterization and determine the functionality of the T6SS putative
effectors/immuny proteins in EAEC
2. Determine the crystal structure of the T6SS putative effectors/immuny proteins in
EAEC
3. Identification, characterization and determine the functionality of the T6SS
effector/immunity proteins in AIEC
4. Determine the crystal structure of the T6SS putative effectors/immuny proteins in
AIEC
37
RESULTS
CHAPTER 1: A phospholipase A1
antibacterial Type VI secretion effector interacts
directly with the C-terminal domain of the VgrG spike protein for delivery
Foreword
Type VI secretion system is a versatile nano-machine that targets not only prokaryotic
cells but also eukaryotic cells. To meet this goal, diverse effectors can be delivered by the
system. However, each T6SS is specialized and is targets either an eukaryotic or a procaryotic
cell, and not both. However, bacteria exhibit a certain versatility since T6SS+ bacteria can
harbour up to six T6SS gene clusters in their genome.
T6SS is composed of 13 core-components and several accessory proteins that are
required for full activity of the system. This machinery resembles an inverted phage structure
with a phage-homologous tail that is anchored to a trans-membrane complex. It functions by
the contraction of the tail-sheath that provides energy to propel the inner tail-tube, decorated
with protein effectors, towards the target cell, where it injects the toxic proteins.
We engaged in a collaboration with Laure Journet, in the group of Eric Cascales, who
works on Enteroaggregative Escherichia coli (EAEC) sci1 T6SS. Enteroaggregative
Escherichia coli (EAEC) was first described in 1987 [120] and as the most common bacterial
pathogen identified in diarrheal stool samples. This emerging pathogen is becoming
increasingly recognized as a leading cause of sporadic diarrhea in healthy adults and children
[121, 122]. EAEC cause not only acute and persistent diarrheal disease but can also persists in
the human intestine subclinically. EAEC may be the most common cause of acute diarrheal
illness among all age groups in the United States [122], and the recent deadly outbreak of
Shiga toxin-encoding EAEC in Europe suggests that it may become a cause of significant
morbidity and mortality [123, 124]. EAEC may be the second most common cause of
traveler’s diarrhea [125, 126].
The EAEC strain 042, which was isolated from a child with diarrhea in the course of
an epidemiologic study in Lima, Peru in 1983 [127], caused diarrhea and became the
prototypical EAEC strain for the study of virulence factors and EAEC pathogenicity [128].
Thus, numerous virulence factors such as adhesins, toxins and proteins have been described in
38
EAEC 042. EAEC 042 possesses the major inner membrane translocation machines namely
Tat and Sec, for export of proteins to the periplasm. It is not surprising to find within EAEC
042 a complete repertoire of E. coli Sec components. However, investigations suggest that
EAEC 042 does not possess any virulence factors which require the Tat system for secretion
[129]. Recent studies on EAEC 042 have shown that among the 551 genes EAEC specific,
many of them are putative novel virulence factors [129]. Noteworthy, analysis of the EAEC
042 genome has shown that all members of the major protein secretion systems (Type 1–6
and the Chaperone-Usher pathways) required for translocation of proteins to the exterior of
the cell are represented [130].
Within the genome of 042, three predicted type VI secretion systems (T6SS) were
found. Two of them were characterized with the latter locus under the control of AggR [131]
which is a global regulator of EAEC virulence determinants. Meanwhile, the third predicted
T6SS encoded within the EAEC 042 genome is more widely distributed than the other two
T6SS1 and T6SS2. However, there is no experimental evidence that most of the genes [132-
134] of this third T6SS are expressed. Dudley et al. showed in 2006 that the T6SS in a 117-kb
pathogenicity island may be an important mediator involved in aggregative adherence to host
cell surfaces [131].
AggR is a transcriptional regulator of EAEC and has been proposed as the defining
factor for typical EAEC strains. AggR is required not only for the expression of plasmid-
encoded genes but aslo for activates the expression of chromosomal genes. Morin et al.
showed in 2013 that there are at least 44 AggR-regulated genes [135]. Expression of multiple
putative virulence factors, including the aggregative adherence fimbriae (AAF), dispersin, the
dispersin translocator Aat, and the Aai type VI secretion system, have been found to be
regulated by AggR [135]. Besides, Bernard et al. also showed in 2011 that σ54
and bEBP bind
and regulate the expression of orphan hcp and vgrG genes, demonstrating that the core T6SS
cluster and accessory elements are coregulated [136].
In our team we are using as model the Gram-negative bacterium EAEC 042
specifically studying on Sci1 gene cluster (TSS1) to gain structural information on the T6SS
components and also its effector/immunity proteins which are main targetss in my thesis. Bio-
informatics and in vivo data indicated that EAEC sci1 T6SS cluster has two pairs of effector-
immunity proteins, that were named sciY-sciX and SciW-SciV. Later, SciW-SciV was found
to be a Tle1-Tli1 couple, and named accordingly. SciY and Tle1 were predicted to be an
amidase and phospholipase, respectively. At the beginning of my PhD, my initial goal was to
39
express, purify, characterize and crystallize these effector-immunity proteins. Unfortunately,
sciY and sciX could not be purified in soluble forms even with huge effort. In contrast, Tle1
and Tli1 were purified in high yield and exhibited required qualities for further studies. A
large crystallization screening of each protein alone or in complex was performed. Only one
crystal form of the Tle1-Tli1 complex was obtained. Unfortunately, these crystals only
diffracted up to low resolution (>6Å). In order to promote the crystallization of the Tle1-Tli1
complex, llama nanobodies against Tle1 were raised.
From the intensive characterization of Tle1, Tli1 and the Tle1-Tli1 complex, and
together with biochemical and in vivo results obtained by our co-workers, we published an
article in Molecular Microbiology. In this work, I managed to prepare and characterize the
proteins and complexes and I performed the protein crystallization experiments. Silvia
Spinelli helped me in protein crystallography: data collection and data reduction. Aline
Desmyter generated and selected the positive nanobodies and Christine Kellenberger
determined the affinities and the epitopes competition between proteins and nanobodies.
Christian Cambillau and Alain Roussel guided me in the experiments. The in vivo
experiments were performed by our co-workers from LISM, and phospholipase activity was
performed by our co-workers from IMM, Stephane Canaan.
Flaugnatti N, Le TT, Canaan S, Aschtgen MS, Nguyen VS, Blangy S, Kellenberger C,
Roussel A, Cambillau C, Cascales E et al: A phospholipase A1 antibacterial Type VI
secretion effector interacts directly with the C-terminal domain of the VgrG spike protein for
delivery. Molecular microbiology 2016, 99(6):1099-1118.
40
A phospholipase A1
antibacterial Type VI secretion effector interacts directly with the
C-terminal domain of the VgrG spike protein for delivery
Nicolas Flaugnatti1, Thi Thu Hang Le
2,3, Stéphane Canaan
4, Marie-Stéphanie Aschtgen
1¶,
Van Son Nguyen2,3
, Stéphanie Blangy2,3
, Christine Kellenberger2,3
, Alain Roussel2,3
,
Christian Cambillau2,3
, Eric Cascales1 and Laure Journet
1*
1 Laboratoire d'Ingénierie des Systèmes Macromoléculaires, CNRS – Aix-Marseille Université,
UMR 7255, Institut de Microbiologie de la Méditerranée, 31 Chemin Joseph Aiguier, 13402
Marseille Cedex 20, France
2 Architecture et Fonction des Macromolécules Biologiques, CNRS – UMR 7257, Campus de
Luminy, Case 932, 13288 Marseille Cedex 09, France
3 Architecture et Fonction des Macromolécules Biologiques, Aix-Marseille Université, Campus de
Luminy, Case 932, 13288 Marseille Cedex 09, France
4 CNRS – Aix-Marseille Université – Laboratoire d'Enzymologie Interfaciale et de Physiologie de
la Lipolyse, UMR 7282, 31 Chemin Joseph Aiguier, 13402 Marseille Cedex 20, France.
¶ Present address: Laboratoire des Sciences de l'Environnement Marin (LEMAR), Institut
Universitaire Européen de la Mer (IUEM), Université de Bretagne Occidentale, CNRS, IRD,
Ifremer – UMR 6539, Technopôle Brest Iroise, 29280 Plouzané, France.
* To whom correspondence should be addressed: Laure Journet. Laboratoire d'Ingénierie des
Systèmes Macromoléculaires (UMR7255), CNRS, Aix-Marseille Université, 31 Chemin Joseph
Aiguier, 13402 Marseille cedex 20, France. Tel.: 33491164156; Fax: 33491712124; E-mail:
41
SUMMARY
The Type VI secretion system (T6SS) is a multi-protein machine that delivers protein effectors in
both prokaryotic and eukaryotic cells, allowing interbacterial competition and virulence. The
mechanism of action of the T6SS requires the contraction of a sheath-like structure that propels an
inner tube capped by a spike toward target cells, allowing the delivery of protein effectors. Here,
we provide evidence that the entero-aggregative Escherichia coli Sci-1 T6SS is required to
eliminate competitor bacteria. We further identify Tle, a toxin effector encoded by this cluster and
showed that Tle1 possesses phospholipase A1 and A2 activities required for the inter-bacterial
competition. Self-protection of the attacker cell is secured by an outer membrane lipoprotein, Tli1,
which binds Tle1 in a 1:1 stoichiometric ratio with nanomolar affinity, and inhibits its
phospholipase activity. Tle1 is delivered into the periplasm of the prey cells using the VgrG1
spike protein as carrier. Further analyses demonstrate that the C-terminal extension domain of
VgrG1, including a transthyretin-like domain, is responsible for the interaction with Tle1 and its
subsequent delivery into target cells. Based on these results, we propose an additional mechanism
of transport of T6SS effectors in which cognate effectors are selected by specific motifs located in
the C-terminal regions of VgrG proteins.
42
INTRODUCTION
The T6SS is built by the assembly of at least 13 proteins encoded by usually clustered
genes. A trans-membrane complex anchors to the cell envelope a phage-like tail complex that
extends from the membrane in the cytoplasm (Coulthurst, 2013; Zoued et al., 2014; Ho et al.,
2014; Basler, 2015). The membrane complex serves as docking station for assembly of the tail
complex (Durand et al., 2015), a dynamic tubular structure functionally and structurally
homologous to the contractile tail of bacteriophages (Bönemann et al., 2009; Leiman et al., 2009;
Bönemann et al., 2010; Basler et al., 2012). It is constituted of an inner tube made of stacked
hexameric rings of the Hcp protein, whose three-dimensional structure is very similar to that of the
bacteriophage tail tube gpV (Mougous et al., 2006; Pell et al., 2009; Ballister et al., 2008; Brunet
et al., 2014; Douzi et al., 2014). This Hcp edifice resembles a channel-like tubular structure with a
40-Å internal diameter and is surrounded by a contractile sheath made of the TssB and TssC
proteins (Kudryashev et al., 2015). The inner tube/sheath structure is built on an assembly
platform – the baseplate – that contacts the membrane complex (Brunet et al., 2015). The TssBC
sheath is highly dynamic. Cycles of sheath assembly, contraction and disassembly were visualized
by time-lapse fluorescence microscopy using fluorescent TssB–sfGFP fusion constructs (Basler et
al., 2012; Brunet et al., 2013; Kapitein et al., 2013). The inner tube is capped by the spike
composed of a VgrG (valine glycine repeat protein) trimer. This complex is structurally
homologous to the bacteriophage T4 gp27-gp5 cell-puncturing device (Leiman et al., 2009). The
VgrG trimer global fold consists of a gp27-like trimer, followed by the N-terminal OB fold
domain of gp5 and a three-stranded β-helix that forms the needle of the spike complex. In some
T6SSs, an additional component called PAAR (Pro-Ala-Ala-Arg motif-containing protein)
assembles a conical structure at the tip of the VgrG protein (Shneider et al., 2013). This
component was proposed to sharpen the VgrG spike, assist folding and to stabilize the β-helix
domain of VgrG or to be used as an adaptor component mediating interaction between VgrG and
effector proteins (Shneider et al., 2013). The contractile structure assembles in an elongated
metastable state. Upon contact with a target cell, the sheath contraction is thought to propel the
Hcp inner tube towards the target cell, piercing the membrane using the VgrG/PAAR spike
complex, hence leading to effector delivery (Cascales, 2008; Silverman et al., 2012; Coulthurst,
2013; Ho et al., 2014; Zoued et al., 2014). In agreement with this mechanism of action, time-lapse
fluorescence recordings demonstrated that contraction of the sheath is correlated with lysis of the
prey cell (Brunet et al., 2013).
43
The T6SS is a versatile machinery as it has been shown to have roles in both pathogenesis
and inter-bacterial competition, and effectors that have eukaryotic or prokaryotic targets have been
identified and characterized (Russell et al., 2014; Durand et al., 2014, Alcoforado Diniz et al.,
2015). For examples, the Vibrio cholerae and Aeromonas hydrophila T6SSs disable eukaryotic
cells by delivering specific effector modules that interfere with the actin cytoskeleton dynamics
(Pukatzki et al., 2006; Pukatzki et al., 2007; Suarez et al., 2010; Durand et al., 2012) while a
growing number of T6SSs have been demonstrated to have anti-bacterial activities (Hood et al.,
2010; Schwarz et al., 2010; MacIntyre et al., 2010; Murdoch et al., 2011; Gueguen et al., 2013;
Brunet et al., 2013; Carruthers et al., 2013). In fact, bacteria do not live alone in their
environment: they share the same ecological niche, socialize, but also display antagonistic
behaviours and compete with each other. The T6SS is one of the key players during the bacterial
warfare by delivering anti-bacterial effectors directly into bacterial competitor cells (Coulthurst,
2013; Ho et al., 2014; Durand et al., 2014, Alcoforado Diniz et al., 2015). Among these effectors,
the Tae (type VI secretion amidase effector) and Tge (type VI secretion glycoside hydrolase
effector) effectors degrade the peptidoglycan of the target cells, the Tle (type VI lipase effectors)
toxins hydrolyse the membrane phospholipids of the target cells whereas the Tde (type VI DNase
effectors) are nucleases (Russell et al., 2014; Benz and Meinhart, 2014; Durand et al., 2014).
Recently, the P. aeruginosa Tse6 T6SS effector has been demonstrated to have NAD(P)+
glycohydrolase activity, hence depleting the NAD(P)+ pool of the target cell (Whitney et al.,
2015). The cell wall degrading effectors target the peptide stem (Tae) or the glycan strands (Tge)
of the peptidoglycan and can be divided in several families with different hydrolysed bond
specificities (Russell et al., 2014; Benz and Meinhart, 2014; Durand et al., 2014). Tle toxins
consist to a large group of enzymes that could be divided into five divergent families bearing
phospholipase A1, A2 or D activities (Russel et al., 2013). Interestingly, whereas the Tae and Tge
toxins are anti-bacterial only, members of the Tle or Tde toxin families target macromolecules
present in both eukaryotic and prokaryotic cells. Indeed, a number of Tle toxins have been shown
to cause damages in eukaryotic cells, such as the P. aeruginosa PldB (Tle5bPA
) protein that
promotes invasion of eukaryotic cells by activation of the AKT/PI3pathway (Jiang et al., 2013) or
the Tle2VC
toxin that is necessary for V. cholerae to escape amoeba predation (Dong et al., 2013).
To prevent self-intoxication, Tae, Tge, Tle and Tde anti-bacterial effectors are produced
concomitantly with cognate immunity proteins, called Tai, Tgi, Tli and Tdi, respectively (Russell
et al., 2014; Benz and Meinhart, 2014; Durand et al., 2014). Usually, the immunity protein resides
in the compartment in which the toxin is delivered, binds to the toxin with high (nanomolar)
44
affinity and inhibits it, either by the occlusion of the catalytic site or by preventing access to its
target (Russell et al., 2014; Benz and Meinhart, 2014; Durand et al., 2014).
T6SS toxins exist either as independent proteins or additional modules fused to the Hcp,
VgrG or PAAR components. These effector modules are hence delivered into the target cell upon
sheath contraction. The cargo mechanisms by which independent effectors are delivered into the
target cell is less known. The current transport models propose that these effectors bind to the Hcp,
VgrG or PAAR proteins directly or via adaptor proteins, and therefore that these structural
components of the machine are used as carriers (Silverman et al., 2013; Shneider et al., 2013;
Hachani et al., 2014; Durand et al., 2014). Indeed, the P. aeruginosa Tae1 and Tge1 effectors are
embedded into the lumen of the Hcp ring and are stored into the Hcp inner tube before sheath
contraction (Silverman et al., 2013). It has been suggested that several P. aeruginosa effectors
bind directly or indirectly to VgrG (Dong et al., 2013; Whitney et al., 2014; Hachani et al., 2014).
Hachani et al. suggested that VgrG/effector combinations are not interchangeable and that
selection of the effector depends on specific motifs on VgrG (Hachani et al., 2014). More recently,
conserved T6SS adaptor proteins linking VgrG and cognate effectors were identified (Alcoforado
Diniz and Coulthurst, 2015; Unterweger et al., 2015 ; Liang et al., 2015).
In the recent years we have characterized the regulatory mechanisms and the structural
architecture of the entero-aggregative E. coli (EAEC) Sci-1 Type VI secretion system (T6SS-1).
However, the function of this T6SS has remained elusive and no T6SS-1 substrate has been
identified. The EAEC sci-1 gene cluster encodes the 13 core components, a PAAR protein, the
TagL accessory protein, and 6 genes of unknown function. Among these genes, a gene encoding a
putative Tle1 effector followed by a gene encoding a putative lipoprotein is found downstream
vgrG1. In this study, we demonstrate that the Sci-1 T6SS is required for EAEC antibacterial
activity in minimal medium and that Tle1EAEC
possesses phospholipase A1 (PLA1) and A2 (PLA2)
activities responsible for the antibacterial activity of Sci-1 T6SS. We then show that Tli1EAEC
is an
outer membrane immunity lipoprotein that binds tightly to Tle1EAEC
and inhibits its PLA activity.
Finally, we demonstrate that Tle1EAEC
is a cargo effector and is delivered into target cells using the
VgrG1 spike as carrier through direct interaction with the VgrG1 C-terminal extension.
Results
The EAEC T6SS sci-1 gene cluster has anti-bacterial activity in minimal medium.
45
To gain insights into the function of the Sci-1 T6SS, we compared the EAEC 17-2 wild-
type (WT) strain with its derivative strain deleted of the entire sci-1 gene cluster (∆T6SS-1) for (i)
virulence towards eukaryotic cells using the Cænorhabditis elegans model of infection and (ii)
antagonism against competitor bacteria. The experiments were performed in NGM or sci-1
inducing minimal media [137] to allow maximal expression of the sci-1 gene cluster (Brunet et al.,
2011). In these conditions, the growth rates of the WT strain and its isogenic ∆T6SS-1 mutant
were comparable (data not shown).
In C. elegans, the WT EAEC 17-2 cells were moderately virulent, with a lethal dose 50%
(LD50) of 7 days (compared to 3 days for Burkholderia cenocepacia K56-2). Identical values were
obtained when ∆T6SS-1 cells were used as feeding source for the worms (Figure 1A). These data
suggest that the T6SS-1 is not involved in the virulence of EAEC 17-2 in the C. elegans model of
infection. The anti-bacterial activity was therefore tested in SIM. The E. coli K-12 strain W3110
(devoid of T6SS genes) engineered to constitutively produce the green fluorescent protein (GFP)
and to resist kanamycin was used as prey. Attacker and prey cells were mixed in a 4:1 ratio and
the mixtures were spotted on SIM agar plates. After a 4-hour incubation at 37°C, the fluorescence
levels and the number of kanamycin-resistant colony-forming units (cfu) were measured to
estimate the survival of the prey cells (Figure 1B). In these conditions, the deletion of the sci-1
gene cluster increased the recovery of prey cells. We therefore concluded that the T6SS-1 machine
provides anti-bacterial activity in minimal medium.
The EC042_4534 gene product encodes an anti-bacterial effector with phospholipase A1 and
A2 activities.
To identify potential effector toxins, we screened the sci-1 gene cluster. The sci-1 gene
cluster encodes the 14 T6SS core-components and a number of genes of unknown function (Fig
2A). Among those, a group of three genes, EC042_4534, EC042_4535 and EC042_4536 is found
between the vgrG1 (EC042_4533) and PAAR (EC042_4537) genes. EC042_4534 (NCBI Gene
Identifier (GI): 284924255) is encoded directly downstream vgrG1. Computer analysis of the
EC042_4534 gene product using Pfam predicts it carries a DUF2235 domain (uncharacterized /
hydrolase domain, amino-acid 36 to 333, E-value 9.7 e-24). A phylogenetic reconstruction of
EC042_4534 with members of Type VI lipase effector (Tle) families 1 to 5 (as defined by Russell
et al., 2013) showed that EC042_4534 segregates with Tle1 members (Figure S4). Multiple
alignments of EC042_4534 with Tle1 members revealed the characteristic GXSXG motif usually
46
found in lipases and some phospholipases (Figure S5). Fold recognition servers (such as Phyre2,
Kelley and Sternberg, 2009) suggest significant homologies between EC042_4534 and the D1
catalytic domain of the Pseudomonas aeruginosa Tle1PA
protein (Hu et al., 2014; data not shown).
The structure of EC042_4534 was modelled using the structure of Tle1PA
(Protein Data Bank
(PDB) identifier 4O5P, Hu et al., 2014) as template (Fig 2B). As expected, the structural model
predicts that the catalytic triad is composed of the Ser-197, Asp-245 and His-310 amino acids
(Figure 2B). EC042_4534 protein is thus a putative member of the Tle1 family of T6SS effectors
and was therefore named hereafter Tle1EAEC
.
In order to biochemically characterize Tle1EAEC
, its coding sequence was cloned into the
pETG20A E. coli expression vector, fused to a N-terminal hexahistidine-tagged thioredoxin
domain followed by a Tobacco Etch Virus (TEV) cleavage site. The wild-type Tle1EAEC
protein
(Tle1EAEC(WT)
) and a variant bearing a mutation in the putative catalytic triad (Tle1
EAEC(S197A)) were
purified to homogeneity using ion metal affinity chromatography and gel filtration, and the
recombinant Tle1EAEC(WT)
and Tle1EAEC(S197A)
proteins were obtained upon cleavage using the TEV
protease (see inset in Figure 2C). Since the two Tle1 family members characterized so far, B.
thailandensis Tle1BT
and Tle1PA
, have phospholipase A2 (PLA2) activity (Russell et al, 2013; Hu
et al., 2014), the activity of the purified Tle1EAEC
and Tle1EAEC(S197A)
were tested on fluorogenic
phospholipid substrates (Figure 2C). Tle1EAEC
possesses both phospholipase A1 (PLA1) activity
(specific activity (SA) = 1338 pmole.min-1
.mg-1
) and a 12.5 lower PLA2 activity (SA = 107
pmole.min-1
.mg-1
). By contrast, the purified Tle1 protein has undetectable phospholipase C and
triacylglycerol lipase activities (data not shown). The PLA1 and PLA2 activities were abolished
when the putative catalytic Ser-197 residue was substituted by an alanine (S197A mutant) (Figure
2C). Finally, to gain insight into Tle1EAEC
specificity, the rate of hydrolysis of major lipids of
bacterial membranes, phosphatidylethanolamine (DLPE) and phosphatidylglycerol (DLPG), as
well as phosphatidylcholine (DLPC) and phosphatidylserine (DLPS), was tested on monolayer
films (Table 1). Tle1EAEC
– but not its S197A mutant – has significant activity on DLPC, DLPE
and DLPS while DLPG hydrolysis was undetectable. Based on bioinformatic analyses and
biochemical results, we conclude that the EC042_4534 protein is a member of the Tle1 family of
T6SS effectors having PLA1/PLA2 activity.
In order to test whether Tle1EAEC
was involved in the anti-bacterial activity of the Sci-1
T6SS, we constructed a tle1 deletion mutant strain. Due to technical genetic constraints (see
Experimental Procedures), this strain is also deleted of the following tli1 gene (that encodes its
cognate immunity, see below). The Hcp release assay, which reflects proper assembly and
47
function of the T6SS, demonstrated that Tle1 (and Tli1) is not necessary for T6SS assembly and
function (Fig S1). However, as shown in Figure 2D, the absence of Tle1 and Tli1 decreased the
anti-bacterial activity of EAEC against E. coli K-12 cells to the same extent as the T6SS-1
deletion mutant. The anti-bacterial activity was restored by the trans-expression of wild-type tle1,
but not with that of tle1S197A
(Figure 2D). Therefore these results suggest that the anti-bacterial
toxicity of the sci-1 T6SS gene cluster towards E. coli is conferred by the phospholipase activity
of Tle1EAEC
.
Tli1EAEC
(EC042_4535) assures self-protection by inhibiting Tle1EAEC
phospholipase activity.
In T6SS, protection against kin cells is secured by the production of immunity proteins that
specifically bind and inhibit their cognate toxins. Usually, effector/immunity genes are found in
tandem in genomes. In the sci-1 gene cluster, the tle1EAEC
gene is followed by a duplicated region
encoding two putative immunity proteins: EC042_4535 and EC042_4536 (GIs: 284924256 and
284924257 respectively). In the sequenced 042 strain, the 225-amino-acid EC042_4535 and
EC042_4536 proteins only differ by a few residues at their extreme C-termini (Figure S2A).
However, in the 17-2 strain used in this study, the EC042_4536 gene has a frameshift mutation at
nucleotide 386 that yields a 155-amino-acid truncated protein (Figure S2A and S2B) that was not
further analysed. To test if the product of the EC042_4535 gene assures protection against
Tle1EAEC
, the EC042_4535 gene was expressed from the pBAD18 vector in W3110 gfp+ reporter
prey cells. Anti-bacterial competition experiments showed that the production of EC042_4535 in
the E. coli K-12 prey conferred full protection against the anti-bacterial activity of the Sci-1 T6SS
(Figure 3A). In addition to showing that EC042_4535 confers protection, this result also
demonstrates that Tle1EAEC
is delivered into target cells and suggests that Tle1EAEC
is the major
anti-bacterial Sci-1 T6SS effector in these conditions. In agreement with this result, when used as
prey, the ∆EC042_4535-4536 mutant was killed by the wild-type EAEC but not by the ∆T6SS-1
or the tle1-tli1 deletion mutant strain (Figure 3B). Finally, production of EC042_4535 in
∆EC042_4535-4536 prey cells protected them against EAEC killing. It is worthy to note that the
recovered fluorescence of complemented ∆EC042_4535-4536 cells is higher than in non
complemented cells, likely due to the sickness of the ∆EC042_4535-4536 prey cells (Figure 3B).
To confirm this result biochemically, the EC042_4535 protein was purified to homogeneity and
tested for its ability to interfere with the Tle1EAEC
activity. Figure 3C and Table 1 show that the
Tle1EAEC
phospholipase activity was inhibited by EC042_4535 in a dose-dependent manner.
Interestingly, the Tle1EAEC
activity was completely abolished with a Tle1EAEC
:EC042_4535
48
molecular ratio of 1:1 (Fig 3C), which is the highest ratio that can be expected between an enzyme
and a specific inhibitor. Taken together, these results confirm that EC042_4535 is an immunity
protein that protects against the phospholipase activity of Tle1EAEC
, and therefore EC042_4535
was named Tli1EAEC
.
Tli1EAEC
-dependent Tle1EAEC
inhibition is mediated by tight binding of Tli1EAEC
to Tle1EAEC
.
The Tli1EAEC
-mediated inhibition of Tle1EAEC
activity strongly suggests that Tli1EAEC
binds to
Tle1EAEC
. To test this hypothesis, we first performed a bacterial two-hybrid (BACTH) assay. The
Tle1EAEC
coding sequence was cloned downstream the T18 or T25 domains of the Bordetella
adenylate cyclase, whereas the Tli1EAEC
coding sequence (deleted of its lipoprotein signal
sequence) was cloned upstream the T18 and T25 domains. When the Tle1EAEC
and Tli1EAEC
fusion
proteins were co-produced, the expression of the reporter gene was activated demonstrating that
Tli1EAEC
and Tle1EAEC
interact (Figure 4A). The BACTH assay also suggested that Tle1EAEC
and
Tli1EAEC
are monomeric as no Tle1-Tle1 or Tli1-Tli1 interactions were detected (Figure 4A). To
validate these results by alternative approaches, the purified Tle1EAEC
and Tli1EAEC
proteins, and
the mixture of the two purified proteins, were subjected to gel filtration and on-line multi-angle
laser light scattering/quasi-elastic light scattering/absorbance/refractive index
(MALS/QELS/UV/RI). Size exclusion chromatography demonstrated that Tle1EAEC
and Tli1EAEC
have apparent molecular masses of 66 kDa and 23.9 kDa, respectively (Figure 4B). These values,
in agreement with their theoretical molecular weights of 62.3 and 24 kDa, further indicate that
these proteins are monomeric in solution. Analysis of the Tle1EAEC
/Tli1EAEC
mixture showed the
apparition of an additional peak containing both proteins (peak 1, Figure 4B), demonstrating
complex formation between the two proteins in vitro. With an apparent molecular mass of ~ 82
kDa, this complex likely corresponds to a Tle1EAEC
-Tli1EAEC
heterodimer (calculated molecular
weight: 86 kDa). Analyses of purified Tle1EAEC
, Tli1EAEC
and Tle1EAEC
-Tli1EAEC
complex by
MALS/QELS/UV/RI confirmed that both Tle1EAEC
and Tli1EAEC
are monomeric and that the
Tle1EAEC
-Tli1EAEC
complex has a 1:1 stoichiometry (Figure 4C). To determine the strength of the
Tle1EAEC
-Tli1EAEC
interaction, the association of the two partners was monitored by Biolayer
Interferometry (BLI). The Tli1EAEC
protein was biotinylated, coupled to a streptavidin sensortip
and the association and dissociation of Tle1EAEC
were recorded for 600 sec. and 1,800 sec.
respectively (Figure 4D). Based on the curve fitting and assuming a 1:1 Tle1EAEC
:Tli1EAEC
heterodimer, Tle1EAEC
and Tli1EAEC
associates with a KD constant value of 1.5 ±0.05 nM. The Kon
and Koff values (Kon=1.65×105 M
-1s
-1 and Koff=2.2×10
-4 s
-1) are in agreement with a rapid
49
association of the two proteins, but a slow dissociation of the complex. Taken together, these
results demonstrate that Tle1EAEC
and Tli1EAEC
are monomeric in solution and interact to form a
stable 1:1 heterodimer with nanomolar affinity.
Tli1EAEC
is an outer membrane lipoprotein protecting against delivery of cytoplasmic
Tle1EAEC
in the periplasm of target cells.
The tle1EAEC
gene is predicted to encode a 62.3-kDa cytoplasmic protein, with no predicted
signal peptide or trans-membrane segment. Fractionation of EAEC cells producing Tle1EAEC
fused
to a C-terminal VSVG epitope (Tle1VSVG) showed that Tle1EAEC
co-localizes with the EF-Tu
cytoplamic marker hence confirming its cytoplasmic localization (Figure 5A). Tle1VSVG was not
detected in the supernatant fraction (data not shown). By contrast, the tli1EAEC
gene is predicted to
encode a protein with a signal sequence bearing a characteristic lipobox motif, such as other Tli1
homologues (Figure 5B). In agreement with this prediction, Tli1EAEC
processing was inhibited by
the signal peptidase II (SPII) inhibitor globomycin (Figure 5C). In Gram-negative bacteria,
lipoprotein sorting is controlled by the Lol complex and the final localization of the lipoprotein
follows the “+2 rule”, i.e., depends on the residue immediately downstream the acylated cysteine
(Zückert, 2014). In Tli1EAEC
, the +2 residue is an asparagine suggesting an outer membrane
destination (Figure 5B). To determine its sub-cellular localization, total membranes of tli1EAEC
cells producing a functional C-terminally FLAG-tagged Tli1EAEC
protein (Tli1FL) were subjected
to sedimentation density sucrose gradient separation. By comparison with the behaviour of control
proteins in the sucrose gradient (the outer membrane porin OmpF and the inner membrane NADH
oxidase), we concluded that Tli1EAEC
co-fractionates with outer membrane proteins (Figure 5D).
Taken together these experiments defined that Tle1EAEC
is a cytoplasmic protein whereas Tli1EAEC
is an outer membrane lipoprotein.
To explain the interaction between Tle1EAEC
and Tli1EAEC
despite their different
localization, we predicted that Tle1EAEC
should be delivered into the periplasm of the target cell,
and thus be a periplasmic-acting toxin. To test this hypothesis, we tested the effects of the
heterologous production of Tle1EAEC
in the cytoplasm or periplasm of E. coli K-12. Tle1EAEC
was
readily produced in the cytoplasm of E. coli without any toxic effect (Figure 6A). By constrast, we
did not succeed to construct a vector allowing the artificial periplasmic production of Tle1EAEC
by
fusing the tle1EAEC
-coding sequence downstream the ompA signal sequence (sp-Tle1EAEC
). The
construction was only obtained when the cloning steps were performed in cells producing Tli1EAEC
from the pBAD33 vector in the presence of arabinose. Indeed, periplasmic targeting of Tle1EAEC
is
50
highly toxic in the absence of arabinose (Figure 6B). The sp-Tle1EAEC
toxicity is efficiently
counteracted by the co-production of Tli1EAEC
. In agreement with the in vitro and in vivo studies
(Figure 2C and 2D), the periplasmic production of the Tle1EAEC
catalytic mutant Tle1S197A
had
roughly no effect on cell viability (Figure 6B), confirming that Tle1EAEC
bacterial toxicity is
conferred by its catalytic activity.
Tle1EAEC
interacts with the VgrG1 C-terminal extension.
The cargo model proposes that independent effectors are secreted through direct or indirect
(via adaptor proteins) interactions with VgrG, Hcp or PAAR protein components (Silverman et al.,
2013; Shneider et al., 2013; Whitney et al., 2014; Hachani et al., 2014; Durand et al., 2014;
Alcoforado Diniz and Coulthurst, 2015; Unterweger et al., 2015; Liang et al., 2015).
Bioinformatic analyses of sci-1 genes of unknown function showed that no putative adaptor
protein is encoded in this cluster. To gain insights into the secretion mechanism of Tle1EAEC
, we
therefore tested direct pair-wise interactions between Tle1EAEC
and the sci-1-encoded Hcp1,
VgrG1, and PAAR proteins by bacterial two-hybrid. Tle1EAEC
was fused downstream or upstream
the T25 domain whereas the Hcp1, VgrG1 and PAAR proteins were fused to the T18 domain. The
results presented in Figure 7A indicate that Tle1EAEC
interacts directly with VgrG1, but not with
the Hcp1 or PAAR proteins. The Tle1 EAEC
-VgrG1 interaction could not be detected with the T25-
Tle1 fusion protein suggesting fusion to the N-terminus of Tle1EAEC
may causes a steric hindrance
preventing complex formation. To validate these data biochemically, we tested the VgrG1-
Tle1EAEC
interaction using a co-immunoprecipitation assay. To visualize direct interactions, the
two proteins were produced into the E. coli K-12 heterologous host. Western-blot analyses of the
eluted material showed that VSVG-tagged Tle1EAEC
specifically co-immunoprecipitated with
FLAG-tagged VgrG1, but not with Hcp1FLAG (Figure 7B and Figure S3). Taken together, the
BACTH and co-immunoprecipitation assays demonstrate that Tle1EAEC
interacts with VgrG1.
Computer predictions and structural characterization of VgrG proteins showed that these
proteins resemble the gp27-gp5 spike of bacteriophages (Pukatzki et al., 2007; Leiman et al.,
2009). In EAEC, VgrG1 carries an additional C-terminal domain (CTD) separated from the gp27-
gp5 common core by a predicted coiled-coil region (Figure 7C). To assess the importance of this
domain for the VgrG1-Tle1EAEC
interaction we constructed VgrG truncated derivatives lacking
this region (VgrG1-615 lacking the CTD and VgrG1-573 lacking both CTD and the predicted coiled-
coil region). Co-immunoprecipitation and BACTH analyses showed that these truncations abolish
Tle1EAEC
binding, suggesting that the VgrG1 C-terminal domain is necessary for the VgrG1-
51
Tle1EAEC
interaction (Figure 7B and 7D). Furthermore, this domain alone is sufficient to mediate
the interaction with Tle1EAEC
, as shown by co-immunoprecipitation and BACTH experiments
using the VgrG574-841 and VgrG616-841 derivatives (Figure 7B and 7D). The VgrG1 CTD possesses a
domain of unknown function of the DUF2345 family (amino-acid 609-765). DUF2345 domains
are commonly found associated with T6SS VgrG proteins (Boyer et al., 2009). Further
bioinformatic analyses of this CTD using fold recognition servers such as Phyre2 and HHPred
predicted that VgrG1 CTD amino-acid 611-766 region is constituted of a regular repetition of
small short-strands, that are reminiscent to the C-terminal domain of gp5 and likely extends the
VgrG spike. This additional -prism domain is followed by a 62-amino-acid region (residues 780
to 841) predicted to fold as a transthyretin [138]-like domain. To test the contribution of this
domain to the VgrG1-Tle1EAEC
interaction, we constructed a VgrG1 truncated variant lacking the
TTR-like region (VgrG1-778). Figure 7D shows that VgrG1-778 interaction with Tle1EAEC
is
undectable by BATCH analysis. However, VgrG1-778 weakly interacts with Tle1EAEC
using the co-
immunoprecipitation assay (Figure 7B) suggesting that the absence of the TTR domain strongly
affects but not completely abolishes the interaction with Tle1EAEC
. Further BACTH experiments
suggest that this domain is sufficient to mediate the interaction with Tle1EAEC
as the VgrG771-
841/Tle1EAEC
combination activates the expression of the reporter gene (Figure 7D). Attempts to
confirm by co-immunoprecipitation that the TTR domain of VgrG1 is sufficient for this
interaction was unsuccessful, as this truncated variant was undectable by Western blot (data not
shown). Collectively, these results show (i) that the VgrG1 CTD is necessary and sufficient for
the interaction with Tle1EAEC
, (ii) that the TTR domain of VgrG1 CTD is involved in the
interaction but (iii) that a second interaction motif located within the 615-778 DUF2345 region
stabilizes the VgrG1-Tle1 interaction.
To further test the importance of these interaction motifs for the delivery of Tle1EAEC
, we
tested the ability of the VgrG1 C-terminal truncated derivatives to complement a vgrG1 knock-out
mutant in Hcp secretion and killing assays (Figure 8). The co-production of full length VgrG1 and
Tle1 complemented the killing defect of the vgrG knock-out strain (Figure 8A). By constrast,
production of the truncated VgrG1-615 or VgrG1-778 variants (with or without the co-production of
Tle1) did not restore killing (Figure 8A). However, this result is not due to the assembly of a non-
functional T6S apparatus as deletion of the VgrG1 CTD did not affect Hcp release (Figure 8B).
Taken together, these results showed that deletion of the Tle1-binding motif within VgrG1 does
not impair T6SS assembly but abolishes killing of prey cells, hence supporting the idea that Tle1
interaction with the C-terminal extension of VgrG is required for Tle1 export into target cells
52
Discussion
In this study we show that the EAEC Sci-1 T6SS is required for inter-bacterial competition
and report the full characterization of EC042_4534, the first T6SS toxin to be identified in EAEC,
and EC042_4535, its cognate immunity protein. We demonstrate that EC042_4534 has
phospholipase activity, belongs to family 1 of the T6SS lipase effectors (Russell et al., 2013), and
was therefore named Tle1EAEC
. A number of Tle proteins, delivered by the T6SS have been
recently identified on the basis of their vicinity to vgrG genes. They consist of different enzymes
divided into five divergent families (Russel et al., 2013). Tle families 1-4 contain a characteristic
GXSXG motif found in lipases and some phospholipases. In vitro studies have shown that the
Burkholderia thailandensis Tle1
BT and P. aeruginosa Tle1
PA effectors have PLA2 activity (Russel
et al., 2013; Hu et al., 2014). By contrast, the Vibrio cholerae Tle2VC
toxin carries PLA1 activity
(Russel et al., 2013). No enzymatic activity has been assigned so far for members of the Tle3 and
Tle4 families. Tle5 family members contain a duplicated HXDXXXXG motif characteristic of
phospholipase D (PLD) superfamily, and can be divided in two subfamilies (Tle5a and Tle5b),
with Tle5a being eukaryotic-like PLD (Russel et al., 2013; Jiang et al., 2014; Egan et al., 2015;
Spencer and Brown, 2015). The PLD activities of P. aeruginosa Tle5aPA
(known as PldA) and
Tle5bPA
(known as PldB) have been demonstrated (Russel et al., 2013; Jiang et al., 2014, Spencer
and Brown, 2015), while Klebsiella pneumoniae Tle5bKP
T6SS toxin presents a cardiolipin
synthase activity (Lery et al., 2014).
Our analyses showed that Tle1EAEC
has PLA1 and to a lesser extent PLA2 activity, which
contrasts with the previously characterized Tle1 members, Tle1BT
and Tle1PA
, which have PLA2
activity only (Russel et al., 2013; Hu et al., 2014). It is noteworthy that the Tle classification was
built on their protein sequences and phylogenetic distribution, and not on their activity. It remains
possible that Tle1 members may not have all the same selectivity for the sn-1 and sn-2 positions.
Tle1 toxins consist to very heterologous proteins in terms of size (~ 500-900 residues). They all
possess a DUF2235 (α/β hydrolase fold domain) likely forming the catalytic module but bear
distinct additional domains. For example, Tle1PA
has a putative C-terminal membrane-anchoring
module (Hu et al., 2014) that has been shown to be critical for the catalytic activity, suggesting
that the activity of Tle1 proteins might also be regulated by these additional domains. Further
biochemical and structural characterizations of Tle1EAEC
and other Tle1 proteins are therefore
required to better understand the differences in substrate selectivity on the phospholipid sn-1 and
53
sn-2 moieties.
Our results also showed that Tle1EAEC
is required for the anti-bacterial activity conferred by
the Sci-1 T6SS. Tle1BT
and Tle4APEC
were previously shown to be required for the anti-bacterial
activity of B. thailandensis and avian pathogenic Escherichia coli, respectively (Russel et al.,
2013; Ma et al., 2014). By contrast, Tle5aPA
(PldA) and Tle5bPA
(PldB) are required for both
bacterial competition and virulence (Jiang et al., 2014). Tle5bPA
delivery into eukaryotic host cells
promotes invasion through the activation of the AKT/PI3pathway (Jiang et al., 2014). Similarly,
the Tle2VC
toxin is necessary for the anti-bacterial activity of V. cholerae and its ability to escape
amoeba predation (Dong et al., 2013). In K. pneumonia, Tle5KP
is required for full virulence in a
mouse model of infection (Lery et al., 2014). Although we did not observe any effect of the Sci-1
T6SS on C. elegans viability, we cannot rule out that the Tle1EAEC
toxin might be delivered into
eukaryotic host cells to create damages. It would be interesting to test the role of the Sci-1 T6SS
and of its specific Tle1EAEC
toxin on epithelial intestinal cells.
Our results also demonstrated that the production of Tle1EAEC
in the cytoplasm of E. coli
K-12 has no effect on its viability. By contrast, cells do not survive when Tle1EAEC
is exported to
the periplasm. This result is consistent with the observations that the heterologous periplasmic
production of Tle1PA
, Tle5aPA
and Tle5bPA
is toxic (Jiang et al., 2014; Hu et al., 2014). One may
hypothesize that (i) Tle1EAEC
targets specific lipids found in the outer leaflet of the inner
membrane or in the inner leaflet of the outer membrane and/or (ii) that dedicated periplasmic
proteins are required to activate Tle1EAEC
, as previously shown for the colicin M toxin (Hullmann
et al., 2008). The activity of Tle toxins in the periplasm is in agreement with the synthesis of a
cognate immunity protein called Tli that are usually periplasmic soluble proteins or membrane-
anchored lipoprotein (Russel et al., 2013). In this work, we have shown that Tli1EAEC
(EC042_4535) confers protection against Tle1EAEC
. Fractionation, isopycnic centrifugation and
processing inhibition assays demonstrated that Tli1EAEC
is an outer membrane lipoprotein.
Tli1EAEC
-mediated inhibition of Tle1EAEC
occurs by protein-protein contacts and is very efficient,
as a molecular ratio of 1:1 totally abolishes Tle1EAEC
phospholipase activity. The interaction
between the two partners was observed in vivo by bacterial two-hybrid analyses and biochemical
approaches. Other immunities to phospholipases characterized so far have been shown to inhibit
the action of the effector by direct protein-protein contacts, such as the P. aeruginosa Tle5/Tli5
pairs (Russell et al., 2013; Jiang et al., 2014). In vitro analyses of the purified Tle1EAEC
/Tli1EAEC
54
complex by gel filtration, MALS/QELS/UV/RI and biolayer interferometry collectively
demonstrated that Tle1EAEC
interacts with Tli1EAEC
with a 1:1 stoichiometry and a KD of 1.5 nM.
This tight binding, with a very fast association and a very slow dissociation is in accordance with
the role of Tli1EAEC
as a specific immunity protein, as the control of Tle1EAEC
activity should be
strict among bacteria from the same species and should occur rapidly after delivery of Tle1EAEC
.
This nanomolar affinity is in the same range as other T6SS effector/immunity pairs such as
Tae1/Tai1 couples (Ding et al., 2012; Shang et al., 2012) and suggests an extensive surface of
contact between the two partners. Indeed, the crystal structure of the putative P. aeruginosa Tle4
effector in complex with its putative immunity protein (Tli4PA
) revealed that Tli4PA
covers a
surface area of ~ 2,800 Å2 and use a grasp mechanism to prevent the interfacial activation of
Tle4PA
(Lu et al., 2014). Further structural characterization of the EAEC Tle1/Tli1 complex is
required to better understand the molecular bases for this efficient inhibition.
In this work, we also addressed the secretion mechanism of Tle1EAEC
. Several mechanisms
have been identified or proposed for independent effectors, and all of them involve direct or
indirect contacts with the Hcp rings, the VgrG spike or the PAAR protein that sits on VgrG
(Silverman et al., 2013; Shneider et al., 2013; Whitney et al., 2014; Hachani et al., 2014; Durand
et al., 2014). Recently, conserved adaptor proteins of the DUF4123 family interacting with both
VgrG and the effector were shown to be required for the translocation of a number of T6SS
effectors (Liang et al., 2015 ; Unterweger et al., 2015). Here, co-immunoprecipitation and
bacterial two-hybrid analyses demonstrated that Tle1EAEC
interacts directly with VgrG1 and that
this interaction is required for proper Tle1EAEC
delivery. A gene encoding a putative PAAR protein,
EC042_4537, is found immediately downstream Tli1. However, PAAR is not required for the
VgrG1/Tle1EAEC
direct interaction, suggesting that it constitutes a structural element at the tip of
the VgrG spike or that it is involved in the recruitment and transport of a yet unidentified effector.
In the cargo transport hypothesis, the VgrG proteins are used as carriers to deliver effectors into
the target cell (Durand et al., 2014). In P. aeruginosa, several toxins have been shown to be
dependent on dedicated VgrG proteins for their delivery suggesting that VgrG proteins bear
specific sequences to select cognate effectors (Whitney et al., 2014; Hachani et al., 2014). Our
results support this idea as we found that the C-terminal extension of VgrG1 is necessary and
sufficient to mediate binding to and transport of Tle1EAEC
. Particularly, we identified a region of
62 amino-acids at the C-terminus of VgrG1 involved in this interaction. This short domain is
predicted to fold as a transthyretin-like [138] domain. TTR domains are putative protein-protein
55
interaction modules. Interestingly, TTR domains were previously identified as PAAR protein
extensions, and were proposed to be adaptors to mediate interaction with effector proteins
(Shneider et al., 2013). However, deletion of the TTR domain of VgrG1 did not totally abolish the
interaction with Tle1EAEC
, suggesting that another motif may be present in the DUF2345 domain
of VgrG1. Collectively, these results demonstrate that DUF2345/TTR domains are involved in
selection and transport of T6SS effectors. One may hypothesize that distinct motifs should be
involved in the recruitment of effectors to confer specificity. Indeed, although this needs to be
experimentally verified, a disordered loop within the C-terminal β-helix of the E. coli O157
VgrG1 protein was proposed to be an interaction site with effectors (Uchida et al., 2014).
Similarly, the recent release of the structure of the P. aeruginosa VgrG1 spike (PDB: 4MTK)
reveals the existence of a small C-terminal helix that folds along the VgrG β-helix. We therefore
propose that additional C-terminal domains in VgrG and likely on PAAR proteins might be
considered as internal adaptors for interaction with effectors and that the EAEC VgrG1
DUF2345/TTR domain represents such a motif. Further experiments on different T6SS effectors
will likely highlight the diversity of these selection modules.
Experimental Procedures
Bacterial strains, growth conditions and chemicals
The E. coli strains and plasmids used in this study are listed in Supplemental Table S1. The
entero-aggregative E. coli EAEC strain 17-2 and its ∆T6SS-1, ∆tle1-tli1, ∆tli1-tli1b isogenic
derivatives were used for this study. E. coli K-12 DH5α, W3110, BTH101, and T7-Iq pLys strains
were used for cloning steps, co-immunoprecipitation, bacterial two-hybrid and protein purification
respectively. The E. coli K-12 W3110 strain carrying the pUA66-rrnB plasmid (gfp under the
control of the constitutive rrnB ribosomal promoter, specifying strong and constitutive
fluorescence, and kanamycin resistance (Zaslaver et al., 2006)) was used as prey in antibacterial
competition experiments. Strains were routinely grown in LB rich medium (or Terrific broth
medium for protein purification) or in Sci-1 inducing medium (SIM; M9 minimal medium,
glycerol 0.2%, vitamin B1 1 µg/ml, casaminoacids 100 µg/ml, LB 10%, supplemented or not with
bactoagar 1.5%) with shaking at 37°C. Nematode growth plates (NGM) were used for the C.
elegans infection assay. Plasmids were maintained by the addition of ampicillin (100 μg/mL for E.
coli K-12, 200 μg/mL for EAEC), kanamycin (50 μg/mL) or chloramphenicol (30 μg/mL).
Expression of genes from pBAD, pOK12 and pASK-IBA vectors was induced at exponential
phase for one hour with 0.1% of L-arabinose (Sigma-Aldrich), 100 μM of isopropyl--D-thio-
56
galactopyrannoside (IPTG, Eurobio) and 0.1 µg/mL of anhydrotetracyclin (AHT, IBA
Technologies) respectively. 5-bromo-4-chloro-3-indolyl--D-galactopyranoside (X-Gal, Eurobio)
was used at 40 µg/mL. Globomycin (a kind gift of Dr. Danièle Cavard) was used at 50 µg/mL.
Oligonucleotides and plasmids used in this study are listed in Supplemental Table S1.
Strain construction
The tle1 (EC042_4534) and tli1 (EC042_4535) genes were deleted into the EAEC 17-2 wild-type
strain using a modified one-step inactivation procedure (Datsenko et al., 2000) as previously
described (Aschtgen et al., 2008) using oligonucleotide pairs DEL-4534-5-DW/DEL-4534-3-DW.
This oligonucleotide pairs carry 50-nucleotide 5’ extensions homologous to regions adjacent to
tle1. Because tli1 is duplicated (tli1b has the same 5’ sequence than tli1), attempts to delete tle1
only yielded to the deletion of both tle1 and tli1 genes (∆tle1-tli1). The tli1 (EC042_4535) and
tli1b (EC042_4536) genes were deleted into 17-2 wild-type strain as described above using
oligonucleotide pairs DEL-4535-5-DW/DEL-4535-3-DW. Kanamycin resistant clones were
selected and verified by colony-PCR. The kanamycin cassette was then excised using plasmid
pCP20. The deletions of the gene of interest were confirmed by colony-PCR and complementation
studies.
Plasmid construction
Custom oligonucleotides were synthesized by Sigma Aldrich and are listed in Supplemental Table
S1. EAEC E. coli 17-2 chromosomal DNA was used as a template for all PCRs. E. coli strain DH5
was used for cloning procedures. Polymerase Chain Reactions (PCR) were performed using a
Biometra thermocycler using the Q5 High fidelity DNA polymerase (New England Biolabs). All
the plasmids (except pETG20A and pOK12 derivatives) have been constructed by restriction-free
cloning (van den Ent and Löwe, 2006) as previously described (Aschtgen et al., 2010a). Briefly,
genes of interest were amplified with oligonucleotides introducing extensions annealing to the
target vector. The double-stranded product of the first PCR has then been used as oligonucleotides
for a second PCR using the target vector as template. For the pETG20A-Tle1 and pETG20A-Tli1
constructs, the genes encoding Tle1EAEC
and signal sequence-less Tli1EAEC
were amplified by PCR
using specific Gateway® primers containing attB sequences, which allow insertion into the
pDONR201 cloning vector by the BP recombination reaction, and then introduced into the
pETG20A vector. The final constructs allow the production of Tle1EAEC
or Tli1EAEC
fused to an N-
57
terminal hexahistidine-tagged thioredoxin (TRX) followed by a Tobacco etch virus (TEV)
protease cleavage site. For pOK-TliHA, pOK-VgrGF, pOK-VgrG1-573F and pOK-VgrG1-615F, the
coding sequence of tli, vgrG and the different vgrG domains were amplified by PCR using
oligonucleotides introducing EcoRI and XhoI restriction sites and cloned into the pOK12-
derivative vector pMS600 (Aschtgen et al., 2008) digested by the same enzymes. In addition, the 3’
oligonucleotide contains the sequence encoding the FLAG epitope, allowing C-terminal in-frame
fusion of the vgrG derivatives with the FLAG epitope. The Ser197-to-Ala substitution was
introduced in the pETG20A-Tle1, pBAD-Tle1-Tli1 and the pIBA-sp-Tle1 plasmids by
QuickChange PCR-based targeted mutagenesis (Supplemental Table S1). Mutations were
confirmed by DNA sequencing (GATC Biotech or Eurofins).
Caenorhabditis elegans infection assay
Virulence towards C. elegans was tested by a slow killing assay. L4 to adult stage nematods
grown on E. coli OP50 were placed on unseeded NGM plates for 24 hours at 25°C. Twenty-five
worms were then picked and placed onto lawns of bacteria to be tested. The viability of each
individual was evaluated on a daily basis and the number of surviving nematods was plotted over
time. The E. coli K-12 OP50 and B. cenocepacia K56-2 strains have been used as controls.
Interbacterial competition assay
The antibacterial growth competition assay was performed as described for the studies on the
Citrobacter rodentium and EAEC Sci-2 T6SSs (Gueguen et al., 2013; Brunet et al., 2013) with
modifications. The wild-type E. coli strain W3110 bearing the pUA66-rrnB plasmid (KanR
(Zaslaver, 2006)) was used as prey in the competition assay. The pUA66-rrnB plasmid provides a
strong constitutive green fluorescent (GFP+) phenotype. Attacker and prey cells were grown for 16
hours in SIM medium, and then diluted in SIM to allow maximal expression of the sci-1 gene
cluster (Brunet et al., 2011). Once the culture reached an OD600nm ~ 0.8, cells were harvested and
normalized to an OD600nm of 10 in SIM. Attacker and prey cells were mixed to a 4:1 ratio and 15-
µl drops of the mixture were spotted in triplicate onto a pre-warmed dry SIM agar plate
supplemented or not with arabinose 0.02% or IPTG 20 M. After 4-hour incubation of the plates
at 37°C, fluorescent images were recorded with a LI-COR Odyssey imager. The bacterial spots
were scratched off, and cells were resuspended in LB medium supplemented with
chloramphenicol and normalized to an OD600nm of 0.5. Triplicates of 150 µl were transferred into
58
wells of a black 96-well plate (Greiner) and the absorbance at 600 nm and fluorescence (excitation,
485 nm; emission, 530 nm) were measured with a Tecan Infinite M200 microplate reader. The
relative fluorescence was expressed as the intensity of fluorescence divided by the absorbance at
600 nm, after subtracting the values of a blank sample. These results are given in arbitrary units
(AU) because the intensity of fluorescence is acquired with a variable gain and hence varies from
one experiment to the other. For estimation of cfu, fluorescent KanR colonies were enumerated
under UV light. The experiments were done in triplicate, with identical results, and we report here
the results of a representative experiment.
Computer algorithms for phylogenetic analyses and Tle1EAEC
structure modelling
Phylogenetic tree reconstruction has been made using Phylogeny.fr (Dereeper et al., 2008). The
homology model of Tle1EAEC
was built with Coot (Emsley et al., 2010) based on a Multalin
alignment with the published P. aeruginosa effector Tle1PA
(PDB: 4O5P, Hu et al., 2014). The
regions present in Tle1EAEC
but not in 4O5P were not included in the modelling.
Bacterial two-hybrid assay (BACTH)
The adenylate cyclase-based bacterial two-hybrid technique (Karimova et al., 1998) was used as
previously published (Battesti and Bouveret, 2012). Briefly, pairs of proteins to be tested were
fused to the isolated T18 and T25 catalytic domains of the Bordetella adenylate cyclase. After
transformation of the two plasmids producing the fusion proteins into the reporter BTH101 strain,
plates were incubated at 30°C for 48 hours. Three independent colonies for each transformation
were inoculated into 600 μL of LB medium supplemented with ampicillin, kanamycin and IPTG
(0.5 mM). After overnight growth at 30°C, 10 μL of each culture were dropped onto LB plates
supplemented with ampicillin, kanamycin, IPTG and X-Gal and incubated for 16 hours at 30 °C.
The experiments were done at least in triplicate and a representative result is shown.
Purification of Tle1, Tle1S197A
and Tli1
E. coli T7 Iq pLys cells carrying the pETG20A-Tle1, pETG20A-Tle1S197A
or pETG20A-Tli1
plasmids were grown at 37°C in terrific broth to an OD600 ~ 0.9 and tle1, tle1S197A
or tli1
expression was induced with IPTG (0.5 mM) for 16 hours at 17°C. Cells were harvested by
centrifugation and stored at -80C. The cell pellet was resuspended in Tris-HCl 20 mM pH 8.0,
59
NaCl 300 mM, glycerol 5% (v/v), lysozyme (0.25 mg/ml), DNase (2 g/ml), MgSO4 20 mM, and
phenylmethylsulfonyl fluoride (PMSF) 1 mM and cells were lysed by ultrasonication on ice. The
insoluble material was discarded by centrifugation at 20,000 × g for 60 min at 4C. The soluble
thioredoxin 6×His-tagged Tle1, Tle1S197A
or Tli1 fusion proteins were purified by affinity
chromatography on a nickel–nitrilotriacetic acid resin (Bio-Rad) and the tag was removed after
dialysis by overnight hydrolysis with the TEV protease and reloading in presence of 10 mM
imidazole. The proteins were further purified by gel filtration chromatography (Superdex 75,
10/30 GE Healthcare) equilibrated in Tris-HCl 20 mM pH 8.0, NaCl 300 mM, dithiothreitol
(DTT) 2 mM using an AKTA purifier System (Amersham). The purified protein fractions were
pooled and concentrated to ~15 mg/ml by ultrafiltration using the Amicon technology (Millipore,
California, USA). For phopholipase activity assays, the purified Tle1 and Tle1S197A
proteins were
concentrated to 0.6 mg/mL.
Purification of the Tle1EAEC
-Tli1EAEC
complex
Purified Tle1EAEC
and Tli EAEC
were mixed together in a molar ratio of 1:1.2. The complex was
purified by gel filtration chromatography (Superdex 75, 10/30 GE Healthcare) equilibrated in a
Tris-HCl 20 mM pH 8.0, NaCl 150 mM buffer using an AKTA purifier System (Amersham). The
fractions containing the complex were pooled and concentrated to ~15 mg/mL as described above.
MALS/QELS/UV/RI-coupled size exclusion chromatography
Size exclusion chromatography (SEC) was performed on an Alliance 2695 HPLC system (Waters)
using a pre-calibrated KW802.5 column (Shodex) run in Hepes 25 mM pH 7.3, NaCl 250 mM at
0.5 ml/min. MALS, UV spectrophotometry, QELS and RI were achieved with MiniDawn Treos
(Wyatt Technology), a Photo Diode Array 2996 (Waters), a DynaPro (Wyatt Technology) and an
Optilab rEX (Wyatt Technology), respectively, as described (Sciara et al., 2008). Mass and
hydrodynamic radius calculation was done with ASTRA software (Wyatt Technology) using a
dn/dc value of 0.185 mL/g.
Biolayer interferometry (BLI)
The purified Tli1EAEC
protein was first biotinylated using the EZ-Link NHS-PEG4-Biotin kit
(Perbio Science, France). The reaction was stopped by removing the excess of biotin using a Zeba
60
Spin Desalting column (Perbio Science, France). BLI studies were performed in black 96-well
plates (Greiner) at 25°C using an OctetRed96 (ForteBio, USA). Streptavidin biosensor tips
(ForteBio, USA) were first hydrated with 0.2 mL Kinetic Buffer (KB, ForteBio, USA) for 20 min
and then loaded with biotinylated Tli1EAEC
(10 g/mL in KB). The association of Tli1EAEC
with
various concentrations of Tle1 (0.4 nM, 1 nM, 2.56 nM, 6.4 nM, 16 nM and 40 nM) was
monitored for 600 sec, and the dissociation was followed for 1,800 sec in KB.
Phospholipase A1 and A2 fluorescent assays
Phospholipase activities of Tle1EAEC
and Tle1S197A
were performed using fluorogenic phospholipid
substrates. Phospholipase A1 and A2 activities were monitored continuously using BODIPY® dye-
labeled phospholipids: PED-A1 (N-((6-(2,4-DNP)Amino)Hexanoyl)-1-(BODIPY® FL C5)-2-
Hexyl-sn-Glycero-3-Phosphoethanolamine) and red/green BODIPY® PC-A2 (1-O-(6-BODIPY-
®558⁄568-Aminohexyl)-2-BODIPY®FLC5-sn-Glycero-3-Phosphocholine), respectively (Darrow
et al., 2011; Farber et al., 2001). The sn-2 fatty acyl group in PED-A1 is a non-hydrolyzable alkyl
chain, and PED-A1 substrate was used to specifically measure the PLA1 activity. The red/green
BODIPY® PC-A2 has a sn-1 uncleavable alkyl chain. Substrate stock solutions (50 µM) were
prepared in ethanol. All enzyme activities were assayed in Tris-HCl 10 mM pH 8.0, NaCl 150 mM,
CaCl2 1 mM and Triton X-100 0.1%. Enzymatic reactions were performed at 20°C for 25 min in a
final volume of 200 µL containing 20 µg of Tle1 purified protein (from a 0.6 mg.mL-1
stock
solution) and 5 µM of the substrate. During pilot studies, we noted that concentrations of the
purified Tle1 protein above 1mg.mL-1
led to a decrease in the measured Tle1 PLA1 specific
activity. The release of BODIPY®
(BFCL5) (Life Technologies) was recorded at λexc = 485 nm
and λem = 538 nm using a 96-well plate fluorometer (Fluoroskan ascent, Thermoscientific).
Enzymatic activities were quantified using a BFCL5 calibration curve (0.08 - 200 pmoles in
activity buffers) and expressed in pmol of fatty acid (or BFLC5) released per minute per mg of
protein (pmol min‑ 1
mg-1
). PLA1 from Thermomyces lanuginosus and Bee venom PLA2 (Sigma-
Aldrich, Saint-Quentin Fallavier, France) were used as positive standards for PLA1 and PLA2
activities, respectively. Inhibition studies with Tli1EAEC
were performed by incubating 20 µg of
Tle1EAEC
with various molar ratio of Tli1EAEC
(xI=0.125, 0.25, 0.5, 1 and 2). The residual activity
was measured as described above.
Activities on phospholipid monolayer films
61
All experiments were performed using the KSV5000 system (KSV, Helsinki, Finland) equipped
with a Langmuir film balance to measure the surface pressure (Π) and monitored by the KSV
Device Server Software v.3.50 running under Windows 7® as previously described (Point et al.,
2013). A Teflon “zero-order” trough (Verger and de Haas, 1973) was filled with Tris-HCl 10 mM
pH 8.0, NaCl 100 mM, CaCl2 21 mM and EDTA 1 mM. The phospholipid monolayer was formed
at the desired surface pressure (Π) of 20 mN.m-1
by spreading a few microliters of a phospholipid
solution (1 mg/mL in chloroform containing 0.4% v/v methanol) and further incubation for > 10
min (chloroform evaporation). Hydrolysis rate measurements were performed with a Teflon “zero-
order” trough with two compartments: a reaction compartment (volume 43 mL; surface area, 38.5
cm2) and a reservoir compartment (volume 203 mL; surface area, 156.5 cm
2) connected to each
other by a small surface channel. The purified Tle1EAEC
protein was injected into the subphase of
the reaction compartment only (11 nM enzyme final concentration), whereas the phospholipid
lipid film spread at the air-water interface covers both of them. When using medium chain
phospholipids, such as DLPC, DLPG, DLPS or DLPE, soluble lipolysis products are released
upon the action of phospholipase and a drop in surface pressure can be recorded. Using the
barostat mode available on the KSV5000 instrument, an automatically driven Teflon mobile
barrier can be moved over the reservoir to compress the phospholipid film and compensate for the
substrate molecules removed from the film by the enzyme hydrolysis, thus keeping the surface
pressure constant (here at Π = 20 mN.m-1
). The kinetics of hydrolysis were recorded for at least 60
min and PLA activities (mol.cm‑ 2
.min-1
.M-1
) were expressed as the number of moles of substrate
hydrolyzed per time unit [139] and per surface unit (cm2) of the reaction compartment of the
“zero-order” trough and for an arbitrary enzyme concentration of 1 M (de la Fournière et al.,
1994).
Globomycin treatment.
EAEC cells producing HA-tagged Tli1 from pOK- Tli1HA were grown to an optical density at 600
nm (OD600) of ~ 0.6 prior to the addition of 50 g/mL of globomycin. After 10 min of treatment,
IPTG was then added at a final concentration of 100 M, and cells were further incubated for 30
min at 37°C. Cells were harvested, and samples were analyzed by SDS-PAGE and
immunoblotting.
Escherichia coli K-12 toxicity assays
62
Cells were grown in LB at 37°C for 16 hours. Bacterial suspensions were normalized to an OD600
of 2, serially diluted and 15 µL drops of each dilution were spotted onto selective LB agar plates
containing or not arabinose 0.2%.
Co-immunoprecipitation experiments
One hundred mL of W3110 cells producing the proteins of interest from independent plasmids
were grown to and OD600 ~ 0.4 and the expression of the cloned genes was induced with IPTG
100 M or arabinose 0.2% for one hour. The cells were harvested and the pellets were frozen in
liquid nitrogen and stored at -80°C for 1 hour. Pellets were then resuspended in Tris-HCl 20 mM
pH 8.0, NaCl 100 mM, sucrose 30%, lysozyme 100 g/mL, EDTA 1 mM, DNase 100 μg/mL,
RNase 100 μg/mL, Complete protease inhibitor cocktail (Roche) to an OD600 ~ 80 and incubated
on ice for 15 min. An equal volume of Tris-HCl 20 mM pH 8.0, NaCl 100 mM, MgCl2 5 mM was
then added, and the cells were lysed by two passages at the French Press (800 psi). Lysates were
clarified by centrifugation at 13,000 × g for 10 min. Supernatants were used for co-
immunoprecipitation using Anti-FLAG®
M2 affinity gel (Sigma-Aldrich). After 3 hours of
incubation, the beads were washed twice with 1 mL of Tris-HCl 20 mM pH 8.0, NaCl 100 mM,
sucrose 15%, and once with Tris-HCl 20 mM pH 8.0, NaCl 100 mM. Beads were recovered and
resuspended in 25 μL of SDS-loading buffer and heated for 10 min at 96°C prior to SDS-PAGE
and Western-blot analyses.
Miscellaneous
Fractionation, sedimentation sucrose gradient assays, NADH oxidase activity measurements and
Hcp release assay have been performed as previously described (Aschtgen et al., 2008; Aschtgen
et al., 2010). SDS-Polyacrylamide gel electrophoresis was performed using standard protocols.
For immunostaining, proteins were transferred onto 0.2 µm nitrocellulose membranes (Amersham
Protran), and immunoblots were probed with primary antibodies (see below) and goat secondary
antibodies coupled to alkaline phosphatase and developed in alkaline buffer in presence of 5-
bromo-4-chloro-3-indolylphosphate and nitroblue tetrazolium. The anti-TolB, anti-TolA, anti-
OmpA and anti-OmpF polyclonal antibodies are from our laboratory collection, while the anti-HA
(3F10 clone, Roche), anti-FLAG (M2 clone, Sigma Aldrich), anti-EF-Tu (Roche) and anti-VSVG
(Sigma-Aldrich) monoclonal antibodies and alkaline phosphatase-conjugated goat anti-rabbit,
mouse or rat secondary antibodies (Millipore) have been purchased as indicated.
63
Acknowledgments
We thank Dr. Danièle Cavard for providing globomycin, Steve Garvis for help with the
Caenorhabditis elegans model. We thank Emmanuelle Bouveret, James Sturgis, Abdelrahim
Zoued and the members of the Cascales, Lloubès, Bouveret and Sturgis research groups for
insightful discussions, Annick Brun, Isabelle Bringer and Olivier Uderso for technical assistance.
Work in E.C. laboratory is supported by the Centre National de la Recherche Scientifique (CNRS),
the Aix-Marseille Université and grants from the Agence Nationale de la Recherche (ANR-10-
JCJC-1303-03, ANR-14-CE14-006-02). Work in C.C. laboratory is supported by the CNRS, the
Aix-Marseille Université and by grants from the Marseille-Nice Genopole, IBiSA and the
Fondation de la Recherche Médicale (FRM DEQ2011-0421282). Ph.D studies of N.F., T.T.H.L.
and V.S.N. are supported by the ANR-14-CE14-0006-02 grant, a fellowship from the University
of Sciences and Techniques of Hanoi (USTH) and a fellowship from the French Embassy in
Vietnam, respectively. M.S.A. was supported by a Ph.D fellowship from the French Ministry of
Research.
REFERENCES
Alcoforado Diniz, J., and Coulthurst, S.J. (2015) Intraspecies competition in Serratia marcescens
is mediated by Type VI-secreted Rhs effectors and a conserved effector-associated accessory
protein. J Bacteriol. 197: 2350-60.
Alcoforado Diniz, J., Liu, Y.C., Coulthurst, S.J. (2015) Molecular weaponry: diverse effectors
delivered by the Type VI secretion system. Cell Microbiol. doi: 10.1111/cmi.12532.
Aschtgen, M.S., Bernard, C.S., de Bentzmann, S., Lloubes, R., and Cascales, E. (2008) SciN is an
outer membrane lipoprotein required for type VI secretion in enteroaggregative Escherichia coli. J.
Bacteriol. 190: 7523-7531.
Aschtgen, M.S., Gavioli, M., Dessen, A., Lloubes, R., and Cascales, E. (2010) The SciZ protein
anchors the enteroaggregative Escherichia coli Type VI secretion system to the cell wall. Mol.
Microbiol. 75: 886-899.
Ballister, E.R., Lai, A.H., Zuckermann, R.N., Cheng, Y., and Mougous, J.D. (2008) In vitro self-
assembly of tailorable nanotubes from a simple protein building block. Proc. Natl. Acad. Sci. U. S.
64
A. 105: 3733-3738.
Basler, M. (2015) Type VI secretion system: secretion by a contractile nanomachine. Philos Trans
R Soc Lond B Biol Sci 370: 20150021.
Basler, M., Pilhofer, M., Henderson, G.P., Jensen, G.J., and Mekalanos, J.J. (2012) Type VI
secretion requires a dynamic contractile phage tail-like structure. Nature 483: 182-186.
Battesti, A., and Bouveret, E. (2012) The bacterial two-hybrid system based on adenylate cyclase
reconstitution in Escherichia coli. Methods 58: 325-334.
Benz, J., and Meinhart, A. (2014) Antibacterial effector/immunity systems: it's just the tip of the
iceberg. Curr Opin Microbiol 17: 1-10.
Bonemann, G., Pietrosiuk, A., Diemand, A., Zentgraf, H., and Mogk, A. (2009) Remodelling of
VipA/VipB tubules by ClpV-mediated threading is crucial for type VI protein secretion. EMBO J.
28: 315-325.
Bonemann, G., Pietrosiuk, A., and Mogk, A. (2010) Tubules and donuts: a type VI secretion story.
Mol. Microbiol. 76: 815-821.
Boyer, F., Fichant, G., Berthod, J., Vandenbrouck, Y., and Attree, I. (2009) Dissecting the
bacterial type VI secretion system by a genome wide in silico analysis: what can be learned from
available microbial genomic resources? BMC Genomics 10: 104.
Brunet, Y.R., Bernard, C.S., Gavioli, M., Lloubes, R., and Cascales, E. (2011) An epigenetic
switch involving overlapping fur and DNA methylation optimizes expression of a type VI
secretion gene cluster. PLoS Genet. 7: e1002205.
Brunet, Y.R., Espinosa, L., Harchouni, S., Mignot, T., and Cascales, E. (2013) Imaging type VI
secretion-mediated bacterial killing. Cell Rep. 3: 36-41.
Brunet, Y.R., Henin, J., Celia, H., and Cascales, E. (2014) Type VI secretion and bacteriophage
tail tubes share a common assembly pathway. EMBO Rep 15: 315-321.
Brunet, Y.R., Zoued, A., Boyer, F., Douzi, B., and Cascales, E. (2015) The Type VI secretion
TssEFGK-VgrG phage-like baseplate is recruited to the TssJLM membrane complex via multiple
contacts and serves as assembly platform for tail tube/sheath polymerization. PLoS Genet. 11:
e1005545.
Carruthers, M.D., Nicholson, P.A., Tracy, E.N., and Munson, R.S., Jr. (2013) Acinetobacter
baumannii utilizes a type VI secretion system for bacterial competition. PLoS One 8: e59388.
65
Cascales, E. (2008) The type VI secretion toolkit. EMBO Rep. 9: 735-741.
Coulthurst, S.J. (2013) The Type VI secretion system - a widespread and versatile cell targeting
system. Res Microbiol 164: 640-654.
Darrow, A.L., Olson, M.W., Xin, H., Burke, S.L., Smith, C., Schalk-Hihi, C., et al. (2011) A
novel fluorogenic substrate for the measurement of endothelial lipase activity. J Lipid Res 52:
374-382.
Datsenko, K.A., and Wanner, B.L. (2000) One-step inactivation of chromosomal genes in
Escherichia coli K-12 using PCR products. Proc. Natl. Acad. Sci. U.S.A. 97: 6640-6645.
de la Fournière, L., Ivanova, M.G., Blond, J.-P., Carrière, F., Verger, R. (1994) Surface behaviour
of human pancreatic and gastric lipases. Colloids Surf B. 2: 585-593.
Dereeper, A., Guignon, V., Blanc, G., Audic, S., Buffet, S., Chevenet, F., et al. (2008)
Phylogeny.fr: robust phylogenetic analysis for the non-specialist. Nucleic Acids Res. 36: W465-
469.
Ding, J., Wang, W., Feng, H., Zhang, Y., and Wang, D.C. (2012) Structural insights into the
Pseudomonas aeruginosa type VI virulence effector Tse1 bacteriolysis and self-protection
mechanisms. J Biol Chem 287: 26911-26920
Dong, T.G., Ho, B.T., Yoder-Himes, D.R., and Mekalanos, J.J. (2013) Identification of T6SS-
dependent effector and immunity proteins by Tn-seq in Vibrio cholerae. Proc Natl Acad Sci U S A
110: 2623-2628.
Douzi, B., Spinelli, S., Blangy, S., Roussel, A., Durand, E., Brunet, Y.R., et al. (2014) Crystal
structure and self-interaction of the type VI secretion tail-tube protein from enteroaggregative
Escherichia coli. PLoS One 9: e86918.
Durand, E., Derrez, E., Audoly, G., Spinelli, S., Ortiz-Lombardia, M., Raoult, D., et al. (2012)
Crystal structure of the VgrG1 actin cross-linking domain of the Vibrio cholerae type VI secretion
system. J Biol Chem 287: 38190-38199.
Durand, E., Cambillau, C., Cascales, E., and Journet, L. (2014) VgrG, Tae, Tle, and beyond: the
versatile arsenal of Type VI secretion effectors. Trends Microbiol 22: 498-507.
Durand, E., Nguyen, V.S., Zoued, A., Logger, L., Péhau-Arnaudet, G., Aschtgen, M.S, et al.
(2015) Biogenesis and structure of a type VI secretion membrane core complex. Nature 523: 555-
60.
66
Egan, F., Reen, F.J., and O'Gara, F. (2015) Tle distribution and diversity in metagenomic datasets
reveal niche specialization. Environ Microbiol Rep 7: 194-203.
Emsley, P., Lohkamp, B., Scott, W.G., and Cowtan, K. (2010) Features and development of Coot.
Acta Crystallogr D Biol Crystallogr 66: 486-501.
Farber, S.A., Pack, M., Ho, S.Y., Johnson, I.D., Wagner, D.S., Dosch, R., et al. (2001) Genetic
analysis of digestive physiology using fluorescent phospholipid reporters. Science 292: 1385-1388.
Finn, R.D., Clements, J., Eddy, S.R. (2011) HMMER web server: interactive sequence similarity
searching. Nucl Acids Res 39: W29-37.
Gueguen, E., and Cascales, E. (2013) Promoter swapping unveils the role of the Citrobacter
rodentium CTS1 type VI secretion system in interbacterial competition. Appl Environ Microbiol
79: 32-38.
Hachani, A., Allsopp, L.P., Oduko, Y., and Filloux, A. (2014) The VgrG proteins are "a la carte"
delivery systems for bacterial type VI effectors. J Biol Chem 289: 17872-17884.
Ho, B.T., Dong, T.G., and Mekalanos, J.J. (2014) A view to a kill: the bacterial type VI secretion
system. Cell Host Microbe 15: 9-21.
Hood, R.D., Singh, P., Hsu, F., Guvener, T., Carl, M.A., Trinidad, R.R., et al. (2010) A type VI
secretion system of Pseudomonas aeruginosa targets a toxin to bacteria. Cell Host Microbe 7: 25-
37.
Hu, H., Zhang, H., Gao, Z., Wang, D., Liu, G., Xu, J., et al. (2014) Structure of the type VI
secretion phospholipase effector Tle1 provides insight into its hydrolysis and membrane targeting.
Acta Crystallogr D Biol Crystallogr 70: 2175-2185.
Hullmann, J., Patzer, S.I., Romer, C., Hantke, K., and Braun, V. (2008) Periplasmic chaperone
FkpA is essential for imported colicin M toxicity. Mol Microbiol 69: 926-937.
Jiang, F., Waterfield, N.R., Yang, J., Yang, G., and Jin, Q. (2014) A Pseudomonas aeruginosa type
VI secretion phospholipase D effector targets both prokaryotic and eukaryotic cells. Cell Host
Microbe 15: 600-610.
Kapitein, N., Bonemann, G., Pietrosiuk, A., Seyffer, F., Hausser, I., Locker, J.K., and Mogk, A.
(2013) ClpV recycles VipA/VipB tubules and prevents non-productive tubule formation to ensure
efficient type VI protein secretion. Mol. Microbiol. 87: 1013-1028.
Karimova, G., Pidoux, J., Ullmann, A., and Ladant, D. (1998) A bacterial two-hybrid system
67
based on a reconstituted signal transduction pathway. Proc. Natl. Acad. Sci. U.S.A. 95: 5752-5756.
Kelley, L.A., and Sternberg, M.J. (2009) Protein structure prediction on the Web: a case study
using the Phyre server. Nat Protoc 4: 363–371.
Kudryashev, M., Wang, R.Y., Brackmann, M., Scherer, S., Maier, T., Baker, D., et al. (2015)
Structure of the type VI secretion system contractile sheath. Cell 160: 952-62.
Leiman, P.G., Basler, M., Ramagopal, U.A., Bonanno, J.B., Sauder, J.M., Pukatzki, S., et al.
(2009) Type VI secretion apparatus and phage tail-associated protein complexes share a common
evolutionary origin. Proc. Natl. Acad. Sci. U.S.A. 106: 4154-4159.
Lery, L.M., Frangeul, L., Tomas, A., Passet, V., Almeida, A.S., Bialek-Davenet, S., et al. (2014)
Comparative analysis of Klebsiella pneumoniae genomes identifies a phospholipase D family
protein as a novel virulence factor. BMC Biol 12: 41.
Liang, X., Moore, R., Wilton, M., Wong, M.J., Lam, L., and Dong, T.G. (2015) Identification of
divergent type VI secretion effectors using a conserved chaperone domain. Proc Natl Acad Sci U S
A 112: 9106-9111
Lu, D., Zheng, Y., Liao, N., Wei, L., Xu, B., Liu, X., and Liu, J. (2014) The structural basis of the
Tle4-Tli4 complex reveals the self-protection mechanism of H2-T6SS in Pseudomonas aeruginosa.
Acta Crystallogr D Biol Crystallogr 70: 3233-3243.
Ma, J., Bao, Y., Sun, M., Dong, W., Pan, Z., Zhang, W., Lu, C. and Yao, H. (2014) Two
functional type VI secretion systems in avian pathogenic Escherichia coli are involved in different
pathogenic pathways. Infect Immun 82: 3867-3879
MacIntyre, D.L., Miyata, S.T., Kitaoka, M., and Pukatzki, S. (2010) The Vibrio cholerae type VI
secretion system displays antimicrobial properties. Proc Natl Acad Sci U S A 107: 19520-19524.
Mougous, J.D., Cuff, M.E., Raunser, S., Shen, A., Zhou, M., Gifford, C.A., et al. (2006) A
virulence locus of Pseudomonas aeruginosa encodes a protein secretion apparatus. Science 312:
1526-1530.
Murdoch, S.L., Trunk, K., English, G., Fritsch, M.J., Pourkarimi, E., and Coulthurst, S.J. (2011)
The opportunistic pathogen Serratia marcescens utilizes type VI secretion to target bacterial
competitors. J Bacteriol 193: 6057-6069.
Pell, L.G., Kanelis, V., Donaldson, L.W., Howell, P.L., and Davidson, A.R. (2009) The phage
lambda major tail protein structure reveals a common evolution for long-tailed phages and the type
VI bacterial secretion system. Proc. Natl. Acad. Sci. U.S.A. 106: 4160-4165.
68
Point, V., Benarouche, A., Jemel, I., Parsiegla, G., Lambeau, G., Carriere, F., and Cavalier, J.F.
(2013) Effects of the propeptide of group X secreted phospholipase A(2) on substrate specificity
and interfacial activity on phospholipid monolayers. Biochimie 95: 51-58.
Pukatzki, S., Ma, A.T., Sturtevant, D., Krastins, B., Sarracino, D., Nelson, W.C., et al. (2006)
Identification of a conserved bacterial protein secretion system in Vibrio cholerae using the
Dictyostelium host model system. Proc. Natl. Acad. Sci. U.S.A. 103: 1528-1533.
Pukatzki, S., Ma, A.T., Revel, A.T., Sturtevant, D., and Mekalanos, J.J. (2007) Type VI secretion
system translocates a phage tail spike-like protein into target cells where it cross-links actin. Proc
Natl Acad Sci U S A 104: 15508-15513.
Russell, A.B., LeRoux, M., Hathazi, K., Agnello, D.M., Ishikawa, T., Wiggins, P.A., et al. (2013)
Diverse type VI secretion phospholipases are functionally plastic antibacterial effectors. Nature
496: 508-512.
Russell, A.B., Peterson, S.B., and Mougous, J.D. (2014) Type VI secretion system effectors:
poisons with a purpose. Nat Rev Microbiol 12: 137-148.
Schwarz, S., West, T.E., Boyer, F., Chiang, W.C., Carl, M.A., Hood, R.D., Rohmer, L., et al.
(2010) Burkholderia Type VI Secretion Systems Have Distinct Roles in Eukaryotic and Bacterial
Cell Interactions. PLoS Pathog 6: e1001068.
Sciara, G., Blangy, S., Siponen, M., Mc Grath, S., van Sinderen, D., Tegoni, M., et al. (2008) A
topological model of the baseplate of lactococcal phage Tuc2009. J. Biol. Chem. 283: 2716-2723.
Shang, G., Liu, X., Lu, D., Zhang, J., Li, N., Zhu, C. et al. (2012) Structural insight into how
Pseudomonas aeruginosa peptidoglycan hydrolase Tse1 and its immunity protein Tsi1 function.
Biochem J 448: 201-211
Shneider, M.M., Buth, S.A., Ho, B.T., Basler, M., Mekalanos, J.J., and Leiman, P.G. (2013)
PAAR-repeat proteins sharpen and diversify the type VI secretion system spike. Nature 500: 350-
353.
Silverman, J.M., Brunet, Y.R., Cascales, E., and Mougous, J.D. (2012) Structure and regulation of
the type VI secretion system. Annu. Rev. Microbiol. 66: 453-472.
Silverman, J.M., Agnello, D.M., Zheng, H., Andrews, B.T., Li, M., Catalano, C.E., Gonen, T., and
Mougous, J.D. (2013) Haemolysin coregulated protein is an exported receptor and chaperone of
type VI secretion substrates. Mol Cell 51: 584-593.
Spencer, C., and Brown, H.A. (2015) Biochemical Characterization of a Pseudomonas aeruginosa
69
Phospholipase D. Biochemistry 54: 1208-1218.
Suarez, G., Sierra, J.C., Erova, T.E., Sha, J., Horneman, A.J., and Chopra, A.K. (2010) A type VI
secretion system effector protein, VgrG1, from Aeromonas hydrophila that induces host cell
toxicity by ADP ribosylation of actin. J Bacteriol 192: 155-168.
Uchida, K., Leiman, P.G., Arisaka, F., and Kanamaru, S. (2014) Structure and properties of the C-
terminal beta-helical domain of VgrG protein from Escherichia coli O157. J Biochem 155: 173-
182.
Unterweger, D., Kostiuk, B., Ötjengerdes, R., Wilton, A., Diaz-Satizabal, L. and Pukatzki, S.
(2015) Chimeric adaptor proteins translocate diverse type VI secretion system effectors in Vibrio
cholerae. EMBO J. 34: 2198-2210.
van den Ent, F., and Lowe, J. (2006) RF cloning: a restriction-free method for inserting target
genes into plasmids. J. Biochem. Biophys. Methods 67: 67-74.
Verger, R., and de Haas, G.H. (1973) Enzyme reactions in a membrane model. 1: A new technique
to study enzyme reactions in monolayers. Chem Phys Lipids 10: 127-136.
Whitney, J.C., Beck, C.M., Goo, Y.A., Russell, A.B., Harding, B.N., De Leon, J.A., et al. (2014)
Genetically distinct pathways guide effector export through the type VI secretion system. Mol
Microbiol 92: 529-542.
Whitney, J.C, Quentin, D., Sawai, S., LeRoux, M, Harding, B.N, Ledvina, H.E. et al. (2015) An
interbacterial NAD(P)(+) glycohydrolase toxin requires elongation factor Tu for delivery to target
cells. Cell 163: 607-619
Zaslaver, A., Bren, A., Ronen, M., Itzkovitz, S., Kikoin, I., Shavit, S., et al. (2006) A
comprehensive library of fluorescent transcriptional reporters for Escherichia coli. Nat Methods 3:
623-628.
Zoued, A., Brunet, Y.R., Durand, E., Aschtgen, M.S., Logger, L., Douzi, B., et al. (2014)
Architecture and assembly of the Type VI secretion system. Biochim Biophys Acta 1843: 1664-
1673.
Zuckert, W.R. (2014) Secretion of bacterial lipoproteins: through the cytoplasmic membrane, the
periplasm and beyond. Biochim Biophys Acta 1843: 1509-1516.
70
Table 1. Rates of hydrolysis of DLPC, DLPE, DLPS and DLPG monolayers at a
constant surface pressure of 20 mN m-1
.
Phospholipid
Substrates
enzyme activity
(µmole cm-2
min-1
M-1
)
Tle1 Tle1S197A Tle1 + Tli1
(1:2 mol/mol)
DLPC 148 (± 9) no activity no activity
DLPE 109 (± 5) no activity no activity
DLPS 97 (± 4) no activity no activity
DLPG no activity no activity no activity
Assays were carried out in a ‘zero order’ trough as described in the Experimental Procedures
section. The final concentration of Tle1 and Tle1S197A was 11 nM for the lipolysis of each
phospholipid. Enzyme activities are expressed as the number of moles of substrate hydrolyzed
by time unit and surface unit of the reaction compartment of the “zero order” trough for an
arbitrary lipase concentration of 1 M. All data are presented as mean values standard
deviations of at least 2 independent assays (CV% < 6%). Buffer: 10 mM Tris (pH 8.0), 100
mM NaCl, 21 mM CaCl2 and 1 mM EDTA.
71
FIGURE LEGENDS
Figure 1. The sci-1 T6SS gene cluster
contributes to EAEC antibacterial activity.
A. The EAEC sci-1 T6SS is not required for
virulence towards the C. elegansmodel of
infection. The number of nematods surviving on
lawn of the indicated strain (closed squares, OP50;
closed circles, EAEC 17-2; open circles,
EAEC ΔT6SS-1; closed triangles, B.
cenocepacia K56-2) was plotted as percentage (%
survival) over time (days).
B. Antibacterial assay. Prey cells (W3110 gfp+,
kanR) were mixed with the indicated attacker cells,
spotted onto sci-1-inducing medium [137] agar
plates and incubated for 4 h at 37°C. The image of
a representative bacterial spot is shown and the
relative fluorescent levels (in AU) are indicated in
the upper graph. The number of recovered E.
coli prey cells is indicated in the lower graph (in
log10 of cfu). The circles indicate values from
three independent assays, and the average is
indicated by the bar.
72
Figure 2. EC042_4534 phospholipase activity is responsible for Sci-1-mediated antibacterial
activity. (A) Schematic representation of the EAEC 17-2 sci-1 T6SS gene cluster. The numbers
on top refer to the genes locus tag (EC042_XXXX). Genes encoding core components (identified
by their names on bottom) are colored grey. Genes of unknown function are indicated in white.
The EC042_4534 gene encoding Tle1 is indicated in green whereas the two genes encoding the
Tli1 and Tli1b immunity proteins are colored in blue. The asterisk below Tli1b indicates that a
frameshift mutation on the 17-2 chromosome yields a truncated protein (Supplemental Figure S2).
(B) Homology-based structural model of EC042_4534 based on the crystal structure of the P.
aeruginosa Tle1PA
effector (PDB: 4O5P). The magnification highlights the positions and
orientations of the putative catalytic triad amino-acid side chains (Ser-197, His-245 and Asp-310).
(C) Specific phospholipase A1 (PLA1) and A2 (PLA2) activity measurements of the EC042_4534
Tle1EAEC
WT and S197A mutant proteins using fluorescent phospholipids. Specific activities were
calculated from the velocity slope obtained for 25 minutes using 20 µg of purified protein. Data
are expressed as mean values ± standard deviations of three independent assays (CV% < 5%). The
SDS-PAGE analysis of the purified EC042_4534 (Tle1WT
) and S197A (Tle1S197A
) proteins (40 μg)
73
after Coomassie blue staining is shown in the inset (molecular weight markers (in kDa) are
indicated on the right). (D) EC042_4534 (Tle1EAEC
) phospholipase activity is responsible for Sci-
1-mediated antibacterial activity. The anti-bacterial activity was assessed by mixing W3110 gfp+
prey cells with the indicated attacker cells for 4 hours at 37°C in sci-1-inducing medium
containing 0.02 % arabinose (WT (EAEC 17-2), ∆T6SS-1 (sci-1 gene cluster deletion derivative)
and ∆tle1-tli1 (17-2 deleted of the tle1 and tli1 genes) carrying the pBAD18 empty vector, or
∆tle1-tli1 producing wild-type Tle1 (Tle1WT
) and Tli1 or the S197A Tle1 mutant (Tle1S197A
) and
Tli1 from the pBAD18-Tle1-Tli1 or pBAD18-Tle1S197A
-Tli1, respectively). The image of a
representative bacterial colony is shown and the relative fluorescent levels (in arbitrary units, AU)
are indicated in the upper graph. The number of recovered E. coli prey cells is indicated in the
lower graph (in log10 of colony-forming units (cfu)). The circles indicate values from three
independent assays, and the average is indicated by the bar.
74
Figure 3. Tli1EAEC
inhibits Tle1EAEC
anti-bacterial phospholipase activity. (A) Anti-bacterial
assay. The anti-bacterial activity was assessed by mixing W3110 gfp+ prey cells producing
(pBAD-Tli1EAEC
) or not (pBAD18) the EC042_4535 (Tli1EAEC
) protein from the pBAD promoter
with the indicated attacker cells for 4 hours at 37°C in sci-1-inducing medium containing 0.02%
arabinose. The image of a representative bacterial colony is shown and the relative fluorescent
levels (in arbitrary units, AU) are indicated in the upper graph. The number of recovered E. coli
prey cells is indicated in the lower graph (in log of colony-forming units (cfu)). The circles
indicate values from three independent assays, and the average is indicated by the bar. (B) The
anti-bacterial activity was assessed by mixing ∆tli1-tli1b (17-2 deleted of the tli1 and tli1b genes)
gfp+ prey cells producing (pBAD-Tli1
EAEC) or not (pBAD18) the EC042_4535 (Tli1
EAEC) protein
from the pBAD promoter with the indicated attacker cells for 4 hours at 37°C in sci-1-inducing
medium containing 0.02% arabinose. The image of a representative bacterial colony is shown and
the relative fluorescent levels (in arbitrary units, AU) are indicated in the upper graph. The number
of recovered E. coli prey cells is indicated in the lower graph (in log10 of colony-forming units
(cfu)). The circles indicate values from three independent assays, and the average is indicated by
the bar. (C) Tli1EAEC
inhibition of Tle1EAEC
PLA1 activity. The rate of hydrolysis of PED-A1 by
purified Tle1EAEC
at 20°C in presence of increasing concentrations of Tli1EAEC
was plotted against
the molar excess of Tli1EAEC
. The SDS-PAGE analysis of the purified Tli1EAEC
(Tli1) protein after
Coomassie blue staining is shown in the inset (molecular weight markers (in kDa) are indicated on
the right).
75
Figure 4. Tli1EAEC
binds Tle1EAEC
with nanomolar affinity. (A) Bacterial two-hybrid analysis.
BTH101 reporter cells producing the indicated Tle1EAEC
(Tle1) or Tli1EAEC
(Tli1) proteins fused to
the T18 or T25 domain of the Bordetella adenylate cyclase were spotted on X-Gal indicator plates.
The blue color of the colony reflects the interaction between the two proteins. Controls include
T18 and T25 fusions to TolB and Pal, two proteins that interact but unrelated to the T6SS. (B) Gel
filtration analysis on a calibrated Superdex 75 (10/30) column. The purified Tli1EAEC
(left panel),
Tle1EAEC
(middle panel) proteins and the Tle1EAEC
/Tli1EAEC
mixture (right panel) were separated
by gel filtration. Tli1EAEC
, Tle1EAEC
and the Tle1EAEC
-Tli1EAEC
complex eluted at 12.8 ml (~ 24
kDa), 10.37 ml (~ 66 kDa) and 9.8 ml (~ 82 kDa) respectively. The inset is the SDS-PAGE and
Coomassie blue staining analysis of the fractions eluted after Tle1EAEC
-Tli1EAEC
complex
separation. Peak 1 contains the two proteins whereas peak 2 (13.1 ml) contains the excess of
Tli1EAEC
. The molecular weight markers (in kDa) are indicated on the right. (C)
MALS/SEC/UV/RI analysis. The UV absorbance at 280 nm corresponding to Tli1EAEC
(blue line),
Tle1EAEC
(orange line) and to the Tle1EAEC
-Tli1EAEC
complex (red line) was plotted against time
(min. after sample injection in the High Performance Liquid Chromatography system). The traces
indicating the molar mass (indicated on the left, in Da) are shown on each peak. (D) Bio-layer
interferometry analysis. Recordings of the binding of purified Tle1EAEC
(concentrations (in nM)
indicated above the corresponding curve) to the streptavidin chip coupled to biotinylated Tli1EAEC
.
The response (in nm) is plotted versus the time (in sec.). The experimental association and
dissociation curves (blue) are compared to the simulated ones [8]. The calculated KD value is 1.50
± 0.05 nM.
76
Figure 5. Sub-cellular localizations of Tle1EAEC
and Tli1EAEC
. (A) Tle1EAEC
is a cytoplasmic
protein. Total ∆tle1-tli1 cells producing VSVG-tagged Tle1EAEC
(Tle1VSVG) (T) were fractionated
to isolate the periplasmic (P), cytoplasmic (C) and membrane fractions (M). Proteins from 109 (T,
M) and 2×109 (P, C) cells were separated by SDS-PAGE and immunodetected with anti-VSVG
monoclonal (Tle1VSVG), anti-EF-Tu (cytoplasmic marker), TolB (periplasmic marker), and TolA
(membrane marker) antibodies. The position of the immunodetected proteins is indicated on the
right. The molecular weight markers (in kDa) are indicated on the left. (B) Tli1EAEC
bears a
characteristic lipobox motif. ClustalW (T-Coffee) sequence alignment of the N-terminal region of
the Tli1EAEC
protein with that of representative homologous proteins (DUF2931 containing
proteins) identified by HMMER analysis (Finn et al., 2011). The putative lipobox motif (L-
[G/A/S]-[G/A/S]-C) is indicated by a blue box. The position of the N-terminal cysteine residue of
the processed form is underlined in green. (C) Tli1EAEC
processing is dependent on signal
peptidase II. EAEC ∆tli1-tli1b 2×109 cells producing HA-tagged Tli1
EAEC (Tli1
HA) treated (+) or
not (-) with the signal peptidase II inhibitor antibiotic globomycin were subjected to SDS-PAGE
77
and immunodetection with the anti-HA and anti-OmpA antibodies. The unprocessed form of
Tli1HA
is indicated by an asterisk. The molecular weight markers (in kDa) are indicated on the left.
(D) Tli1EAEC
is an outer membrane protein. Total membrane from ∆tli1-tli1b cells producing
FLAG-tagged Tli1EAEC
(Tli1FL) were separated on a discontinuous sedimentation sucrose gradient.
The collected fractions were subjected to measurements of the NADH oxidase activity (inner
membrane marker, represented as relative % to the total activity) (upper graph) and to SDS-PAGE
and immunodetection with the anti-OmpF (outer membrane marker), anti-TolA (inner membrane
marker) and anti-FLAG antibodies. The positions of outer and inner membrane fractions (based on
control markers) are indicated.
78
Figure 6. Tle1EAEC
periplasmic toxicity is counteracted by Tli1EAEC
. (A) Cytoplasmic
production of Tle1EAEC
is not toxic. Serial dilutions (from 0 to 10-7
) of normalized cultures of E.
coli K-12 W3110 cells producing the wild-type (Tle1WT
) or the S197A mutant (Tle1S197A
)
Tle1EAEC
proteins from the pBAD18 vector were spotted on LB agar plates supplemented (right
panel) – or not (left panel) – with 0.2% arabinose. (B) Tli1EAEC
protects the cell against the
toxicity of the periplasmic production of Tle1EAEC
. Serial dilutions (from 0 to 10-7
) of normalized
cultures of E. coli K-12 W3110 cells producing Tli1EAEC
from the pBAD33 vector and the
maltose-binding protein (MBP), the wild-type (Tle1WT
) or the S197A mutant (Tle1S197A
) proteins
fused to a signal peptide (sp) from the pASK-IBA4 vector (periplasmic targeting) were spotted on
LB agar plates supplemented (right panel) – or not (left panel) – with 0.2% arabinose.
79
Figure 7. Tle1EAEC
interacts with the VgrG1 C-terminal extension. (A) Tle1EAEC
interacts with
VgrG, not with Hcp or PAAR. BTH101 reporter cells producing Tle1EAEC
fused to the T25
domain, and Hcp1, VgrG1 or PAAR proteins fused to the T18 domain of the Bordetella adenylate
cyclase were spotted on X-Gal indicator plates. The blue color of the colony reflects the
interaction between the two proteins. Controls include T18 and T25 fusions to TolB and Pal, two
proteins that interact but unrelated to the T6SS. (B) The C-terminal domain of VgrG1 is required
and necessary for Tle1EAEC
co-immunoprecipitation. The soluble lysate from 1011
E. coli K-12
W3110 cells producing VSVG-tagged Tle1EAEC
(Tle1VS) alone (-, empty vector) or mixed with
soluble lysates of W3110 cells producing the FLAG-tagged full-length (VgrGF) or variants of
VgrG1 represented in panel (C) were immunoprecipitated on anti-FLAG-coupled beads. The
immunoprecipitated material was subjected to 12.5%-acrylamide SDS-PAGE and
immunodetected with anti-FLAG (upper panel) and anti-VSVG (lower panel) monoclonal
antibodies. Molecular weight markers (in kDa) are indicated. The asterisks indicate the position of
80
the antibody heavy chain. (C) Schematic representation of the EAEC VgrG1 protein and of the
truncated variants used in this study. The different domains (and their boundaries) are indicated
(gp27 and gp5 structural core; cc, coiled-coil; DUF, DUF2345; TTR, transthyretin-like region).
(D) The TTR C-terminal region of VgrG1 is necessary and sufficient for the interaction with
Tle1EAEC
. BTH101 reporter cells producing Tle1EAEC
fused to the T25 domain, and the indicated
VgrG1 fragments fused to the T18 domain of the Bordetella adenylate cyclase were spotted on X-
Gal indicator plates. The blue color of the colony reflects the interaction between the two proteins.
Controls include T18 and T25 fusions to TolB and Pal, two proteins that interact but unrelated to
the T6SS.
Figure 8. The VgrG1 CTD is required for antibacterial activity but not for T6SS assembly.
81
(A) The VgrG1 CTD is required for Tle1-dependent killing. The anti-bacterial activity was
assessed by mixing W3110 gfp+ prey cells with the indicated attacker cells: WT (EAEC 17-2),
∆T6SS-1 (sci-1 gene cluster deletion derivative) and ∆vgrG (17-2 deleted of vgrG1 gene) carrying
the pBAD18 and pMS600 empty vectors, or producing the indicated proteins (western-blot
analyses shown in the inset), for 4 hours at 37°C in sci-1-inducing medium containing 0.02 %
arabinose. The image of a representative bacterial colony is shown and the relative fluorescent
levels (in arbitrary units, AU) are indicated in the upper graph. The number of recovered E. coli
prey cells is indicated in the lower graph (in log of colony-forming units (cfu)). The circles
indicate values from three independent assays, and the average is indicated by the bar. (B) The
VgrG1 CTD is not required for proper assembly and function of the Sci-1 T6SS. Hcp release was
assessed by separating whole cells (C) and supernatant [140] fractions from WT (EAEC 17-2),
∆T6SS-1 (sci-1 gene cluster deletion derivative) and ∆vgrG1 (17-2 deleted of vgrG1 gene)
carrying the pMS600 empty vector or producing the indicated VgrG1 variant, and pBAD-HcpVSVG.
1×108 total cells and the TCA-precipitated material from the supernatant of 2×10
8 cells were
subjected to SDS-PAGE and immunodetection using anti-VSVG monoclonal antibody (lower
panel) and anti-TolB polyclonal antibodies as a lysis control (upper panel). The molecular weight
markers (in kDa) are indicated on the left.
82
Chapter 1b: supplementary experiments (not published)
In this section, some experiments are described that are still underway and that have not been
published because they are not finalized. These experiments are covering crystallization and data
collection of the Tle1EAEC
- Tli1EAEC
complex, as well as nanobodies generation against Tle1EAEC
and the Xray structure of a nanobody.
1b.1. Crystallization and data collection of the Tle1EAEC
- Tli1EAEC
complex.
The Tle1EAEC
- Tli1EAEC
complex has been subjected to crystallization assays after gel filtration.
The complex was concentrated at 15 mg/mL in Tris-HCl 20 mM pH 8.0, NaCl 150 mM. 100 L
of complex solution was mixed in Greiner plates with 100 or 200 L of a well precipitant solution
containing 0.2M NaCl, 0.1M Na Cacodylate pH 6, and 8% w/v PEG8000. Crystals of ~80 x 80 x
50 appeared after a few days (Figure 1b1).
A crystal of the Tle1EAEC
- Tli1EAEC
complex was mounted in a cryoloop and frozen at 100 K.
It was exposed to X-rays at ESRF (Grenoble, France) beamline ID29. The crystals, however,
did not diffract beyond 6.5 Å resolution. A complete data set was collected allowing to
determine the space-group as cubic, P 213 (Table 1b.1). This low resolution can be explained
by the large crystal cell, with a=b=c=364.1 Å. Such an asymmetric may contain between 8
complexes (Vm=5.82 Å3/Da, Vs=76%) and 20 complexes (Vm=2.33 Å
3/Da, Vs=47%), with
the highest probability for this latter value as indicated by the Matthews Probability
Calculator (C.X. Weichenberger and B. Rupp,
http://www.ruppweb.org/mattprob/default.html) (Figure 1b.2). Despite tremendous efforts,
we could not obtain another crystal form, and diffraction could not be improved.
Figure 1b1 : Mono-crystals of the
Tle1EAEC
- Tli1EAEC
complex. The
size of this crystal is approximatively
~80 x 80 x 50 m
83
Considering the crystallization successes obtained in the group with nanobodies, it was
decided to generate anti Tle1EAEC
llama nanobodies.
Table 1b1. Data collection and refinement
statistics the Tle1EAEC
- Tli1EAEC
complex.
(numbers in brackets refer to the highest
resolution bin)
DATA COLLECTION nbK18
Beam line ESRF ID29
Space group P213 (cubic)
Cell (Å) a=b=c=364.1 Å
Resolution limits (Å) 50-6.52 (6.69-6.52)
Rmerge 0.13 (0.93)
CC1/2 0.99 (0.59)
Unique reflections 31158 (2130)
Mean((I)/sd(I)) 8.3 (1.9)
Completeness (%) 98.6 (92.0)
Multiplicity 4.1 (4.3)
1b.2 Nanobodies generation against Tle1EAEC
.
In order to obtain better crystals of Tle1EAEC
and/or of the Tle1EAEC
- Tli1EAEC
complex, we
raised llama nanobodies against Tle1EAEC
to be co-crystallized with Tle1EAEC
or with Tle1EAEC
- Tli1EAEC
complex. A llama was therefore imunized with Tle1EAEC
.
Five injections of 1 mg purified Tle1EAEC
(in 20 mM Hepes pH 6.8, 500 mM NaCl) were
performed subcutaneously at one-week intervals in one llama (Llama glama from Ardèche
lamas France). Lymphocytes were isolated from blood sample obtained five days after the last
immunization The cDNA was synthesized from purified total RNA by reverse transcription
and was used as a template for PCR amplification to amplify the sequences corresponding to
the variable domains of the heavy-chain antibodies. PCR fragments were then cloned into the
phagemid vector pHEN4 [141] to create a nanobody phage display library. Selection and
screening of nanobodies were performed as described previously [142]. A clear enrichment of
antigen-specific clones was already observed after two consecutive rounds of selection on
solid-phase coated antigen. Twenty-four randomly chosen colonies from respectively panning
Figure 1b2. Probability of the number of
complexes in the asymmetric unit. The
most probable number is 20.
84
2 and panning 3 were grown for expression of their specific nanobody as soluble protein. Of
these crude periplasmic extracts tested in an ELISA, 5/24 from pan2 and 13/24 from pan3
were shown to be specific towards the Tle1EAEC
protein. After sequence analysis, two
different positive clones, nb14 and nb15, were chosen to be sub-cloned into pHEN6
expression vector downstream the pelB signal peptide and fused to a C-terminal 6×His tag
(Figure 1b.3). RF sub-cloning, expression and purification of those 2 clones were performed.
Nanobodies expression and purification were performed as described [143] .
Figure 1b.3. Sequences of the 2 anti-Tle1EAEC
llama nanobodies. CDR1 to 3 are red,
green and blue, respectively.
.......10!.......20!.......30!.......40!.......50!..abcdef.....60!..
nb14 QVQLVESGGGLVQAGGSLRLSCAASGRTFSSYAMGWFRQAPGKEREFVAAISW-----SGGSTYYADS
nb15 QVQLVESGGGLVRAGGSLTLSCTVSGSTVSDYAMGWFRQAPEKERVFVAAVSL-----TGRSTRYADS
.....70!.......80!..abc.....90!......100!abcdefghijklmnop......110!...
nb14 VKGRFTISRDNAKNTVYLQMNSLKPEDTAVYYCAAASASSWYPAFIPGEYDY-------WGQGTQVTVSS
nb15 VKGRFTISRDNAKNTVYLQMNSLKPEDTALYYCAAGPAYGTAYADQMAYDY--------WGQGTQVTVSS
1b.3 Structure of the nanobody nbTle1-15 against Tle1EAEC
.
We decided to crystallize and solve the structure of nb15 against Tle1EAEC
, as this structure would
be of utmost importance for molecular replacement if we could obtain nb15/ Tle1EAEC
crystals.
The nb15 nanobody has been subjected to crystallization assays. The nanobody was concentrated
at 15 mg/mL in Tris-HCl 20 mM pH 8.0, NaCl 150 mM. 100 L of nanobody solution was mixed
in Greiner plates with 100 or 200 L of a well precipitant solution containing 0.2M NaCl, 0.1M
Na Cacodylate pH 6, and 8% w/v PEG8000. Crystals of ~150 x 100 x 80 appeared after a few
days.
A crystal of the nb15 nanobody was mounted in a cryoloop and frozen at 100 K. It was exposed to
X-rays at ESRF (Grenoble, France) beamline ID29. The crystals diffracted up to 1.30 Å resolution
and a complete data set was collected at this resolution. The crystals belong to the monoclinic
space group I 1 2 1, with cell dimensions a=38.9 Å, b=30.4 Å, c = 50.0 Å and β=104.2° (Table
1b.2). One molecule per asymmetric unit yields a Vm of 1.86 Å3/Da and Vs=33.7%.
85
Table 1b.2. Data collection and refinementof nb15. (numbers in brackets refer to the highest
resolution bin)
DATA COLLECTION REFINEMENT
PDB Resolution (Å) 38.8-1.30 (1.35-1.30)
Source Number of reflections 27475 (2710)
Space group I 1 2 1 protein / water atoms 1062 / 172
Cell (Å) a=38.9, b=30.4, c = 50.0
Å Test set reflections 1405
Angles (°) α= γ=90; β=104.2 Rwork/Rfree 22.1/24.4 (21.7/23.0)
Resolution limits (Å) 40-1.30 (1.35-1.30) r.m.s.d.bonds (Å)/angles
(°) 0.008 / 1.06
Rmerge 0.063 (0.67) B-Wilson / B-mean Å 10.4 / 16.1
CC1/2 0.995 (0.80) Ramachandran: preferred /
allowed / outliers (%) 99.3 / 0.7 / 0
Unique reflections 27490 (2725)
Mean((I)/sd(I)) 8.5 (1.7)
Completeness (%) 93.4 (75.0)
Multiplicity 2.8 (2.3)
Figure 1b.4. X-ray structure of the nanobody anti Tle1EAEC
, nb15. A/ Ribbon view of the
nanobody in which the three CDRs, the disulfide bridge and the His tag have been labelled. B/ A
region of the 2Fo-Fc map around Trp36.
The structure was solved by molecular replacement using Molrep [144] with a nanobody structure
as search model, in which the CDRs were removed. Refinement was performed with AutoBuster
[145] alternated with manual building using Coot [146]. The chain could be completely traced
between all the nanobody residues (1 to 125) and also 5 His residues from the His6 tag.
86
Conclusion and perspectives of Chapter 1b.
Our crystallization experiments have shown that the complex Tle1EAEC
- Tli1EAEC
is stable
enough to produce diffracting crystals. However, the large unit cell of the crystals was
probably responsible of their weak diffraction. To turn around this problem, we thought to use
nanobodies as they have been shown to favour well packed crystals. However, due to a lack of
time, it was not possible to test extensively the effect of the two nanobodies (nb14 and nb15)
on crystallization. Nevertheless, all the elements are documented for further crystallization
experiments of Tle1EAEC
/nanobodies or Tle1EAEC
- Tli1EAEC
/nanobodies complexes.
87
CHAPTER 2 : Purification, characterization, crystallization and preliminary X-ray
studies of the type VI secretion effector/immunity protein Tle3AIEC
/ Tli3AIEC
from
Adherent-Invasive Escherichia coli (Manuscript in preparation)
Foreword
In another effort to structurally investigate T6SS phospholipase effectors, I focused on
Tle3-Tli3 from AIEC (Adherent-invasive E. coli) LF82 T6SS1 for the second half of my PhD.
Adherent invasive E. coli (AIEC) is associated with the ileal-Crohn’s disease (CD). This
bacterium shares many genetic and phenotypic features with group of extra-intestinal
pathogenic E. coli (ExPEC) which is associated with urinary tract infections and neonatal
meningitis [147, 148]. AIEC bacteria have been isolated from inflammatory bowel disease
patients and has been increasingly implicated in the ileal-Crohn’s disease patients [149].
Studies indicated that AIEC is more prevalent in CD patents than in controls in many
countries, France, United Kingdom, Spain and USA [150].
AIEC is characterized by the ability to adhere and invade intestinal epithelial cells, the
ability to survive and replicate expansively in large vacuoles within macrophages without
triggering host cell death, and the ability to induce the release of tumor necrosis factor alpha
by infected macrophages [148, 151]. AIEC strain LF82 was shown to disrupt the integrity of
epithelial cells in in vivo or in vitro cell models. Other studies demonstrated that AIEC LF82
induces aggregation of infected macrophages, forming multinucleated giant cells and
subsequent recruitment of lymphocytes [152]. CD patients need to be treated with specific
personal strategies because of different etiologies. CD patients colonized by AIEC could be
effectively treated by eradicating these bacteria. More generally, biomarkers that can reliably
predict the responsiveness and efficacy of treatments is urgent. Such biomarkers that
recognize the microbial presence is useful for diagnosis, monitoring disease activity, and
therapeutic orientation [153].
Genomic analysis of AIEC revealed several pathogenicity islands encoding a variety of
virulence factors (4 putative pathogenic islands carrying virulence-related genes), but it lacks
most of the virulence genes found in ExPEC strains [92]. In 2010, whole genomes of AIEC
strains were sequenced and genes encoded for complete T6SSs were identified [92, 147]. The
AIEC LF82 chromosome carries two putative T6SSs: T6SS1 and T6SS2 located on PAI I and
PAI III, respectively [92].
88
These T6SS clusters have been indicated to play a role in the virulence of the AIEC
strains, that increase the adhesion and invasion capacity and limit survival in macrophages
[154]. Deletion of the T6SS pathogenicity island (null mutants of T6SS-1, T6SS-2 or both)
did not alter the morphology or growth of AIEC LF82 bacteria. However, these mutants
showed a dramatic motility defect, in particular LF82- deltaT6SS-1 and double LF82-
deltaT6SS1+2. The deletion of these clusters results in reduction of AIEC LF82 ability to
adhere to and to invade intestinal epithelial T84 cells. Besides, these systems have been
proved to be essential in the process and the maintenance of infection by the AIEC strains.
Determination of the secreted proteins by these T6SSs is important and this may provide new
bacterial targets for controlling the virulence of the AIEC strains [154].
The AIEC LF82 T6SS1 gene cluster encodes the 14 T6SS core-components and a
number of genes of unknown function. Tle3 was predicted to be a phospholipase. It was
expressed and purified with high yield while Tli3, bearing 6 cysteines, had to be expressed in
periplasm with very low yield. Co-crystallization of Tle3-Tli3 resulted in nice crystals that
diffracted up to 3.8 A (Seleno-Met anomalous dataset). The crystals belonged to space group
P 21, with unit-cell parameters a = 66.985, b = 445.238, c = 116.001 Å. The structural
determination of the complex is still in progress.
In another direction, nanobodies against Tle3 have been generated and selected in
order to obtain better crystals with higher diffraction of the proteins. Seven Tle3-specifically-
bound nanobodies were selected. We are now performing the characterization and co-
crystallization of Tle3 with various nanobodies.
Preliminary results on Tle3 and Tli3 are now gathered for publication: Purification,
characterization, crystallization and preliminary X-ray studies of the putative type VI
secretion effector/immunity protein Tle3AIEC/ Tli3AIEC from Adherent-invasive Escherichia
coli. The manuscript is in preparation.
From the intensive characterization of Tle3, Tli3 and the Tle3-Tli3 complex, I
managed to prepare and characterize the proteins and complexes and I performed the protein
crystallization experiments. Silvia Spinelli helped me in protein crystallography: data
collection and data reduction. Aline Desmyter generated and selected the positive nanobodies
and Christine Kellenberger determined the affinities and the epitopes competition between
proteins and nanobodies. Christian Cambillau and Alain Roussel guided me in the
experiments. The phospholipase activity was performed by our co-workers from IMM,
Stephane Canaan.
89
90
Publication (in preparation): Purification, characterization, crystallization and preliminary
X-ray studies of the type VI secretion effector/immunity protein Tle3AIEC
/ Tli3AIEC
from
Adherent-Invasive Escherichia coli
Abstract
Two proteins from Adherent-Invasive Escherichia coli (AIEC) have recently been identified
as a putative Tle3AIEC
phospholipase effector and its cognate immunity protein - an outer
membrane lipoprotein Tli3AIEC
from a type VI secretion system (T6SS). We further showed
that Tle3AIEC
possesses phospholipase A1 activity required for the interbacterial competition.
And, the toxicity of Tle3AIEC
protein can be neutralized by the cognate immunity protein
Tli3AIEC
which binds Tle3 in a 1:1 stoichiometric ratio and inhibits its phospholipase activity.
Here, the expression, purification, characterization, crystallization of Tle3AIEC
and/or Tli3AIEC
and preliminary crystallographic analysis of the complex Tle3AIEC
- Tli3AIEC
are reported. X-
ray diffraction data were collected from selenomethionine-derivatize Tle3AIEC SeMet
- Tli3AIEC
crystals to a resolution of 3.8 Å. The crystals belonged to space group P 21, with unit-cell
parameters a = 66.985, b = 445.238, c = 116.001 Å
1. Introduction
Encoded by more than 25% of Gram-negative species [138], the type VI secretion system
(T6SS) is a nanomachine that delivers virulence effector proteins into both eukaryotic and
prokaryotic cells [155]. To protect themselves from the effectors, the cells produce cognate
immunity proteins to neutralize the effector toxicities [46]. Effectors can be classified based
on what component they target. Effectors targeting bacterial cells were sub-divided into three
groups: cell wall-degrading enzymes (murein hydrolases), membrane-targeting proteins
(phospholipases and pore-forming proteins), and nucleases [74]. Recently, a superfamily of
T6SS lipase effectors named Tle1–Tle5 (type VI lipase effectors 1–5) and belonging to the
membrane-targeting group has been investigated. In this superfamily, Tle proteins have been
characterized that belong to Tle1 group (the Burkholderia
thailandensis Tle1BT
, P.
aeruginosa Tle1PA
and Enteroaggregative Escherichia coli Tle1EAEC
effectors [47, 108, 109]),
Tle2 group (Vibrio cholerae Tle2VC
[47]), Tle4 group (Pseudomonas aeruginosa Tle4 [110] )
or Tle5 group (P. aeruginosa Tle5aPA
and Tle5bPA
[47, 111, 112], Klebsiella pneumoniae
Tle5bKP
[113]). However, no member of the Tle3 group has been investigated.
A pathogenic group of E. coli, called Adherent-Invasive E. coli (AIEC), has been implicated
91
as having a role in human Crohn’s disease (CD) which is currently one of the most important
players in the pathogen story [156]. There are at least two gene clusters in the AIEC LF82
genome that encode T6SS clusters, named AIEC LF82 T6SS1 and AIEC LF82 T6SS2 [157,
158]. The protein 435 from the pathogen AIEC LF82 T6SS1 gene cluster has been predicted
to be a putative phospholipase Tle3 effector with PLA1 activity [158]. Its toxicity can be
neutralized by the cognate immunity protein 434, a putative Tli3, as they can bind together
and form a Tle3 - Tli3 protein complex. This pair of effector/immunity protein, belonging to
the Tle3 group has not been characterized to date. The two separated proteins and their
complex have been called Tle3AIEC
, Tli3AIEC
and Tle3AIEC
- Tli3AIEC
complex, respectively. In
order to further investigate Tle3AIEC
and Tli3AIEC
, we report their expression, purification
complexation and enzymatic characterization. Crystallization of the two proteins and
preliminary X-ray crystallographic studies of the Tle3AIEC
- Tli3AIEC
complex are also
reported.
2. Results
To identify potential effector toxins, we screened the AIEC LF82 T6SS1 gene cluster. The
AIEC LF82 T6SS1 gene cluster encodes the 14 T6SS core-components and a number of
genes of unknown function. Among those, a group of two genes, LF82_p434 and LF82_p435
is found between the putative vgrG (LF82_p433) and putative PAAR (LF82_p436) genes
[158] Ma 2014 (Figure S1). LF82_p435 (NCBI Gene Identifier (GI): 222034514) is encoded
downstream putative vgrG. Computer analysis of the LF82_p435 gene product using HHpred
indicates that it has the signature of an uncharacterized / hydrolase catalytic domain.
HHpred analysis suggests that this LF82_p435 gene might be a lipase/phospholipases.
Figure S1. The AIEC LF82 T6SS1 gene cluster. Encodes the 14 T6SS core-components and a number of
genes of unknown function.
A phylogenetic reconstruction of LF82_p435 with members of the Type VI lipase effector
(Tle) families 1 to 5 (as defined by Russell et al., 2013) showed that AIEC LF82_p435
protein is 100% identical to NP_755270 corresponding to protein c3395, a protein from
UPEC strain CFT073 (that was identified by Russel et al) that segregates with Tle3 members
92
(Figure S2). AIEC LF82_p435 protein is thus a putative member of the Tle3 family of T6SS
effectors and was therefore named hereafter Tle3AIEC
.
Figure S2. Phylogenetic reconstruction of
LF82_p435. LF82_p435 was reconstructed with
members of Type VI lipase effector (Tle) families
1 to 5 (as defined by Russell et al., 2013) showed
that AIEC LF82_p435 protein was 100%
identical to NP_755270
In T6SS, protection against kin cells is secured by the production of immunity proteins that
specifically bind and inhibit their cognate toxins. Usually, effector/immunity genes are found
as tandem in genomes. In the AIEC LF82 T6SS1 gene cluster, LF82_p434 (NCBI Gene
Identifier (GI): 222034513) is encoded directly downstream a putative vgrG gene. Computer
analysis of the LF82_p435 gene product using the lipo1.0 predict that LF82_p435 is a
lipoprotein. To confirm biochemical result , the LF82_p434 protein was purified to
homogeneity and tested for its ability to interfere with the Tle3AIEC
activity.
Figure 3C show that the Tle3AIEC
phospholipase activity was inhibited by LF82_p434 in a
dose-dependent manner. Interestingly, the Tle3AIEC
activity was completely abolished with a
Tle3AIEC
: LF82_p434 molecular ratio of 1:1 (Fig 3C), which is the highest ratio that can be
expected between an enzyme and a specific inhibitor. Taken together, these results confirm
93
that LF82_p434 is an immunity protein that protects against the phospholipase activity of
Tle3AIEC
, and therefore LF82_p434 was named Tli3AIEC
.
2.1 Cloning and protein expression
The gene encoding Tle3AIEC
and Tli3AIEC
(without the N-terminal 25-residue signal peptide)
were amplified from AIEC LF82 genomic DNA. Tle3AIEC
was cloned into the pETG20A
vector that incorporates a TRX fusion protein followed by N-terminal His tag tand a TEV [84]
cleavage site. The expresstion plasmid was transformed into Escherichia coli T7 Iq pLys
strain for expression. Cells were grown in a 2 l fermenter at 37°C in Terrific Broth medium
supplemented with 100 g/ml ampicillin. When the OD600 reached 0.8, the temperature was
reduced to 17°C and expression was induced overnight by the addition of 0.5 mM IPTG.
Expression of selenomethionine (SeMet)-labelled Tle3AIEC
(Tle3AIEC SeMet
) was performed in
Escherichia coli T7 Iq pLys cells using the method of methionine biosynthesis pathway
inhibition [159]. In both cases, cells were harvested by centrifugation and stored at - 80°C.
Tli3AIEC
was inserted into a pET22b vector (Novagen) with a C-terminal His6 tag in order to
be addressed to the periplasm. E. coli C41 (DE3) strain (Sigma-aldrich) was transformed with
pET22b-Tli3 for expression. E. coli C41 (DE3) cells carrying pET22b- Tli3AIEC
were grown
at 37oC in Terrific Broth medium containing 0.1% glucose and 100 g/ml ampicillin to an
optical density (OD600nm) ~ 0.8. Induction of the culture was then carried out with the
addition of 1 mM IPTG and incubation for 16 hours at 28oC. The periplasmic fraction
containing the Tli3 was prepared by osmotic shock according to Skerra & Pluckthun [160].
The cells were pelleted by centrifugation at 4000 rpm for 15 min at 4°C. The pellet was
resuspended in 9ml cold TES buffer (0.2M Tris-HCl pH 8.0, 0.5mM EDTA, 0.5M sucrose)
and kept on ice for 1 hour. The periplasmic proteins were removed by osmotic shock by
addition of 13.5ml cold TES diluted 1/4 into H2O. After 1-2 hours on ice, the suspension was
centrifugated 2 times at 13000 rpm for 30 min at 4°C. The supernatant containing the
expressed Tli3AIEC
was stored at - 80°C.
2.2. Protein purification
2.2.1. Purification of Tle3AIEC
, Tle3AIEC SeMet
The Tle3AIEC
or Tle3AIEC SeMet
cell pellet was resuspended in Tris-HCl 20 mM pH 8.0, NaCl
300 mM, glycerol 5% (v/v), lysozyme (0.25 mg/ml), DNase (2 g/ml), MgSO4 20 mM, and
phenylmethylsulfonyl fluoride (PMSF) 1 mM and cells were lysed by ultrasonication on ice.
94
The insoluble material was discarded by centrifugation at 20,000 × g for 60 min at 4C. The
soluble thioredoxin 6×His-tagged Tle3AIEC
or Tle3AIEC SeMet
fusion proteins were purified by
affinity chromatography on a nickel–nitrilotriacetic acid resin (Bio-Rad) and the tag was
removed after dialysis by overnight hydrolysis with the TEV protease and reloading in
presence of 10 mM imidazole. The proteins were further purified by gel filtration
chromatography (Superdex 200, 16/60 GE Healthcare) equilibrated in Hepes 10 mM pH 6.8,
NaCl 150 mM using an AKTA purifier System (Amersham). The recombinant Tle3
chromatogramme exhits two peaks : a small one, corresponding to an apparent molecular
mass of ~236 kDa (peak 1) and a large one, corresponding to an apparent molecular mass of
~73.8 kDa (peak 2) on a Superdex 200, 16/60 size exclusion chromatography (GE Healthcare)
(Figure 1a). The theoretical mass of Tle3 being 74 kDa (Figure 1b), these peaks likely
correspond to a trimer and a monomer in solution, respectively.
Figure 1. Solution characteristics of Tle3AIEC
. (a) Purified Tle3AIEC
eluted from gel-filtration
chromatography (Superdex 200, 16/60 GE Healthcare) at 66.92 ml and 79.52 ml corresponding to
molecular mass of ~140 kDa and 70 kDa respectively. (b) SDS–PAGE analysis of purified Tle3AIEC
visualized using Coomassie Blue. (c) Comparision : Purified Tle3AIEC
eluted from the frist gel-
filtration chromatography at 66.92 ml and 79.52 ml (diagram in pink). Putative dimer of Tle3AIEC
eluted from reinjected gel-filtration chromatography at 68 ml (diagram in blue). Putative monomer of
Tle3AIEC
eluted from reinjected gel-filtration chromatography at 77.5 ml (diagram in green).
The Tle3AIEC SeMet
protein exhibited the same characteristic as the native protein. The purified
protein fractions of the putative Tle3AIEC
trimer and monomer were pooled separately and
reinjected one by one to the same gel-filtration chromatography column. The reinjected
130
72
43
34
26
17
55
95
Peak1Tle3
Peak2Tle3
Peak1Tle3
Peak2Tle3
(a)(b)(c)
Peak1Tle3
Peak2Tle3
95
Tle3AIEC
dimer and monomer keep nearly the same range of elution at 68 ml and 77.5 ml,
respectively (Figure 1c), meaning that they do not reequilibrate (at least during the experiment
time) and keep their oligomerisation state. The purified protein fractions were pooled and
concentrated to ~20 mg/ml for crystallization by ultrafiltration using an Amicon membrane
(Millipore, California, USA). For phopholipase activity assays, the purified Tle3AIEC
protein
was concentrated to ~1 mg/mL. In all following experiments the monomeric Tle3AIEC
protein
was used.
2.2.2 Purification of Tli3AIEC
The soluble C-terminus fusion hexahistidine-tagged Tli3AIEC
was purified by immobilized
metal affinity chromatography on a 5-mL Ni-NTA column equilibrated in Tris-HCl 20 mM
pH 8.0, NaCl 300 mM, 10 mM imidazole. Tli3AIEC
were eluted in 250m Mimidazole and
concentrated (Amicon- Ultra 10-kDa cut-off) prior to be further purified by gel-filtration
chromatography (Superdex 75, 10/30 GE Healthcare) equilibrated in buffer consisting of 10
mM Hepes pH 6.8, 150 mM NaCl, using an AKTA purifier System (Amersham). The
recombinant Tli3AIEC
has an apparent molecular mass of ~32 kDa using a Superdex 75, 10/30
size exclusion chromatography column (GE Healthcare ) (Figure 2a). Considering its
molecular weight of 29 kDa, it is likely to be monomer in solution (Figure 2b). Highly
purified protein fractions were pooled and concentrated to 1.35 mg/ml. Notably, Tli3AIEC
alone is not stable, and therefore was used for crystallization only in complex with Tle3, as it
is stable associated with this enzyme. For phopholipase activity assays, the purified putative
monomer Tli3AIEC
proteins were concentrated to ~1 mg/mL, and used extemporaneously.
2.2.3 Purification of the Tle3AIEC
- Tli3AIEC
complex
The purified monomers of Tle3 and Tli3AIEC
were mixed together with a molar ratio of 1:1.3. The
complex was purified by gel filtration chromatography (Superdex 200, 16/60 GE Healthcare)
equilibrated in a Hepes 10 mM pH 6.8, NaCl 150 mM buffer using an AKTA purifier System
(Amersham). The Tle3AIEC
– Tli3AIEC
complex has an apparent molecular mass of 100 kDa on a
Superdex 200, 16/60 size exclusion chromatography column (GE Healthcare) (Figure 2c).
Considering its theoretical molecular weight of 109 kDa (Figure 2d), it is likely to be a 1:1 Tle3-
Tli3AIEC
heterodimer in solution. Highly purified protein fractions were pooled and concentrated
for crystallization by ultrafiltration with an Amicon cell (Millipore, California, USA) to 5.6
mg/ml, as quantified by the Bio-Rad protein assay kit.
96
Figure 2 Solution characteristics of Tli3AIEC
and Tle3AIEC
- Tli3AIEC
complex. (a) Purified Tli3AIEC
eluted from gel-filtration chromatography (Superdex 75, 10/30 GE Healthcare) at 11.3 ml
corresponding to molecular mass of ~40 kDa. (b) SDS–PAGE analysis of purified Tli3AIEC
visualized
using Coomassie Blue. (c) Purified Tle3AIEC
- Tli3AIEC
eluted from gel-filtration chromatography at
75.33 ml corresponding to molecular mass of ~109 kDa. (d) SDS–PAGE analysis of purified Tle3AIEC
- Tli3AIEC
and access of Tli3AIEC
visualized using Coomassie Blue.
2.3. Phospholipase A1 fluorescent assays and inhibition studies
A number of Tle proteins, delivered by the T6SS have been recently identified on the basis of
their vicinity to vgrG genes in the T6SS cluster. They are enzymes that were divided into five
divergent families [47] in which Tle families 1-4 contain a characteristic e fold
GXSXG motif found in lipases and some phospholipases. In vitro studies have shown that the
Burkholderia thailandensis Tle1
BT and P. aeruginosa Tle1
PA effectors have PLA2 activity [47,
108]. By contrast, the Vibrio cholerae Tle2VC
toxin exhibits a PLA1 activity[47]. Interestingly,
in our recent study (Chaper 1, above), Enteroaggregative Escherichia coli Tle1EAEC
has both
phospholipase A1 and phospholipase A2 (PLA2) [109]. Notably, no enzymatic activity has
been assigned so far for members of the Tle3 and Tle4 families. Therefore, the activity of the
purified Tle3AIEC
was tested on fluorogenic phospholipid substrates (Figure 3A).
Phospholipase A1 and A2 activities were monitored continuously using BODIPY® dye-
labeled phospholipids: PED-A1 (N-((6-(2,4-DNP)Amino)Hexanoyl)-1-(BODIPY® FL C5)-2-
Hexyl-sn-Glycero-3-Phosphoethanolamine) 3 and red/green BODIPY® PC-A2 (1-O-(6-
BODIPY-®558⁄568-Aminohexyl)-2-BODIPY®FLC5-sn-Glycero-3-Phosphocholine),
respectively [161, 162]. The sn-2 fatty acyl group in PED-A1 is a non-hydrolyzable alkyl
chain, and PED-A1 substrate was used to specifically measure the PLA1 activity. The
(a)(b)(c)(d)
Peak2Tli3
access
Peak1Tle3-Tli3
130
72
43
34
26
17
55
95
Peak1Tle3-Tli3
Peak2Tli3
access
97
red/green BODIPY®
PC-A2 has a sn-1 uncleavable alkyl chain. Substrate stock solutions (50
µM) were prepared in ethanol. All enzyme activities were assayed in 10 mM Tris-HCl pH 8.0,
150 mM NaCl, 1 mM CaCl2 and 0.1% Triton X-100. Enzymatic reactions were performed at
20°C for 25 min in a final volume of 200 µL containing 20 µg of Tle3 purified protein (from a
0.5 mg.mL-1
stock solution) and 5 µM of the substrate. The release of BODIPY® (BFCL5)
(Life Technologies) was recorded at λexc = 485 nm and λem = 538 nm using a 96-well plate
fluorometer (Fluoroskan ascent, Thermo Fissher Scientific). Enzymatic activities were
quantified using a BFCL5 calibration curve (0.08 - 200 pmoles in activity buffers) and
expressed in pmol of fatty acid (or BFLC5) released per minute per mg of protein (pmol min-1
mg-1
). PLA1 from Thermomyces lanuginosus and Bee venom PLA2 (Sigma-Aldrich, Saint-
Quentin Fallavier, France) were used as positive standards for PLA1 and PLA2 activities,
respectively. Results showed that Tle3 possesses phospholipase A1 (PLA1) activity (specific
activity (SA) = 7,65 nmole.min-1
.mg-1
). Beside, experiments have also been performed with
PLA2 substrate and in contrast to the case of Tle1EAEC
, no PLA2 activity was detected. This
result confirms our previous remark that the Tle classification was based on their protein
sequences and phylogenetic distribution, and not on their activity[109].
Purified Tli3AIEC
was tested for its ability to interfere with the Tle3AIEC
activity by incubating
30 μg of Tle3AIEC
with various molar ratio of Tli3AIEC
(xi=0, 0.25, 0.5, 1 and 2) (Figure 3B).
The activity was measured immediately or after 30 min incubation at room temperature.
Figure 3C show that the Tle3AIEC
phospholipase activity was inhibited by Tli3AIEC
in a dose-
dependent manner. Interestingly, the Tle3AIEC
activity was nearly abolished with a
Tle3AIEC
:Tli3AIEC
molecular ratio of 1:2 (Fig 3C). Whatever the incubation time, the xi=0.5
molar reaching 50% of activity, allows to demonstrate that Tle3AIEC
interacts with Tli3AIEC
with a 1:1 stoichiometry which is the highest ratio that can be expected between an enzyme
and a specific inhibitor. Tli3AIEC
is able to interact strongly to the Tle3AIEC
, probably inside
the active or in a close vicinity to the active, since no substrate can have access to the active
site to be hydrolysed. Taken together, these results suggest that Tli3AIEC
is an immunity
protein that protects against the phospholipase activity of Tle3AIEC
.
98
Figure 3. Tli3AIEC
inhibits Tle3AIEC
anti-bacterial phospholipase activity. (A) Specific
phospholipase A1 (PLA1) activity measurements of the Tle3AIEC
proteins using fluorescent
phospholipids. (B) Inhibition studies with Tli3AIEC
were performed by incubating 30 μg of Tle3AIEC
with various molar ratio of Tli3AIEC
(xi=0, 0.25, 0.5, 1 and 2). (C) Tli3AIEC
inhibition of Tle3AIEC
PLA1
activity. The rate of hydrolysis of PED-A1 by purified Tle3AIEC
at 20°C in presence of increasing
concentrations of Tli3AIEC
was plotted against the molar excess of Tli1EAEC
. (D) Cross-inhibition
studies with Tli1EAEC
were performed by incubating 20 μg of Tle3AIEC
with various molar ratio of
Tli1EAEC
(xi=0, 0.25, 0.5, 1, 2 and 10).
In order to study the specificity of Tli3 towards Tle3, cross-inhibition studies with Tli1EAEC
(From Entero-aggregative Escherichia coli Sci-1) were performed by incubating 20 μg of
Tle3AIEC
with various molar ratio of Tli1 (xI=0, 0.5, 1, 2 and 10) (Figure 3D). No inhibition by
Tli1EAEC
was detected even with the highest immunity/effector ratio. As expected this result
suggests a strong specificity between Tle and Tli from the same strain without any cross-
reaction.
2.4. Analytical Gel Filtration Analysis and MALS/QELS/UV/RI coupled size exclusion
chromatography
99
The results of gel filtration chromatography indicate that the Tle3AIEC
protein is likely to be
trimeric and monomeric (Figure 1a), while Tli3AIEC
is monomeric in solution (Figure 2b). Gel
filtration experiments have shown (see above) that Tli3AIEC
and Tle3AIEC
bind together in 1:1
molar ratio. To validate these results by a more precice approach, the purified Tle3AIEC
(before
the filtration chromatography step), the putative monomers of Tle3AIEC
and Tli3AIEC
, and the
mixture of them, were subjected to gel filtration with on-line multi-angle laser light
scattering/quasi-elastic light scattering/absorbance/refractive index (MALS/QELS/UV/RI).
The experiment was performed on an Alliance 2695 HPLC system (Waters) using a pre-
calibrated KW802.5 column (Shodex) run in Hepes 10 mM pH 7.4, NaCl 150 mM at 0.5
ml/min. MALS, UV spectrophotometry, QELS and RI were achieved with MiniDawn Treos
(Wyatt Technology), a Photo Diode Array 2996 (Waters), a DynaPro (Wyatt Technology) and
an Optilab rEX (Wyatt Technology), respectively, as already described by Sciara et al.[163].
Mass and hydrodynamic radius calculation was done with the ASTRA software (Wyatt
Technology) using a dn/dc value of 0.185 mL/g.
Concerning the the purified monomer Tle3AIEC
, a unique peak is observed with an apparent
molecular masse of 76 kDa (Figure 4), in agreement with its theoretical molecular weights of
~ 74 kDa. This confirms that this protein is monomeric in solution. Concerning the purified
mixture of Tle3AIEC
and Tli3AIEC
monomers, we observe in the chromatogramme a peak with
an apparent molecular mass of ~ 105 kDa, that likely corresponds to a Tle3AIEC
- Tli3AIEC
heterodimer (calculated molecular weight: 103 kDa). Finally and unfortunately, Tli3AIEC
could not be detected in this experiment, which could be explained by a fast degradating of
the protein on the column and in these conditions, an hypothesis confirmed by SDS PAGE.
In conclusion, the analyses of the purified monomer Tle3AIEC
and of heterodimer Tle3AIEC
-
Tli3AIEC
complex by MALS/QELS/UV/RI confirmed that Tle3AIEC
keeps its monomeric state
and that the Tle3AIEC
- Tli3AIEC
complex has a 1:1 stoichiometry (Figure 4), in agreement with
both their analysis by gel filtration chromatography and the activity inhibition.
100
Figure 4. MALS/SEC/UV/RI analysis. Tle3AIEC
(blue line) and to the Tle3AIEC
- Tli3AIEC
complex
(red line) was plotted against time (min. after sample injection in the High Performance Liquid
Chromatography system). The traces indicating the molar mass (indicated on the left, in Da) are shown
on each peak.
2.5. Crystallization of the Tle3AIEC
- Tli3AIEC
complex
Having well produced, stable and well behaved Tle3AIEC
- Tli3AIEC
complex encouraged us to
engage in a structural study of this complex by X-ray crystallography. We therefore
performed crystallization screening assays with Hampton Research and Qiagen kits using the
sitting-drop vapour-diffusion method at 20°C. The purified Tle3AIEC
- Tli3AIEC
protein was
crystallized under PEG-containing conditions. The Tle3AIEC
- Tli3AIEC
crystals were obtained
using a reservoir solution consisting of 20 % v/v PEG 3350; Bis Tris propane pH 6.5; 0.2 M
sodium acetate trihydrate after 44 days at 20°C. The Tle3AIEC
- Tli3AIEC
crystals quality
were optimized by varying the pH, the protein and precipitant concentrations and the protein /
precipitant ratio. The best condition was obtained by mixing 300 l of protein solution and
100 l of mother liquor (0.1 M Bis Tris propane pH 7.5, 25% PEG 3350, 0.2 M sodium
acetate trihydrate). Sharped crystals appeared after from 10 days to 3 weeks and reached
maximum dimensions of 100 x 40 x 40 m within about two months (Figure 5). The Tle3AIEC
- Tli3AIEC
crystals were socked with different heavy atoms: CsI (quick shock), TaBr (3 days-
shock), NaBr (quick shock) and cryocold in Ethylenglycol 12% to 14%. The best condition
for rescreen of SeMet Tle3SeMet _
Tli3 crystals were in 70 mM Bis-Tris Propane, 30mM Tris,
pH 7.6 17% PEG 3350, 130 mM NaAcetate, 9.4% PEG 6000.
2.6. Data collection and processing
Before data collection, the crystals were soaked with heavy atoms Cesium iodide (CsI) in a
101
reservoir solution supplemented with 12%(v/v) ethylenglycerol for a few seconds and then
flash-cooled in liquid nitrogen. The Tle3AIEC
- Tli3AIEC
data were collected at Soleil (Saint-
Aubin, France) on beamline Proxima 1 on a pilatus 6M detector and at a wavelength of
1.5498 Å in order to maximize the anomalous effect of the Cs+ and I
- ions. The final data-
collection and processing statistics are given in Table 1. The Tle3SeMet _
Tli3 data were also
collected at Proxima 2 using a pilatus 6M detector at a wavelength of 0.9791 Å. Before data
collection, the crystals were soaked in reservoir solution supplemented with 12%(v/v)
ethylenglycerol for a few seconds and then flash- cooled in liquid nitrogen. All data were
processed using XDS [164]. The final data-collection and processing statistics are given in
Table 1.
Figure 5 : Mono-crystals of the Tle3SeMet
_Tli3 complex. The size of this crystal is
approximatively ~100 x 40 x 40
2.7. Analysis of preliminary X-ray diffraction results
A total of 375390 measured reflections were merged into 67370 unique reflections. The
merged data set was 98.9% complete to 3.8 Å resolution [164]. The relevant data-collection
statistics are given in Table 1.
DATA COLLECTION Tle3AIEC
- Tli3AIEC
Space group P21
Cell (Å) a=67.9Å, b=449.1 Å,
c =
Resolution limits (Å) 50-3.8 (3.9-3.8)
Rmerge 0.31 (1.12)
CC1/2 0.98 (0.57)
Unique reflections 67370 (5521)
Mean((I)/sd(I)) 5 (1.2)
Completeness (%) 98.9 (92.2)
Multiplicity 3.7 (3.9)
The Tle3AIEC
- Tli3AIEC
crystals belonged to space group P 21 , with unit-cell parameters a =
Table 1. Data collection and refinement
statistics the Tle3AIEC
- Tli3AIEC
complex.
(numbers in brackets refer to the highest
resolution bin)
102
67.9, b = 449.1, c = 116.2 (Table 1). Based on the molecular weight of
101,990 Da and on the cell dimensions, the crystals exhibit a Vm of 17.4 Å3/Da for one
complex per asymmetric unit. A total of 4 to 8 complexes per asymmetric unit should
therefore be expected as this would lead to more reasonnable values of Vm in the range of
4.35-2.17 Å3/Da and solvent contents of 72-43%. However, the data resolution and quality
need to be improved before the crystal structure of Tle3SeMet _
Tli3 could be determined by the
single-wavelength anomalous dispersion (SAD) method.
2.8. Generation of llama nanobodies against Tle3AIEC
In order to obtain better crystals of Tle3AIEC
or Tle3AIEC
- Tli3AIEC
complex, and following the
same reasoning as for Tle1EAEC
– Tli1EAEC
, we decided to obtain llama nanobodies against
Tle3 that could be co-crystallized with Tle3 or with the Tle3AIEC
- Tli3AIEC
complex. To this
end, a llama was immunized with Tle3AIEC
.
Five injections of 1 mg purified Tle3AIEC
(in 20 mM Hepes pH 6.8, 500 mM NaCl) were
performed subcutaneously at one-week intervals in one llama (Llama glama from Ardèche
lamas France). Lymphocytes were isolated from blood sample obtained five days after the last
immunization The cDNA was synthesized from purified total RNA by reverse transcription
and was used as a template for PCR amplification to amplify the sequences corresponding to
the variable domains of the heavy-chain antibodies. PCR fragments were then cloned into the
phagemid vector pHEN4 [141] to create a nanobody phage display library. Selection and
screening of nanobodies were performed as described previously [142]. A clear enrichment of
antigen-specific clones was already observed after two consecutive rounds of selection on
solid-phase coated antigen. Twenty-four randomly chosen colonies were grown for expression
of their specific nanobody as soluble protein. Of these crude periplasmic extracts tested in an
ELISA, 17 were shown to be specific towards the Tle3 protein. After sequence analysis, seven
different positive clones were chosen to be sub-cloned into pHEN6 expression vector
downstream the pelB signal peptide and fused to a C-terminal 6×His tag. RF sub-cloning,
expression and purification of those 7 clones were performed. Nanobodies expression and
purification were performed as described [143].
103
Figure 6. anti-Tle3 llama nanobodies A) The seven nanobodies with non-redundent sequence.
B) Elisa assays (binding threshold 0.15)
.......10!.......20!.......30!.......40!.......50!..abcdef.....60!..
nb-Tle3 01 QVQLVESGGGSVQAGGSRRLSCVASGTIFSINYMAWFRQTPGKQRELVAGMSR------GGSTRYADS
nb-Tle3 02 QVQLVESGGALVQAGGSLRLSCAVSGVTFGSYVMGWFRQAPGKEREFVAAINR-----SGGSTNYANS
nb-Tle3 03 QVQLVESGGGLVQPGGSLRLSCAASGRTSSINVMGWYRQPPGKQRELVARMTT------GGTTNYADS
nb-Tle3 05 QVQLVESGGGLVQAGGSLTLSCTASGRAISDYAIAWFRQAPGKRREFVTKIGT-----KYYYTYYADS
nb-Tle3 08 QVQLVESGGGLVQPGGSLRLSCAASGFTFSGYAMSWVRQAPGKGPEWVSAITS-----GGYMTNYADS
nb-Tle3 17 DVQLVESGGGLVQPGGSLRLSCAASGFTFSNYGMNWVRQAPGKGPEWVSRITS-----GGRITSYSDS
nb-Tle3 20 QVQLVESGGGLVQPGGSLRLSCAASGNFFDIKDMGWYSQAPGKQREVVAAITR------GGSTHYADS
.....70!.......80!..abc.....90!......100!abcdefghijklmnop......110!...
nb-Tle3 01 VKGRFTVSRDNTKKSTYLQMNNLKPEDTAVYYCYAGRLGESNY----------------WGQGTQVTVSS
nb-Tle3 02 VKGRFTISRDYAKKTVYLQMNNLSPEDTAVYSCNAFRIVVGQPQAY-------------WGRGTQVTVSS
nb-Tle3 03 VKGRFTISRDNAKKTVYLQMNSLKSDDTAVYYCAAGGGWSFNSERQYDY----------WGQGTQVTVSS
nb-Tle3 05 VKGRFTISRDNTKNTVALQMNSLRPEDTAVYFCAAGDPNRVVGDRRSVSSEYDD-----WGQGTQVTVSS
nb-Tle3 08 VKGRFTTSRDNAKSTLYLQMNSLKPEDTAIYYCRVRISPNTY-----------------WGQGTQVTVSS
nb-Tle3 17 VKGRFTISRDDAKNAVYLQMSSLRPEDTAVYYCNTARY---------------------WGQGTQVTVSS
nb-Tle3 20 VKGRFTISKDGAKKTVYLQMNDLKPEDTAVYYCYARRSRIWGPEY--------------WGRGTQVTVSS
PBS coating
Block in PBS-skimmed milk 2%
Tle3
Pan2
Nb1
0.56
Nb20.
60
Nb30.
57
Nb40.
61
Nb5
0.34
Nb6
0.62
Nb70.
60
Nb80.
64
Nb90.
57
Nb10
0.58
Nb11
0.57
Nb12
0.25
Nb13
0.60
Nb14
0.22
Nb15
0.22
Nb16
0.61
Nb17
0.66
Nb18
0.10
Nb19
0.58
Nb20
0.65
Nb21
0.11
Nb22
0.08
Nb23
0.07
Nb24
0.56
3. Conclusion
Tle3AIEC
and Tli3AIEC
proteins were successfully expressed, purified and characterized.
Tle3AIEC
protein was obtained with high purity and yield while Tli3AIEC
had to be expressed in
periplasm with very low yield. Tle3AIEC
displays phospholipase activity, specificity PA1, and
belongs to family 3 of the T6SS lipase effectors. Gel filtration and Dynamic light scattering
results showed Tle3AIEC
exist as a monomer or dimer in the solution while Tli3AIEC
exist as a
monomer in the solution. The interaction between the two partners was observed by SDS
PAGE analysis after gelfitration and Dynamic light scattering results of the Tle3AIEC
- Tli3AIEC
complex. In agreement with that, inhibition studies showed that Tle3AIEC
interacts with
Tli3AIEC
with a 1:1 stoichiometry which is the highest ratio that can be expected between an
enzyme and a specific inhibitor. Tli3AIEC
is able to interact strongly to the Tle3AIEC
, probably
104
inside the active or in a close vicinity to the active since no substrate can have access to the
active site to be hydrolysed. Taken together, these results suggest that Tli3AIEC
is an immunity
protein that protects against the phospholipase activity of Tle3AIEC
. Co-crystallization of
Tle3AIEC
- Tli3AIEC
resulted in nice crystals that diffracted up to 3.8 Å (Seleno-Met anomalous
dataset). The structural determination of the complex is in progress.
105
GENERAL CONCLUSION
At the beginning of my PhD, I started to work on the two proteins Tle1EAEC
effector
and its immunity protein Tli1EAEC
encoded by genes in the EAEC Sci-1 T6SS cluster. These
two proteins were successfully expressed, purified and characterized with high purity and
yield. Our co-workers further demonstrate that Tle1EAEC
has phospholipase activity,
specificity PA1 and also PA2 activity, which contrasts with the previously characterized Tle1
members, Tle1BT
and Tle1PA
, which have PLA2 activity only. Gel filtration and Dynamic light
scattering results showed Tle1EAEC
and Tli1EAEC
both exist as a monomer in the solution.
Tli1EAEC
binds Tle1EAEC
in a 1:1 stoichiometric ratio with nanomolar affinity, and inhibits its
phospholipase activity.
Co-crystallization of Tle1EAEC
- Tli1EAEC
resulted in nice crystals. Unfortunately, these
crystals diffracted only to 6 Å (Seleno-Met anomalous dataset). The crystals belonged to
space group as cubic, P 213, with unit-cell parameters a=b=c=364.1 Å. Despite tremendous
efforts, we could not obtain another crystal form, and diffraction could not be improved. In
order to obtain better crystals of Tle1EAEC
and/or of the Tle1EAEC
- Tli1EAEC
complex, we
raised llama nanobodies against Tle1EAEC
to be co-crystallized with Tle1EAEC
or with Tle1EAEC
- Tli1EAEC
complex. Two nanobodies nbTle1-14 and nbTle1-15 has been subjected to
crystallization assays alone or in complex with Tle1EAEC
or with Tle1EAEC
- Tli1EAEC
complex.
Beside, we decided to crystallize and solve the structure of nbTle1-15 against Tle1, as this
structure would be of utmost importance for molecular replacement if we could obtain
nbTle1-15/Tle1 crystals. However, due to a lack of time, it was not possible to test extensively
the effect of the two nanobodies (nb14 and nb15) on crystallization.
At the same time, our co-workers at LISM showed that the EAEC Sci-1 T6SS is
required for inter-bacterial competition. Tle1EAEC
was the first T6SS phospholipase toxin to be
identified in EAEC, as well as Tli1EAEC
, its cognate immunity protein. Tle1EAEC
is required
for the anti-bacterial activity conferred by the Sci-1 T6SS. We also demonstrated that the
production of Tle1EAEC
in the cytoplasm of E. coli K-12 has no effect on its viability. By
contrast, cells do not survive when Tle1EAEC
is exported to the periplasm. Importantly, the
results support that the C-terminal extension of VgrG1 is necessary and sufficient to mediate
binding to and transport of Tle1EAEC
.
In the framework of my PhD, I also worked on Tle3/Tli3 from AIEC (Adherent-
invasive E. coli) LF82 T6SS1 – effector/immunity protein. Tle3AIEC
and Tli3 proteins were
successfully expressed, purified and characterized. Tle3AIEC
protein was obtained with high
106
purity and yield while Tli3AIEC
had to be expressed in periplasm with very low yield. Our co-
workers further demonstrate that Tle3AIEC
has phospholipase activity, specificity PA1, and
that it belongs to family 3 of the T6SS lipase effectors. Gel filtration and Dynamic light
scattering results showed Tle3AIEC
exist as a monomer or dimer in the solution while Tli3AIEC
exist as a monomer in the solution. The interaction between the two partners was observed by
SDS PAGE analysis after gelfitration and Dynamic light scattering results of the Tle3AIEC
-
Tli3AIEC
complex. In agreement with that, inhibition studies showed that Tle3AIEC
interacts
with Tli3AIEC
with a 1:1 stoichiometry which is the highest ratio that can be expected between
an enzyme and a specific inhibitor. Tli3AIEC
is able to interact strongly to the Tle3AIEC
,
probably inside the active or in a close vicinity to the active since no substrate can have access
to the active site to be hydrolysed. Taken together, these results suggest that Tli3AIEC
is an
immunity protein that protects against the phospholipase activity of Tle3AIEC
.
Co-crystallization of Tle3AIEC
- Tli3AIEC
resulted in nice crystals that diffracted up to
3.8 Å (Seleno-Met anomalous dataset). The structural determination of the complex is still in
progress.
In an effort to crystallize Tle3AIEC
/ Tli3AIEC
, nanobodies against Tle3AIEC
have been
generated and selected in order to obtain better crystals with higher diffraction of the proteins.
Seven Tle3AIEC
-specifically-bound nanobodies were selected. The characterization and co-
crystallization of Tle3AIEC
with various nanobodies will be performed in a near future.
107
PERSPECTIVES
While the characterizatio of Tle1, Tli1, Tle3 and Tli3 have been a success,
crystallization experiments have shown similar difficulties. Although nicely shaped crystals of
the complexes Tle1EAEC
/Tli1EAEC
and Tle3AIEC
/Tli3AIEC
have been obtained, they lacked
diffracting power. In both cases, the large unit cell of the crystals was probably responsible of
their weak diffraction and not the stability of the compounds. In an attempt to disrupt the
weak interactions between these complexes in the crystal, leading to large assemblies, we
thought to use llama nanobodies. We thought that these nanobodies would favour well
packed crystals. Only preliminary crystallization attempts were performed with
Tle1EAEC
/Tli1EAEC
and Tle3AIEC
/Tli3AIEC
. However, due to lack of time, it was not possible to
test extensively the effect of their nanobodies on crystallization. Nevertheless, all the elements
are documented and ready for another person to try this challenge.
Improving the Tle3AIEC
/Tli3AIEC
Se-Met crystals is a first possibility. However, considering
the large number of crystal improvement assays performed, I would propose two alternate
strategies:
- Firstly, I would suggest to select nanobodies that do not interfere with the inhibitors -
Tli1EAEC
or Tli3AIEC
- binding to their enzyme. Then to co-crystallize one or two nanobody to
the Tle1EAEC
/Tli1EAEC
and Tle3AIEC
/Tli3AIEC
complexes.
- Second, I would also suggest, in order to increase the chances of getting crystals, to attempt
co-crystallization of the phospholipases alone - Tle1EAEC
and Tle3AIEC
with, in this case,
neutralizing nanobodies. These nanobodies might be selected on the base of their inhibitory
effect on their cognate enzyme. A perfect situation would be to be able to compare a Tle/Tli
and a Tle/nanobody crystal structure!
108
REFERENCES
[1] Kudryashev M, Wang RY, Brackmann M, Scherer S, Maier T, Baker D, et al. Structure of
the type VI secretion system contractile sheath. Cell. 2015;160:952-62.
[2] Dewoody RS, Merritt PM, Marketon MM. Regulation of the Yersinia type III secretion
system: traffic control. Frontiers in cellular and infection microbiology. 2013;3:4.
[3] Alcoforado Diniz J, Coulthurst SJ. Intraspecies Competition in Serratia marcescens Is
Mediated by Type VI-Secreted Rhs Effectors and a Conserved Effector-Associated Accessory
Protein. Journal of bacteriology. 2015;197:2350-60.
[4] Russell AB, Peterson SB, Mougous JD. Type VI secretion system effectors: poisons with
a purpose. Nature reviews Microbiology. 2014;12:137-48.
[5] Durand E, Nguyen VS, Zoued A, Logger L, Pehau-Arnaudet G, Aschtgen MS, et al.
Biogenesis and structure of a type VI secretion membrane core complex. Nature.
2015;523:555-60.
[6] Rego AT, Chandran V, Waksman G. Two-step and one-step secretion mechanisms in
Gram-negative bacteria: contrasting the type IV secretion system and the chaperone-usher
pathway of pilus biogenesis. The Biochemical journal. 2010;425:475-88.
[7] Leo JC, Grin I, Linke D. Type V secretion: mechanism(s) of autotransport through the
bacterial outer membrane. Philosophical transactions of the Royal Society of London Series
B, Biological sciences. 2012;367:1088-101.
[8] Costa TR, Felisberto-Rodrigues C, Meir A, Prevost MS, Redzej A, Trokter M, et al.
Secretion systems in Gram-negative bacteria: structural and mechanistic insights. Nature
reviews Microbiology. 2015;13:343-59.
[9] Desvaux M, Hebraud M, Talon R, Henderson IR. Outer membrane translocation:
numerical protein secretion nomenclature in question in mycobacteria. Trends in
microbiology. 2009;17:338-40.
[10] Chagnot C, Zorgani MA, Astruc T, Desvaux M. Proteinaceous determinants of surface
colonization in bacteria: bacterial adhesion and biofilm formation from a protein secretion
perspective. Frontiers in microbiology. 2013;4:303.
[11] Schneewind O, Missiakas DM. Protein secretion and surface display in Gram-positive
bacteria. Philosophical transactions of the Royal Society of London Series B, Biological
sciences. 2012;367:1123-39.
[12] Chandran Darbari V, Waksman G. Structural Biology of Bacterial Type IV Secretion
Systems. Annual review of biochemistry. 2015;84:603-29.
[13] Abdallah AM, Gey van Pittius NC, Champion PA, Cox J, Luirink J, Vandenbroucke-
Grauls CM, et al. Type VII secretion--mycobacteria show the way. Nature reviews
Microbiology. 2007;5:883-91.
[14] Lillington J, Geibel S, Waksman G. Reprint of "Biogenesis and adhesion of type 1 and P
pili". Biochimica et biophysica acta. 2015;1850:554-64.
[15] Nivaskumar M, Francetic O. Type II secretion system: a magic beanstalk or a protein
escalator. Biochimica et biophysica acta. 2014;1843:1568-77.
[16] Galan JE, Lara-Tejero M, Marlovits TC, Wagner S. Bacterial type III secretion systems:
specialized nanomachines for protein delivery into target cells. Annual review of
microbiology. 2014;68:415-38.
109
[17] Cianfanelli FR, Monlezun L, Coulthurst SJ. Aim, Load, Fire: The Type VI Secretion
System, a Bacterial Nanoweapon. Trends in microbiology. 2016;24:51-62.
[18] Hobley L, Harkins C, MacPhee CE, Stanley-Wall NR. Giving structure to the biofilm
matrix: an overview of individual strategies and emerging common themes. FEMS
microbiology reviews. 2015;39:649-69.
[19] Nakayama K. Porphyromonas gingivalis and related bacteria: from colonial pigmentation
to the type IX secretion system and gliding motility. Journal of periodontal research.
2015;50:1-8.
[20] Kanonenberg K, Schwarz CK, Schmitt L. Type I secretion systems - a story of
appendices. Research in microbiology. 2013;164:596-604.
[21] Lycklama ANJA, Driessen AJ. The bacterial Sec-translocase: structure and mechanism.
Philosophical transactions of the Royal Society of London Series B, Biological sciences.
2012;367:1016-28.
[22] Palmer T, Berks BC. The twin-arginine translocation (Tat) protein export pathway.
Nature reviews Microbiology. 2012;10:483-96.
[23] Papanikou E, Karamanou S, Economou A. Bacterial protein secretion through the
translocase nanomachine. Nature reviews Microbiology. 2007;5:839-51.
[24] Smitha Rao CV, Anne J. Bacterial type I signal peptidases as antibiotic targets. Future
microbiology. 2011;6:1279-96.
[25] Orriss GL, Tarry MJ, Ize B, Sargent F, Lea SM, Palmer T, et al. TatBC, TatB, and TatC
form structurally autonomous units within the twin arginine protein transport system of
Escherichia coli. FEBS letters. 2007;581:4091-7.
[26] Frobel J, Rose P, Muller M. Twin-arginine-dependent translocation of folded proteins.
Philosophical transactions of the Royal Society of London Series B, Biological sciences.
2012;367:1029-46.
[27] Natale P, Bruser T, Driessen AJ. Sec- and Tat-mediated protein secretion across the
bacterial cytoplasmic membrane--distinct translocases and mechanisms. Biochimica et
biophysica acta. 2008;1778:1735-56.
[28] Sargent F. The twin-arginine transport system: moving folded proteins across
membranes. Biochemical Society transactions. 2007;35:835-47.
[29] Junker M, Besingi RN, Clark PL. Vectorial transport and folding of an autotransporter
virulence protein during outer membrane secretion. Molecular microbiology. 2009;71:1323-
32.
[30] Sauer FG, Remaut H, Hultgren SJ, Waksman G. Fiber assembly by the chaperone-usher
pathway. Biochimica et biophysica acta. 2004;1694:259-67.
[31] Busch A, Waksman G. Chaperone-usher pathways: diversity and pilus assembly
mechanism. Philosophical transactions of the Royal Society of London Series B, Biological
sciences. 2012;367:1112-22.
[32] Vetsch M, Puorger C, Spirig T, Grauschopf U, Weber-Ban EU, Glockshuber R. Pilus
chaperones represent a new type of protein-folding catalyst. Nature. 2004;431:329-33.
[33] Barnhart MM, Chapman MR. Curli biogenesis and function. Annual review of
microbiology. 2006;60:131-47.
[34] Evans ML, Chapman MR. Curli biogenesis: order out of disorder. Biochimica et
biophysica acta. 2014;1843:1551-8.
110
[35] von Tils D, Bladel I, Schmidt MA, Heusipp G. Type II secretion in Yersinia-a secretion
system for pathogenicity and environmental fitness. Frontiers in cellular and infection
microbiology. 2012;2:160.
[36] Douzi B, Filloux A, Voulhoux R. On the path to uncover the bacterial type II secretion
system. Philosophical transactions of the Royal Society of London Series B, Biological
sciences. 2012;367:1059-72.
[37] Sato K, Naito M, Yukitake H, Hirakawa H, Shoji M, McBride MJ, et al. A protein
secretion system linked to bacteroidete gliding motility and pathogenesis. Proceedings of the
National Academy of Sciences of the United States of America. 2010;107:276-81.
[38] Kharade SS, McBride MJ. Flavobacterium johnsoniae chitinase ChiA is required for
chitin utilization and is secreted by the type IX secretion system. Journal of bacteriology.
2014;196:961-70.
[39] Sato K, Yukitake H, Narita Y, Shoji M, Naito M, Nakayama K. Identification of
Porphyromonas gingivalis proteins secreted by the Por secretion system. FEMS microbiology
letters. 2013;338:68-76.
[40] Gerlach RG, Hensel M. Protein secretion systems and adhesins: the molecular armory of
Gram-negative pathogens. International journal of medical microbiology : IJMM.
2007;297:401-15.
[41] Thomas S, Holland IB, Schmitt L. The Type 1 secretion pathway - the hemolysin system
and beyond. Biochimica et biophysica acta. 2014;1843:1629-41.
[42] Diepold A, Wagner S. Assembly of the bacterial type III secretion machinery. FEMS
microbiology reviews. 2014;38:802-22.
[43] Cornelis GR. The type III secretion injectisome. Nature reviews Microbiology.
2006;4:811-25.
[44] Waksman G, Fronzes R. Molecular architecture of bacterial type IV secretion systems.
Trends in biochemical sciences. 2010;35:691-8.
[45] Ilangovan A, Connery S, Waksman G. Structural biology of the Gram-negative bacterial
conjugation systems. Trends in microbiology. 2015;23:301-10.
[46] Russell AB, Hood RD, Bui NK, LeRoux M, Vollmer W, Mougous JD. Type VI secretion
delivers bacteriolytic effectors to target cells. Nature. 2011;475:343-7.
[47] Russell AB, LeRoux M, Hathazi K, Agnello DM, Ishikawa T, Wiggins PA, et al. Diverse
type VI secretion phospholipases are functionally plastic antibacterial effectors. Nature.
2013;496:508-12.
[48] Ma LS, Hachani A, Lin JS, Filloux A, Lai EM. Agrobacterium tumefaciens deploys a
superfamily of type VI secretion DNase effectors as weapons for interbacterial competition in
planta. Cell host & microbe. 2014;16:94-104.
[49] Zoued A, Brunet YR, Durand E, Aschtgen MS, Logger L, Douzi B, et al. Architecture
and assembly of the Type VI secretion system. Bba-Mol Cell Res. 2014;1843:1664-73.
[50] Felisberto-Rodrigues C, Durand E, Aschtgen MS, Blangy S, Ortiz-Lombardia M, Douzi
B, et al. Towards a structural comprehension of bacterial type VI secretion systems:
characterization of the TssJ-TssM complex of an Escherichia coli pathovar. PLoS pathogens.
2011;7:e1002386.
[51] Ma LS, Narberhaus F, Lai EM. IcmF family protein TssM exhibits ATPase activity and
energizes type VI secretion. The Journal of biological chemistry. 2012;287:15610-21.
111
[52] Aschtgen MS, Bernard CS, De Bentzmann S, Lloubes R, Cascales E. SciN Is an Outer
Membrane Lipoprotein Required for Type VI Secretion in Enteroaggregative Escherichia coli.
Journal of bacteriology. 2008;190:7523-31.
[53] Shalom G, Shaw JG, Thomas MS. In vivo expression technology identifies a type VI
secretion system locus in Burkholderia pseudomallei that is induced upon invasion of
macrophages. Microbiol-Sgm. 2007;153:2689-99.
[54] Ma LS, Lin JS, Lai EM. An IcmF Family Protein, ImpL(M), Is an Integral Inner
Membrane Protein Interacting with ImpK(L), and Its Walker A Motif Is Required for Type VI
Secretion System-Mediated Hcp Secretion in Agrobacterium tumefaciens. Journal of
bacteriology. 2009;191:4316-29.
[55] Aschtgen MS, Thomas MS, Cascales E. Anchoring the type VI secretion system to the
peptidoglycan TssL, TagL, TagP, ... what else? Virulence. 2010;1:535-40.
[56] Aschtgen MS, Gavioli M, Dessen A, Lloubes R, Cascales E. The SciZ protein anchors
the enteroaggregative Escherichia coli Type VI secretion system to the cell wall. Molecular
microbiology. 2010;75:886-99.
[57] Leiman PG, Basler M, Ramagopal UA, Bonanno JB, Sauder JM, Pukatzki S, et al. Type
VI secretion apparatus and phage tail-associated protein complexes share a common
evolutionary origin. Proceedings of the National Academy of Sciences of the United States of
America. 2009;106:4154-9.
[58] Osipiuk J, Xu X, Cui H, Savchenko A, Edwards A, Joachimiak A. Crystal structure of
secretory protein Hcp3 from Pseudomonas aeruginosa. J Struct Funct Genomics. 2011;12:21-
6.
[59] Brunet YR, Henin J, Celia H, Cascales E. Type VI secretion and bacteriophage tail tubes
share a common assembly pathway. EMBO reports. 2014;15:315-21.
[60] Uchida K, Leiman PG, Arisaka F, Kanamaru S. Structure and properties of the C-
terminal beta-helical domain of VgrG protein from Escherichia coli O157. J Biochem.
2014;155:173-82.
[61] Lin JS, Ma LS, Lai EM. Systematic dissection of the agrobacterium type VI secretion
system reveals machinery and secreted components for subcomplex formation. PloS one.
2013;8:e67647.
[62] Basler M, Pilhofer M, Henderson GP, Jensen GJ, Mekalanos JJ. Type VI secretion
requires a dynamic contractile phage tail-like structure. Nature. 2012;483:182-6.
[63] Chang YW, Chen S, Tocheva EI, Treuner-Lange A, Lobach S, Sogaard-Andersen L, et
al. Correlated cryogenic photoactivated localization microscopy and cryo-electron
tomography. Nat Methods. 2014;11:737-9.
[64] Clemens DL, Ge P, Lee BY, Horwitz MA, Zhou ZH. Atomic structure of T6SS reveals
interlaced array essential to function. Cell. 2015;160:940-51.
[65] Shneider MM, Buth SA, Ho BT, Basler M, Mekalanos JJ, Leiman PG. PAAR-repeat
proteins sharpen and diversify the type VI secretion system spike. Nature. 2013;500:350-3.
[66] Bondage DD, Lin JS, Ma LS, Kuo CH, Lai EM. VgrG C terminus confers the type VI
effector transport specificity and is required for binding with PAAR and adaptor-effector
complex. Proceedings of the National Academy of Sciences of the United States of America.
2016;113:E3931-E40.
112
[67] Cascales E, Cambillau C. Structural biology of type VI secretion systems. Philosophical
transactions of the Royal Society of London Series B, Biological sciences. 2012;367:1102-11.
[68] Zoued A, Durand E, Bebeacua C, Brunet YR, Douzi B, Cambillau C, et al. TssK Is a
Trimeric Cytoplasmic Protein Interacting with Components of Both Phage-like and
Membrane Anchoring Complexes of the Type VI Secretion System. Journal of Biological
Chemistry. 2013;288:27031-41.
[69] English G, Byron O, Cianfanelli FR, Prescott AR, Coulthurst SJ. Biochemical analysis of
TssK, a core component of the bacterial Type VI secretion system, reveals distinct oligomeric
states of TssK and identifies a TssK-TssFG subcomplex. The Biochemical journal.
2014;461:291-304.
[70] Lossi NS, Dajani R, Freemont P, Filloux A. Structure-function analysis of HsiF, a gp25-
like component of the type VI secretion system, in Pseudomonas aeruginosa. Microbiology.
2011;157:3292-305.
[71] Kube S, Kapitein N, Zimniak T, Herzog F, Mogk A, Wendler P. Structure of the VipA/B
type VI secretion complex suggests a contraction-state-specific recycling mechanism. Cell
Rep. 2014;8:20-30.
[72] De Maayer P, Venter SN, Kamber T, Duffy B, Coutinho TA, Smits TH. Comparative
genomics of the Type VI secretion systems of Pantoea and Erwinia species reveals the
presence of putative effector islands that may be translocated by the VgrG and Hcp proteins.
BMC genomics. 2011;12:576.
[73] Whitney JC, Beck CM, Goo YA, Russell AB, Harding BN, De Leon JA, et al.
Genetically distinct pathways guide effector export through the type VI secretion system.
Molecular microbiology. 2014;92:529-42.
[74] Durand E, Cambillau C, Cascales E, Journet L. VgrG, Tae, Tle, and beyond: the versatile
arsenal of Type VI secretion effectors. Trends in microbiology. 2014;22:498-507.
[75] Zheng J, Leung KY. Dissection of a type VI secretion system in Edwardsiella tarda.
Molecular microbiology. 2007;66:1192-206.
[76] Blondel CJ, Jimenez JC, Contreras I, Santiviago CA. Comparative genomic analysis
uncovers 3 novel loci encoding type six secretion systems differentially distributed in
Salmonella serotypes. BMC genomics. 2009;10:354.
[77] Silverman JM, Agnello DM, Zheng H, Andrews BT, Li M, Catalano CE, et al.
Haemolysin coregulated protein is an exported receptor and chaperone of type VI secretion
substrates. Molecular cell. 2013;51:584-93.
[78] Hood RD, Singh P, Hsu F, Guvener T, Carl MA, Trinidad RR, et al. A type VI secretion
system of Pseudomonas aeruginosa targets a toxin to bacteria. Cell host & microbe.
2010;7:25-37.
[79] Li M, Le Trong I, Carl MA, Larson ET, Chou S, De Leon JA, et al. Structural basis for
type VI secretion effector recognition by a cognate immunity protein. PLoS pathogens.
2012;8:e1002613.
[80] Chou S, Bui NK, Russell AB, Lexa KW, Gardiner TE, LeRoux M, et al. Structure of a
peptidoglycan amidase effector targeted to Gram-negative bacteria by the type VI secretion
system. Cell Rep. 2012;1:656-64.
113
[81] Brooks TM, Unterweger D, Bachmann V, Kostiuk B, Pukatzki S. Lytic activity of the
Vibrio cholerae type VI secretion toxin VgrG-3 is inhibited by the antitoxin TsaB. The
Journal of biological chemistry. 2013;288:7618-25.
[82] Barret M, Egan F, Fargier E, Morrissey JP, O'Gara F. Genomic analysis of the type VI
secretion systems in Pseudomonas spp.: novel clusters and putative effectors uncovered.
Microbiology. 2011;157:1726-39.
[83] Koskiniemi S, Lamoureux JG, Nikolakakis KC, t'Kint de Roodenbeke C, Kaplan MD,
Low DA, et al. Rhs proteins from diverse bacteria mediate intercellular competition.
Proceedings of the National Academy of Sciences of the United States of America.
2013;110:7032-7.
[84] Pukatzki S, Ma AT, Revel AT, Sturtevant D, Mekalanos JJ. Type VI secretion system
translocates a phage tail spike-like protein into target cells where it cross-links actin.
Proceedings of the National Academy of Sciences of the United States of America.
2007;104:15508-13.
[85] Miyata ST, Kitaoka M, Brooks TM, McAuley SB, Pukatzki S. Vibrio cholerae requires
the type VI secretion system virulence factor VasX to kill Dictyostelium discoideum.
Infection and immunity. 2011;79:2941-9.
[86] Dong TG, Ho BT, Yoder-Himes DR, Mekalanos JJ. Identification of T6SS-dependent
effector and immunity proteins by Tn-seq in Vibrio cholerae. Proceedings of the National
Academy of Sciences of the United States of America. 2013;110:2623-8.
[87] Unterweger D, Kostiuk B, Otjengerdes R, Wilton A, Diaz-Satizabal L, Pukatzki S.
Chimeric adaptor proteins translocate diverse type VI secretion system effectors in Vibrio
cholerae. The EMBO journal. 2015;34:2198-210.
[88] Liang X, Moore R, Wilton M, Wong MJ, Lam L, Dong TG. Identification of divergent
type VI secretion effectors using a conserved chaperone domain. Proceedings of the National
Academy of Sciences of the United States of America. 2015;112:9106-11.
[89] Ma LS, Hachani A, Lin JS, Filloux A, Lai EM. Agrobacterium tumefaciens Deploys a
Superfamily of Type VI Secretion DNase Effectors as Weapons for Interbacterial
Competition In Planta. Cell host & microbe. 2014;16:94-104.
[90] Hachani A, Allsopp LP, Oduko Y, Filloux A. The VgrG proteins are "a la carte" delivery
systems for bacterial type VI effectors. The Journal of biological chemistry. 2014;289:17872-
84.
[91] Russell AB, Singh P, Brittnacher M, Bui NK, Hood RD, Carl MA, et al. A widespread
bacterial type VI secretion effector superfamily identified using a heuristic approach. Cell
host & microbe. 2012;11:538-49.
[92] Miquel S, Peyretaillade E, Claret L, de Vallee A, Dossat C, Vacherie B, et al. Complete
genome sequence of Crohn's disease-associated adherent-invasive E. coli strain LF82. PloS
one. 2010;5.
[93] Benz J, Sendlmeier C, Barends TR, Meinhart A. Structural insights into the effector-
immunity system Tse1/Tsi1 from Pseudomonas aeruginosa. PloS one. 2012;7:e40453.
[94] Busby JN, Panjikar S, Landsberg MJ, Hurst MR, Lott JS. The BC component of ABC
toxins is an RHS-repeat-containing protein encapsulation device. Nature. 2013;501:547-50.
114
[95] Ding J, Wang W, Feng H, Zhang Y, Wang DC. Structural insights into the Pseudomonas
aeruginosa type VI virulence effector Tse1 bacteriolysis and self-protection mechanisms. The
Journal of biological chemistry. 2012;287:26911-20.
[96] Fulop V, Jones DT. Beta propellers: structural rigidity and functional diversity. Current
opinion in structural biology. 1999;9:715-21.
[97] Zhang H, Zhang H, Gao ZQ, Wang WJ, Liu GF, Xu JH, et al. Structure of the type VI
effector-immunity complex (Tae4-Tai4) provides novel insights into the inhibition
mechanism of the effector by its immunity protein. The Journal of biological chemistry.
2013;288:5928-39.
[98] Benz J, Meinhart, A. . Antibacterial effector/immunity systems : it’s just the tip of the
iceberg. Current opinion in microbiology. 2014;17:1-10.
[99] Srikannathasan V, English G, Bui NK, Trunk K, O'Rourke PE, Rao VA, et al. Structural
basis for type VI secreted peptidoglycan DL-endopeptidase function, specificity and
neutralization in Serratia marcescens. Acta Crystallogr D Biol Crystallogr. 2013;69:2468-82.
[100] Zhang XY, Brunet YR, Logger L, Douzi B, Cambillau C, Journet L, et al. Dissection of
the TssB-TssC interface during type VI secretion sheath complex formation. PloS one.
2013;8:e81074.
[101] Anantharaman V, Aravind L. Evolutionary history, structural features and biochemical
diversity of the NlpC/P60 superfamily of enzymes. Genome Biol. 2003;4:R11.
[102] Whitney JC, Chou S, Russell AB, Biboy J, Gardiner TE, Ferrin MA, et al.
Identification, structure, and function of a novel type VI secretion peptidoglycan glycoside
hydrolase effector-immunity pair. The Journal of biological chemistry. 2013;288:26616-24.
[103] Wang T, Ding J, Zhang Y, Wang DC, Liu W. Complex structure of type VI
peptidoglycan muramidase effector and a cognate immunity protein. Acta Crystallogr D Biol
Crystallogr. 2013;69:1889-900.
[104] Li L, Zhang W, Liu Q, Gao Y, Gao Y, Wang Y, et al. Structural Insights on the
bacteriolytic and self-protection mechanism of muramidase effector Tse3 in Pseudomonas
aeruginosa. The Journal of biological chemistry. 2013;288:30607-13.
[105] Gerard F, Brooks MA, Barreteau H, Touze T, Graille M, Bouhss A, et al. X-ray
structure and site-directed mutagenesis analysis of the Escherichia coli colicin M immunity
protein. Journal of bacteriology. 2011;193:205-14.
[106] Suarez G, Sierra JC, Erova TE, Sha J, Horneman AJ, Chopra AK. A type VI secretion
system effector protein, VgrG1, from Aeromonas hydrophila that induces host cell toxicity by
ADP ribosylation of actin. Journal of bacteriology. 2010;192:155-68.
[107] Yang X, Xu M, Wang Y, Xia P, Wang S, Ye B, et al. Molecular mechanism for self-
protection against the type VI secretion system in Vibrio cholerae. Acta crystallographica
Section D, Biological crystallography. 2014;70:1094-103.
[108] Hu H, Zhang H, Gao Z, Wang D, Liu G, Xu J, et al. Structure of the type VI secretion
phospholipase effector Tle1 provides insight into its hydrolysis and membrane targeting. Acta
crystallographica Section D, Biological crystallography. 2014;70:2175-85.
[109] Flaugnatti N, Le TT, Canaan S, Aschtgen MS, Nguyen VS, Blangy S, et al. A
phospholipase A1 antibacterial Type VI secretion effector interacts directly with the C-
terminal domain of the VgrG spike protein for delivery. Molecular microbiology.
2016;99:1099-118.
115
[110] Lu D, Zheng Y, Liao N, Wei L, Xu B, Liu X, et al. The structural basis of the Tle4-Tli4
complex reveals the self-protection mechanism of H2-T6SS in Pseudomonas aeruginosa. Acta
crystallographica Section D, Biological crystallography. 2014;70:3233-43.
[111] Spencer C, Brown HA. Biochemical characterization of a Pseudomonas aeruginosa
phospholipase D. Biochemistry. 2015;54:1208-18.
[112] Jiang F, Waterfield NR, Yang J, Yang G, Jin Q. A Pseudomonas aeruginosa type VI
secretion phospholipase D effector targets both prokaryotic and eukaryotic cells. Cell host &
microbe. 2014;15:600-10.
[113] Lery LM, Frangeul L, Tomas A, Passet V, Almeida AS, Bialek-Davenet S, et al.
Comparative analysis of Klebsiella pneumoniae genomes identifies a phospholipase D family
protein as a novel virulence factor. BMC biology. 2014;12:41.
[114] Miyata ST, Unterweger D, Rudko SP, Pukatzki S. Dual expression profile of type VI
secretion system immunity genes protects pandemic Vibrio cholerae. PLoS pathogens.
2013;9:e1003752.
[115] Poole SJ, Diner EJ, Aoki SK, Braaten BA, t'Kint de Roodenbeke C, Low DA, et al.
Identification of functional toxin/immunity genes linked to contact-dependent growth
inhibition (CDI) and rearrangement hotspot (Rhs) systems. PLoS genetics. 2011;7:e1002217.
[116] Alteri CJ, Mobley HL. The Versatile Type VI Secretion System. Microbiology
spectrum. 2016;4.
[117] Fritsch MJ, Trunk K, Diniz JA, Guo M, Trost M, Coulthurst SJ. Proteomic
identification of novel secreted antibacterial toxins of the Serratia marcescens type VI
secretion system. Molecular & cellular proteomics : MCP. 2013;12:2735-49.
[118] Unterweger D, Miyata ST, Bachmann V, Brooks TM, Mullins T, Kostiuk B, et al. The
Vibrio cholerae type VI secretion system employs diverse effector modules for intraspecific
competition. Nature communications. 2014;5:3549.
[119] Salomon D, Kinch LN, Trudgian DC, Guo X, Klimko JA, Grishin NV, et al. Marker for
type VI secretion system effectors. Proceedings of the National Academy of Sciences of the
United States of America. 2014;111:9271-6.
[120] Nataro JP, Kaper JB, Robins-Browne R, Prado V, Vial P, Levine MM. Patterns of
adherence of diarrheagenic Escherichia coli to HEp-2 cells. Pediatr Infect Dis J. 1987;6:829-
31.
[121] Cohen MB, Nataro JP, Bernstein DI, Hawkins J, Roberts N, Staat MA. Prevalence of
diarrheagenic Escherichia coli in acute childhood enteritis: a prospective controlled study. J
Pediatr. 2005;146:54-61.
[122] Nataro JP, Mai V, Johnson J, Blackwelder WC, Heimer R, Tirrell S, et al.
Diarrheagenic Escherichia coli infection in Baltimore, Maryland, and New Haven,
Connecticut. Clin Infect Dis. 2006;43:402-7.
[123] Nataro JP. Outbreak of hemolytic-uremic syndrome linked to Shiga toxin-producing
enteroaggregative Escherichia coli O104:H4. Pediatr Res. 2011;70:221.
[124] Rasko DA, Webster DR, Sahl JW, Bashir A, Boisen N, Scheutz F, et al. Origins of the
E. coli strain causing an outbreak of hemolytic-uremic syndrome in Germany. N Engl J Med.
2011;365:709-17.
116
[125] Jiang ZD, Greenberg D, Nataro JP, Steffen R, DuPont HL. Rate of occurrence and
pathogenic effect of enteroaggregative Escherichia coli virulence factors in international
travelers. J Clin Microbiol. 2002;40:4185-90.
[126] Vila J, Vargas M, Henderson IR, Gascon J, Nataro JP. Enteroaggregative escherichia
coli virulence factors in traveler's diarrhea strains. The Journal of infectious diseases.
2000;182:1780-3.
[127] Nataro JP, Baldini MM, Kaper JB, Black RE, Bravo N, Levine MM. Detection of an
adherence factor of enteropathogenic Escherichia coli with a DNA probe. The Journal of
infectious diseases. 1985;152:560-5.
[128] Nataro JP, Deng Y, Cookson S, Cravioto A, Savarino SJ, Guers LD, et al.
Heterogeneity of enteroaggregative Escherichia coli virulence demonstrated in volunteers.
The Journal of infectious diseases. 1995;171:465-8.
[129] Chaudhuri RR, Sebaihia M, Hobman JL, Webber MA, Leyton DL, Goldberg MD, et al.
Complete genome sequence and comparative metabolic profiling of the prototypical
enteroaggregative Escherichia coli strain 042. PloS one. 2010;5:e8801.
[130] Henderson IR, Navarro-Garcia F, Desvaux M, Fernandez RC, Ala'Aldeen D. Type V
protein secretion pathway: the autotransporter story. Microbiol Mol Biol Rev. 2004;68:692-
744.
[131] Dudley EG, Thomson NR, Parkhill J, Morin NP, Nataro JP. Proteomic and microarray
characterization of the AggR regulon identifies a pheU pathogenicity island in
enteroaggregative Escherichia coli. Molecular microbiology. 2006;61:1267-82.
[132] Chen SL, Hung CS, Xu J, Reigstad CS, Magrini V, Sabo A, et al. Identification of
genes subject to positive selection in uropathogenic strains of Escherichia coli: a comparative
genomics approach. Proceedings of the National Academy of Sciences of the United States of
America. 2006;103:5977-82.
[133] Johnson TJ, Kariyawasam S, Wannemuehler Y, Mangiamele P, Johnson SJ, Doetkott C,
et al. The genome sequence of avian pathogenic Escherichia coli strain O1:K1:H7 shares
strong similarities with human extraintestinal pathogenic E. coli genomes. Journal of
bacteriology. 2007;189:3228-36.
[134] Touchon M, Hoede C, Tenaillon O, Barbe V, Baeriswyl S, Bidet P, et al. Organised
genome dynamics in the Escherichia coli species results in highly diverse adaptive paths.
PLoS genetics. 2009;5:e1000344.
[135] Morin N, Santiago AE, Ernst RK, Guillot SJ, Nataro JP. Characterization of the AggR
regulon in enteroaggregative Escherichia coli. Infection and immunity. 2013;81:122-32.
[136] Bernard CS, Brunet YR, Gavioli M, Lloubes R, Cascales E. Regulation of type VI
secretion gene clusters by sigma54 and cognate enhancer binding proteins. Journal of
bacteriology. 2011;193:2158-67.
[137] Morak M, Attarbaschi A, Fischer S, Nassimbeni C, Grausenburger R, Bastelberger S, et
al. Small sizes and indolent evolutionary dynamics challenge the potential role of P2RY8-
CRLF2-harboring clones as main relapse-driving force in childhood ALL. Blood.
2012;120:5134-42.
[138] Boyer F, Fichant G, Berthod J, Vandenbrouck Y, Attree I. Dissecting the bacterial type
VI secretion system by a genome wide in silico analysis: what can be learned from available
microbial genomic resources? BMC genomics. 2009;10:104.
117
[139] Minor ZOaW. Processing of X-ray Diffraction Data Collected in Oscillation Mode
Macromolecular Crystallography. 1997;276:307-26.
[140] van Bastelaar KM, Pouwer F, Cuijpers P, Riper H, Twisk JW, Snoek FJ. Is a severe
clinical profile an effect modifier in a Web-based depression treatment for adults with type 1
or type 2 diabetes? Secondary analyses from a randomized controlled trial. J Med Internet
Res. 2012;14:e2.
[141] Arbabi Ghahroudi M, Desmyter A, Wyns L, Hamers R, Muyldermans S. Selection and
identification of single domain antibody fragments from camel heavy-chain antibodies. FEBS
letters. 1997;414:521-6.
[142] Desmyter A, Farenc C, Mahony J, Spinelli S, Bebeacua C, Blangy S, et al. Viral
infection modulation and neutralization by camelid nanobodies. Proceedings of the National
Academy of Sciences of the United States of America. 2013;110:E1371-9.
[143] Conrath K, Pereira AS, Martins CE, Timoteo CG, Tavares P, Spinelli S, et al. Camelid
nanobodies raised against an integral membrane enzyme, nitric oxide reductase. Protein
Science. 2009;18:619-28.
[144] Vagin A, Teplyakov A. Molecular replacement with MOLREP. Acta Crystallogr D Biol
Crystallogr. 2010;66:22-5.
[145] Blanc E, Roversi P, Vonrhein C, Flensburg C, Lea SM, Bricogne G. Refinement of
severely incomplete structures with maximum likelihood in BUSTER-TNT. Acta Crystallogr
D Biol Crystallogr. 2004;60:2210-21.
[146] Emsley P, Lohkamp B, Scott WG, Cowtan K. Features and development of Coot. Acta
Crystallogr D Biol Crystallogr. 2010;66:486-501.
[147] Nash JH, Villegas A, Kropinski AM, Aguilar-Valenzuela R, Konczy P, Mascarenhas
M, et al. Genome sequence of adherent-invasive Escherichia coli and comparative genomic
analysis with other E. coli pathotypes. BMC genomics. 2010;11:667.
[148] Martinez-Medina M, Mora A, Blanco M, Lopez C, Alonso MP, Bonacorsi S, et al.
Similarity and divergence among adherent-invasive Escherichia coli and extraintestinal
pathogenic E. coli strains. J Clin Microbiol. 2009;47:3968-79.
[149] Rolhion N, Darfeuille-Michaud A. Adherent-invasive Escherichia coli in inflammatory
bowel disease. Inflamm Bowel Dis. 2007;13:1277-83.
[150] Denizot J, Sivignon A, Barreau F, Darcha C, Chan HF, Stanners CP, et al. Adherent-
invasive Escherichia coli induce claudin-2 expression and barrier defect in CEABAC10 mice
and Crohn's disease patients. Inflamm Bowel Dis. 2012;18:294-304.
[151] Glasser AL, Boudeau J, Barnich N, Perruchot MH, Colombel JF, Darfeuille-Michaud
A. Adherent invasive Escherichia coli strains from patients with Crohn's disease survive and
replicate within macrophages without inducing host cell death. Infection and immunity.
2001;69:5529-37.
[152] Meconi S, Vercellone A, Levillain F, Payre B, Al Saati T, Capilla F, et al. Adherent-
invasive Escherichia coli isolated from Crohn's disease patients induce granulomas in vitro.
Cell Microbiol. 2007;9:1252-61.
[153] Berry D, Reinisch W. Intestinal microbiota: a source of novel biomarkers in
inflammatory bowel diseases? Best Pract Res Clin Gastroenterol. 2013;27:47-58.
[154] S. Massier SM, N. Dreux, A. Agus, S. Bouhours, J. Denizot, A. Darfeuille-Michaud, N.
Barnich*. Involvement of type VI secretion systems in virulence of adherent-invasive
118
Escherichia coli isolated from patients with Crohn's disease. Congress of ECCO - European
Crohn’s and Colitis Organisation2015.
[155] Jani AJ, Cotter PA. Type VI secretion: not just for pathogenesis anymore. Cell host &
microbe. 2010;8:2-6.
[156] Agus A, Massier S, Darfeuille-Michaud A, Billard E, Barnich N. Understanding host-
adherent-invasive Escherichia coli interaction in Crohn's disease: opening up new therapeutic
strategies. BioMed research international. 2014;2014:567929.
[157] Massier S. Involvement of type VI secretion systems in virulence of adherent-invasive
Escherichia coli isolated from patients with Crohn's disease. 2014:
[158] Ma J, Bao Y, Sun M, Dong W, Pan Z, Zhang W, et al. Two functional type VI
secretion systems in avian pathogenic Escherichia coli are involved in different pathogenic
pathways. Infection and immunity. 2014;82:3867-79.
[159] Doublie S. Preparation of selenomethionyl proteins for phase determination. Methods in
enzymology. 1997;276:523-30.
[160] Skerra A, Pluckthun A. Assembly of a functional immunoglobulin Fv fragment in
Escherichia coli. Science. 1988;240:1038-41.
[161] Darrow AL, Olson MW, Xin H, Burke SL, Smith C, Schalk-Hihi C, et al. A novel
fluorogenic substrate for the measurement of endothelial lipase activity. Journal of lipid
research. 2011;52:374-82.
[162] Farber SA, Pack M, Ho SY, Johnson ID, Wagner DS, Dosch R, et al. Genetic analysis
of digestive physiology using fluorescent phospholipid reporters. Science. 2001;292:1385-8.
[163] Sciara G, Blangy S, Siponen M, Mc Grath S, van Sinderen D, Tegoni M, et al. A
topological model of the baseplate of lactococcal phage Tuc2009. The Journal of biological
chemistry. 2008;283:2716-23.
[164] Kabsch W. Xds. Acta Crystallogr D Biol Crystallogr. 2010;66:125-32.
RESEARCH ARTICLE
Inhibition of Type VI Secretion by an Anti-TssM Llama NanobodyVan Son Nguyen1,2☯, Laureen Logger3☯, Silvia Spinelli1,2, Aline Desmyter1,2,Thi Thu Hang Le1,2, Christine Kellenberger1,2, Badreddine Douzi1,2¤a, Eric Durand1,2¤b,Alain Roussel1,2, Eric Cascales3*, Christian Cambillau1,2*
1 Architecture et Fonction des Macromolécules Biologiques, Centre National de la Recherche Scientifique(CNRS)—UMR 7257, Marseille, France, 2 Architecture et Fonction des Macromolécules Biologiques, Aix-Marseille Université, Campus de Luminy, Case 932, Marseille, France, 3 Laboratoire d’Ingénierie desSystèmes Macromoléculaires, Institut de Microbiologie de la Méditerranée, Aix-Marseille Université, CNRS—UMR 7255, 31 chemin Joseph Aiguier, Marseille, France
☯ These authors contributed equally to this work.¤a Current address: Department of Biochemistry and Molecular Biology, Faculty of Medicine, 2350 HealthSciences Mall, Vancouver BC, Canada¤b Current address: G5 Biologie Structurale de la Sécrétion Bactérienne, CNRS UMR 3528, Institut Pasteur,Paris, France* [email protected] (CC); [email protected] (EC)
AbstractThe type VI secretion system (T6SS) is a secretion pathway widespread in Gram-negative
bacteria that targets toxins in both prokaryotic and eukaryotic cells. Although most T6SSs
identified so far are involved in inter-bacterial competition, a few are directly required for full
virulence of pathogens. The T6SS comprises 13 core proteins that assemble a large com-
plex structurally and functionally similar to a phage contractile tail structure anchored to the
cell envelope by a trans-membrane spanning stator. The central part of this stator, TssM, is
a 1129-amino-acid protein anchored in the inner membrane that binds to the TssJ outer
membrane lipoprotein. In this study, we have raised camelid antibodies against the purified
TssM periplasmic domain. We report the crystal structure of two specific nanobodies that
bind to TssM in the nanomolar range. Interestingly, the most potent nanobody, nb25, com-
petes with the TssJ lipoprotein for TssM binding in vitro suggesting that TssJ and the nb25
CDR3 loop share the same TssM binding site or causes a steric hindrance preventing
TssM-TssJ complex formation. Indeed, periplasmic production of the nanobodies displacing
the TssM-TssJ interaction inhibits the T6SS function in vivo. This study illustrates the power
of nanobodies to specifically target and inhibit bacterial secretion systems.
IntroductionThe type VI secretion system (T6SS) is a machinery widespread in Gram-negative bacteria anddedicated to the delivery of toxins in bacterial and eukaryotic host cells. By its anti-bacterial an-tagonistic action, the T6SS is one of the main players in the bacterial warfare for the access tonutrients and for colonization of the ecological niche [1, 2]. The T6SS assembles from 13
PLOSONE | DOI:10.1371/journal.pone.0122187 March 26, 2015 1 / 14
a11111
OPEN ACCESS
Citation: Nguyen VS, Logger L, Spinelli S, DesmyterA, Le TTH, Kellenberger C, et al. (2015) Inhibition ofType VI Secretion by an Anti-TssM Llama Nanobody.PLoS ONE 10(3): e0122187. doi:10.1371/journal.pone.0122187
Academic Editor: Erh-Min Lai, Academia Sinica,TAIWAN
Received: December 21, 2014
Accepted: February 12, 2015
Published: March 26, 2015
Copyright: © 2015 Nguyen et al. This is an openaccess article distributed under the terms of theCreative Commons Attribution License, which permitsunrestricted use, distribution, and reproduction in anymedium, provided the original author and source arecredited.
Data Availability Statement: All relevant data arewithin the paper.
Funding: This work was supported by grants fromthe Agence Nationale de la Recherche to EC (ANR-10-JCJC-1303-03) and from the Fondation pour laRecherche Médicale (FRM) to CC (DEQ2011-0421282). VSN is supported by a PhD grant from theFrench Embassy in Vietnam, and TTHL is supportedby a PhD grant from the University of Sciences andTechniques of 1 Hanoi (USTH). LL is supported by adoctoral fellowship from the French Ministry ofResearch. ED was supported by a FRM post-doctoralfellowship (FRM-SPF20101221116). The funders had
conserved components. Architecturally, the T6SS can be seen as a micrometer-long syringe an-chored to the cell membrane by a trans-envelope complex [3–6]. The phage tail-related sy-ringe-like tubular structure is composed of an internal tube tipped by a spike-like complex,wrapped by a contractile sheath and tethered to the membrane through contacts with compo-nents of a trans-envelope multiprotein complex [6, 7]. This membrane-associated complex iscomposed of the TssL and TssM inner membrane proteins and of the TssJ outer membrane li-poprotein [8–13]. The TssM and TssL proteins interact and stabilize each other and share ho-mologies with the Type IVb secretion system IcmF and DotU subunits respectively [12, 14,15]. In enteroaggregative Escherichia coli (EAEC), TssM (accession number: EC042_4539;gene ID: 387609960) is a 1129-amino-acid protein anchored to the inner membrane by threetransmembrane helices and bearing a large ~ 750 amino-acid periplasmic domain (amino-acids 386–1129). The C-terminal extremity of the TssM periplasmic domain interacts with theL1-2 loop of the TssJ lipoprotein with a KD of 2–4 μM [11]. By combining interactions withinner membrane and outer membrane-associated components, the TssM protein crosses thecell envelope and is therefore central to the T6SS membrane complex.
Although the EAEC TssM periplasmic domain purified readily, we did not succeed to gainstructural information [11]. One of the most efficient approaches to improve the crystallizationprocess is to use co-crystallization of the protein of interest with cognate camelid nanobodies.Camelid (llamas, dromaderies and alpacas) antibodies differ from classical antibodies as theyonly associate two heavy-chains, lacking the CH1 domain and terminated by monomeric vari-able antigen-binding VHH domains called nanobodies [16, 17, 18]. By contrast to the conven-tional immunoglobulin domains, these single-domain VHH antibodies are highly convenient:in addition to be the smallest antibodies, they are easy to produce in the E. coli periplasm [19].Therefore, they have remarkable potential in the biotechnology and bio-pharmaceutical fields[18, 20, 21]. More important for structural biologists, they also demonstrated their efficiency toimprove protein solubility and facilitating crystallization when complexed with the protein ofinterest [19], in particular for membrane-associated or flexible proteins [22, 23, 24, 25, 26]. Fi-nally, due to their high affinity and selectivity and their small size, nanobodies are excellent en-zymes and receptors inhibitors and can be used for functional studies.
To gain further information on the EAEC TssM protein, the purified TssM periplasmic do-main was used for llama immunization. Here we report the selection and the structural analysisof two specific nanobodies. These antibodies bind to the TssM periplasmic domain with a KD
in the nanomolar range. One of these nanobodies disrupts the TssM-TssJ interaction in vitroand prevents the proper function of the T6SS apparatus.
Results and Discussion
Selection and crystal structures of TssM-specific nanobodiesNanobodies were raised by immunization of llamas with the purified periplasmic domain ofthe EAEC TssM protein (TssMp). Three strong TssMp binders were identified from the im-mune library by three rounds of panning using phage display coupled to ELISA. Two nanobo-dies, called nb02 and nb25, were selected for further studies based on their high affinity forTssM and on their amino-acid differences in the variable regions, suggesting they bind distinctregions of TssMp (Fig. 1A). The third nanobody, nb42, is very similar to nb25, and was not re-tained for the structural studies. The two selected nanobodies, nb02 and nb25, sharing 77% se-quence identity, were produced in the periplasm of E. coli, purified to homogeneity andconcentrated to 10 mg/ml. Both nb02 and nb25 behaved as monomers in size-exclusion chro-matography and both crystallized readily, allowing to solve their three-dimensional structuresby molecular replacement (Fig. 1B and 1C). Crystal structures were refined to 1.7 Å and 1. 38
Nanobody-Mediated T6SS Inhibition
PLOS ONE | DOI:10.1371/journal.pone.0122187 March 26, 2015 2 / 14
no role in study design, data collection and analysis,decision to publish, or preparation of the manuscript.
Competing Interests: The authors have declaredthat no competing interests exist.
Å resolution with R/Rf values of 25.1/20.5% and 19.1/19.8%, respectively (Table 1). Overall, thetwo structures (nb02, PDB 4QLR; nb25, PDB 4QGY) are very similar. The superimposed struc-tures of nb02 and nb25 differ by a rmsd of 1.13 Å on 118 aligned residues (Fig. 1D). However,despite identical lengths, the conformations of the nb02 and nb25 CDR1 and CDR2 regionsdiffer significantly (Fig. 1D). More importantly, large differences can be observed on the CDR3regions, notably in terms of size (14-residue long for nb02, 19-residue long for nb25). Notewor-thy, the nb25 CDR3 region is stabilized by a disulfide bridge between Cys-106 and CDR2 Cys-50 and extends out of the nanobody core, providing a large putative binding area for the anti-gen (Fig. 1C).
Nanobodies bind TssM with nanomolar affinitiesTo gain further information on the nanobodies, the strengths of their interactions with TssMpwere measured by surface plasmon resonance (SPR). Nb02, nb25 and the nb25 variant nb42were covalently coupled to the CHIP and sensorgrams were recorded after injection of the pu-rified TssMp fragment in the microfluidic channel (Fig. 2). Analysis of the SPR curves indicatethat nb02, nb25 and nb42 bind to TssMp with KD values of 66.8±2 nM, 1.61± 0.1 nM and1.76± 0.1 nM, respectively (Table 2).
Nb25 interferes with TssJ binding to TssMWe previously reported that TssJ binds to TssMp with a KD value of 2–4 μM. This interactionis mediated by the L1-2 loop of TssJ and is essential for the proper function of the T6SS [11].
Fig 1. Structure of the anti-TssM nanobodies. (A) Sequence alignment of nanobodies nb02, nb25 andnb42. Identical amino-acids are in white and underlined in red, similar amino-acids are colored in red,different amino-acids are in black. The CDR1, CDR2 and CDR3 are highlighted in blue, green and yellow,respectively. The green numbers (1 or 2) below the sequence indicate cysteine residues involved indisulphide bridges SS1 (1, Cys22 and Cys96) and SS2 (2, Cys50 and Cys106) disulphide bridges formation.Crystal structure of nanobodies nb02 (B) and nb25 (C), represented as ribbons and colored in rainbowmode.(D) Superimposition of the structures of nb02 (grey) and nb25 (rainbow color). The locations of the CDR1,CDR2 and CDR3 variable regions are indicated as well as the positions of disulphide bridges (SS1 and SS2).
doi:10.1371/journal.pone.0122187.g001
Nanobody-Mediated T6SS Inhibition
PLOS ONE | DOI:10.1371/journal.pone.0122187 March 26, 2015 3 / 14
Based on these results, we hypothesized that the TssJ L1-2 loop will contact a crevice withinTssM [11]. Because nanobodies are known to target enzymatic sites or crevices [27], we soughtto determine whether the nanobodies and TssJ share the same TssM-binding site. We thereforeperformed Bilayer interferometry (BLI) competition experiments between the nanobodies andTssJ on TssMp. The TssJ protein (devoid of its N-terminal Cys acylation residue) was biotiny-lated and coupled to the streptavidine BLI chip. The chip was immersed on solution containingTssMp-nanobody complexes. Due to the 50- to 1000-fold magnitude differences between theKDs of TssJ and that of the nanobodies to TssMp, we expected that the occupation of the TssJbinding site by the nanobody would prevent TssMp binding to TssJ. Fig. 3 shows that theTssMp-nb02 complex can readily interact with TssJ. The formation of the nb02-TssMp-TssJternary complex demonstrates that TssJ and nb02 bind TssM differently. By contrast, theTssMp-nb25 complex does not bind TssJ demonstrating that nb25 interferes with TssJ bindingon TssMp. This result suggests that nb25 and TssJ share the same TssM binding site or thatnb25 binding on TssMp causes steric hindrance preventing TssJ binding.
Nb25 disrupts the TssMp-TssJ interaction in vitroWe hypothesized that if nb25 and TssJ share the same TssM binding site, nb25 should disruptthe TssM-TssJ complex. We therefore analyzed the effect of nb25 on the stability of theTssM-TssJ complex. Competition experiments were performed by incubating the purifiedTssMp-TssJ complex with an excess of nb25, prior to analysis by gel filtration. Three peakswere observed on the chromatogram (Fig. 4). Analysis of the peaks by SDS-PAGE (Fig. 4,inset) showed that peak 1 (~ 90 kDa) contained TssMp and nb25 (theoretical weight of theTssMp-nb25 complex = 98 kDa), while peak 2 (~ 18 kDa) contained TssJ (theoretical
Table 1. Data collection and refinement statistics for the nb02 and nb25 anti-TssM nanobodies.
DATA COLLECTION nb02 (PDB:4QLR) nb25 (PDB:4QGY)
Diffraction source ESRF Soleil PX1
Detector Pilatus 6M Pilatus 6M
Space group P 21 C2221Cell dimensions (Å,°) a = 50.0, b = 48.6, c = 52.9 Å3 β = 118.8° a = 52.0, b = 70.9, c = 145.7 Å3
Resolution rangea (Å) 50–1.70 (1.76–1.70) 50–1.38 (1.42–1.38)
R-mergea (%) 7.0 (52) 3.1 (70)
CC(1/2)a (%) 99.7 (73.1) 100 (85)
Mean((I)/sd(I))a 9.6 (1.7) 29 (2.6)
Total number of reflectionsa 62041 (5333) 395611 (25318)
Number of unique reflectionsa 23496 (2222) 55440 (3886)
Completenessa (%) 94.3 (90.3) 99.6 (95)
Multiplicitya 2.64 (2.4) 7.1 (6.5)
REFINEMENT
Resolutiona (Å) 44.1–1.7 (1.78–1.7) 20–1.38 (1.42–1.38)
Nr of reflectionsa 23498 (2742) 55416 (3814)
Nr protein / water 1931 / 211 1918 / 309
Nr test set reflections 1201 2813
Rwork/Rfreea (%) 20.5/25.1 (24.0/25.4) 19.1/19.8 (32.2/36.7)
r.m.s.d.bonds (Å) / angles (°) 0.01 / 1.02 0.01 / 1.09
B-wilson / B-average 21.0 / 23.6 20.3 / 24.8
a numbers in brackets refer to the highest resolution bin.
doi:10.1371/journal.pone.0122187.t001
Nanobody-Mediated T6SS Inhibition
PLOS ONE | DOI:10.1371/journal.pone.0122187 March 26, 2015 4 / 14
Fig 2. Nanobodies nb02 and nb25 bind TssMpwith nanomolar affinity. Surface Plasmon Resonance recordings representing binding and release of thepurified periplasmic domain of TssM (from bottom to top: 1.95, 3.9, 7.8, 15.6, 31.25, 62.5, 125 and 250 nM) to nanobody nb02 (A), nb25 (B) or nb42 (C)immobilized on the Chip. The variation of the Surface Plasmon Resonance (shown as the experimental and fitting curves) is reported on the y axis (inarbitrary unit, RU) plotted versus the reaction time on the x axis (in sec.). The apparent KDs are indicated on the top of each graph. The kinetic andthermodynamic parameters are indicated in Table 2.
doi:10.1371/journal.pone.0122187.g002
Table 2. Kinetic and thermodynamic parameters of the interactions between anti-TssM nanobodieswith TssMp.
kon (M-1.s-1) koff (s-1) KD (nM) Rmax
nb02 6.02 105 402 10-4 66.8 ± 2.5 734
nb25 3.5 105 5.6 10-4 1.61 ± 0.05 1165
nb42 2.04 105 3.6 10-4 1.76 ± 0.08 3155
doi:10.1371/journal.pone.0122187.t002
Nanobody-Mediated T6SS Inhibition
PLOS ONE | DOI:10.1371/journal.pone.0122187 March 26, 2015 5 / 14
Fig 3. Binding of TssMp:nanobody complexes to TssJ. Bilayer interferometry recordings representingbinding of TssMp alone (blue) or the TssMp:nb02 (red) or TssMp:nb25 (green) complexes to chip coupled toTssJ. The response (in nm) is plotted versus the time (in sec.). The absence of response in the greensensorgram indicates that nb25 prevents attachment of TssMp to TssJ.
doi:10.1371/journal.pone.0122187.g003
Fig 4. Nanobody nb25 disrupts the TssMp:TssJ complex. The pre-formed TssM-TssJ complex was analyzed by size exclusion chromatography before(dash line) and after incubation with an excess of nb25 (plain line). The composition of the different complexes, eluting at different volumes, was analysed bySDS-PAGE and Coomassie blue staining (inset). The fractions corresponding to peak 1–3 are indicated on the top. The molecular masses of the proteinmarkers (in kDa) are indicated on the left. According to the calibration of the column, the apparent molecular weights for peak 1 (9.72 ml), peak 2 (13.02 ml)and peak 3 (14.43 ml) are about 90, 18 and 14 kDa, respectively. The faint band at the position of TssMp in peaks 2 and 3 of the SDS gel arise fromcontamination by the large main TssMp peak 1.
doi:10.1371/journal.pone.0122187.g004
Nanobody-Mediated T6SS Inhibition
PLOS ONE | DOI:10.1371/journal.pone.0122187 March 26, 2015 6 / 14
Fig 5. TssM-specific nanobodies nb25 and nb42 specifically affect T6SS function. (A) Hcp release assay. HA-epitope-tagged Hcp (HcpHA) release wasassessed by separating whole cells (C) and supernatant (S) fractions from wild-type (WT), ΔtssM cells or WT cells producing 6His-tagged nanobodies asindicated. 2×108 cells and the TCA-precipitated material of the supernatant from 5×108 cells were loaded on a 12.5%-acrylamide SDS-PAGE and thenanobodies (lower panel), Hcp (middle panel) and periplasmic TolB (cell integrity control, upper panel) proteins were immunodetected using anti-5His, anti-HA and anti-TolB antibodies respectively. (B) Anti-bacterial assay. The Sci-1 T6SS-dependent anti-bacterial activity was assessed by mixing prey cells(W3110 gfp+, kanR) with the indicated attacker cell (K-12, W3110; WT, EAEC 17–2; ΔtssM, 17–2ΔtssM or WT cells producing the indicated nanobody) for 16hours at 37°C in sci-1-inducing medium (SIM). The recovered fluorescent level (in arbitrary units) is shown in the upper graph (mean of fluorescence levelsper OD600nm obtained from four independent experiments). The number of recovered viable prey cells, expressed in colony forming unit (cfu) is shown in thelower graph (the triangles indicate values from four independent assays, and the average is indicated by the bar).
doi:10.1371/journal.pone.0122187.g005
Nanobody-Mediated T6SS Inhibition
PLOS ONE | DOI:10.1371/journal.pone.0122187 March 26, 2015 7 / 14
weight = 17 kDa) and peak 3 (~ 14 kDa), the excess of nb25 (theoretical weight = 14 kDa).These data confirm that nb25 binds to TssMp and chases TssJ from the pre-assembled TssMp-TssJ complex.
Nb25 prevents formation of a functional T6SSThe TssM-TssJ interaction was previously shown to be indispensable for the proper functionof the Type VI secretion apparatus in enteroaggregative E. coli [11]. As nb25 (and nb42) com-pete with TssJ for TssM binding in vitro, we wondered whether these nanobodies would dis-rupt the TssM-TssJ interaction in vivo. The vectors allowing the periplasmic production of thenb02, nb25 and nb42 nanobodies, as well as a control nanobody (a nanobody targeting the Lac-tococcus lactis phage 1358 receptor binding protein) were introduced into the wild-type EAECstrain 17–2. The function of the T6SS was assessed by the Hcp1 release assay and by the Sci1--dependent antibacterial activity. As shown in Fig. 5, the control nanobody and nb02 had no—or little—effect on the function of the T6SS as 17–2 cells producing nb02 released the Hcp1protein and retained anti-bacterial activities at levels comparable to that of wild-type 17–2cells. By contrast, the periplasmic production of nb25 and nb42 inhibited the function of theT6SS as (i) no Hcp1 was found in the culture supernatant of 17–2 cells producing these VHHsand (ii) these cells did not have any growth advantage when co-cultured with E. coli K-12 cells.Taken together, these results suggest that disruption of the TssM-TssJ interaction prevents theformation of a functional Type VI secretion apparatus in EAEC.
Concluding RemarksIn this study, we raised and selected two camelid variable fragments of heavy-chain antibodiesspecific to the EAEC T6SS TssM periplasmic domain. Although this strategy was originally ini-tiated to help the crystallization of the TssM protein, we used here these nanobodies to gainfunctional information on the assembly of the T6SS apparatus. We showed that the two select-ed nanobodies bind TssMp with nanomolar affinities. These values are in agreement with theaffinities of nanobodies, including that targeting the Salmonella Typhimurium actin ADP-ribosylating toxin or the Pseudomonas aeruginosa flagellin [28, 29]. BLI and gel filtration ex-periments showed that nb25 is a competitor of TssJ in vitro, suggesting that nb25 binds at—orclose to—the TssJ binding site on TssMp or at least causes a steric hindrance or a TssM confor-mational change preventing TssM-TssJ complex formation. Due to the variability of the CDR3protruding loop and the observation that nanobodies are often targeting crevices [27], we hy-pothesize that the nb25 CDR3 loop mediates the interaction with a crevice in TssM, such as theTssJ L1-2 loop does [11]. Significant sequence differences occur in the CDR1 and CDR2 ofnb25 and nb42, while the CDR3 are identical. This observation is in line with our hypothesis ofCDR3 binding to the TssJ binding site of TssM. Sequence alignment between the TssJ LI-2loop and the CDR1, CDR2 and CDR3 of nb25 shows that, except an A-X-G-I motif, very limit-ed similarities exist between TssJ and CDR3. However, the TssJ L1-2 loop being elongatedwhereas the CDR3 having a compact fold, no structural conservation is observed betweenthese loops suggesting that TssJ L1-2 and nb25/nb42 CDR3 do not bind identically to TssM, al-though we cannot exclude loop rearrangements upon binding. In view of these results, the co-crystallization between TssMp and nb25 is therefore essential, not only to gain structural infor-mation on TssM but also to identify the nb25-binding site and to test whether this region isalso responsible for making contacts with the TssJ L1-2 loop. Interestingly, the use of nanobo-dies for co-crystallization was successful in this specific case as we recently obtained co-crystalsbetween a fragment of the periplasmic domain of TssM and nb25 and performed a preliminaryX-ray diffraction analysis [30].
Nanobody-Mediated T6SS Inhibition
PLOS ONE | DOI:10.1371/journal.pone.0122187 March 26, 2015 8 / 14
By contrast, although nb02 has strong affinity for TssM, we showed that nb02 does notcompete with TssJ for TssM binding in vitro. Because of the formation of an nb02-TssMp-TssJternary complex, nb02 is an interesting tool to assist the crystallization of the TssMp-TssJ com-plex. Hcp release and growth competition assays showed that nb02 has no inhibitory effect invivo. Therefore, defining the site of binding of nb02 on TssM is also interesting as this nano-body targets a region that is not critical for the assembly of the T6SS membrane complex.
Nanobodies targeting bacterial multi-protein complexes or toxins have already been re-ported. For examples, specific nanobodies to the Vibrio Type II secretion (T2SS) GspD secretinand EspI/J pseudopilin complex have been selected and used for structural studies but their ef-fect on the assembly of the T2SS in vivo has not been addressed [31]. By contrast, nanobodiesinteracting with the P. aeruginosa or Campylobacter jejuni flagellin decrease the swimming rateand biofilm formation of these strains [29, 32]. Similarly, nanobodies that bind to and interferewith the catalytic site of the S. Typhimurium ADP-ribosylation toxin diminish the cytotoxicityin vivo [28]. Our study not only paves the way to the structural characterization of the TssMperiplasmic domain and of the TssMp-TssJ complex, but also demonstrates that nanobodiescan be used to inhibit the assembly of the T6SS or of secretion systems in general with the goalto diminish or abolish the fitness or the virulence of bacterial pathogens in theirecological niche.
Materials and Methods
Bacterial strains, growth conditions and chemicalsThe entero-aggregative E. coli EAEC strain 17–2 and its ΔtssM isogenic derivative [10] wereused for this study. The E. coli K-12 WK6 strain was used for nanobody production. The E. coliK-12 W3110 strain carrying the pUA66-rrnB plasmid (gfp under the control of the constitutiverrnB ribosomal promoter, specifying strong and constitutive fluorescence, and kanamycin re-sistance [33]) was used as prey in antibacterial competition experiments. Strains were routinelygrown in lysogeny broth (LB) or Terrific broth at 37°C, with aeration. For antibacterial compe-tition assays, cells were grown in sci-1-inducing medium (SIM: M9 minimal medium, glycerol0.2%, vitamin B1 1 μg/ml, casaminoacids 100 μg/ml, LB 10%, supplemented or not with bac-toagar 1.5%). Plasmids were maintained by the addition of ampicillin (200 μg/mL) or kanamy-cin (50 μg/mL). The expression of the nanobody constructs cloned into pHEN6 vectorderivatives was induced by the addition of isopropyl-β-thio-galactoside (IPTG).
Generation of llama nanobodies against TssMThe periplasmic domain of TssM (amino-acids 386–1129; TssMp) was produced and purifiedas described previously [11]. Four injections of 1 mg of recombinant TssMp (in Tris-HCl 20mM (pH 8.0), NaCl 150 mM) were performed subcutaneously with two weeks intervals fol-lowed by a fifth injection one month later in two llamas (Lama glama) (Capralogics Inc., Hard-wick, MA 01037 USA). Lymphocytes were isolated from blood sample obtained 1 week afterthe last immunization. cDNA was synthesized from purified total RNA by reverse transcrip-tion. The cDNA was used as template for PCR amplification to amplify sequences correspond-ing to the variables domains of the heavy-chain antibodies. PCR fragments were cloned intothe phagemid vector pHEN4 [34] to create a nanobody phage display library. Selection andscreening of nanobodies were performed as previously published [35]. Three rounds of pan-ning resulted in the isolation of TssMp-specific binders. Nanobodies nb02, nb25 and its variantnb42 were selected, sequenced and cloned into the pHEN6 expression vector downstream thepelB signal peptide and fused to a C-terminal 6×His tag [36].
Nanobody-Mediated T6SS Inhibition
PLOS ONE | DOI:10.1371/journal.pone.0122187 March 26, 2015 9 / 14
Purification of nanobodiesE. coliWK6 cells carrying the pHEN6 derivatives were grown at 37°C in Terrific Broth mediumcontaining 0.1% glucose and ampicillin to an optical density (OD600nm) ~ 0.8. The expressionof the nanobody was induced by the addition of 1 mM IPTG and incubation for 16 hours at28°C. The periplasmic fraction containing the nanobodies was prepared by osmotic shock. TheHis-tail-containing fusion proteins were purified by immobilized metal affinity chromatogra-phy on a 5-mL Ni-NTA column equilibrated in 50 mMNa/K phosphate (pH 8.0), 300 mMNaCl, 10% glycerol. Nanobodies were eluted in 250mM imidazole and concentrated (Amicon-Ultra 10-kDa cut-off) prior to be loaded on a HiLoad 16/60 Superdex 75 gel filtration columnequilibrated in 20 mM Tris–HCl (pH 8.0), 150 mMNaCl.
Nanobodies structure determinationCrystallization screening experiments were performed with several commercial kits. The nano-drop crystallization experiments were performed in Greiner plates. The reservoirs of the Grei-ner plates were filled up using a TECAN pipetting robot, while the nano-L drops were dis-pensed by a Mosquito robot using nano-crystallization protocols [37]. All crystallizationexperiments were performed at 293K. Nanobody nb02 crystallized at 15 mg/mL in 100 mM so-dium cacodylate (pH 6.5), 25% w/v PEG 8000, 200 mM (NH4)2SO4. Nanobody nb25 crystal-lized at 15 mg/mL in 100 mM CHES (pH 10), 1 M K/Na tartrate, 200 mM Li2SO4. Crystalswere mounted in loops and soaked in their crystallization solution supplemented by 4MTMAO before cryocooling.
Data collection of nb02 and nb25 were performed at synchrotrons ESRF (Grenoble, France)beamline ID29 and Soleil PX1 (saint-Aubin, France), respectively (Table 1). Data were inte-grated with XDSME and scaled with XSCALE. The nb02 crystals belong to space-group P21,with cell dimensions a = 50.0, b = 48.6, c = 52.9 Å3, β = 118.8°. Two molecules in the symmetricunit yield a Vm value of 2.0 Å3/Da and 38% solvent. The nb25 crystals belong to space-groupC2221, with cell dimensions a = 52.0, b = 70. 9, c = 145.7 Å3. Two molecules in the asymmetricunit yielded a Vm value of 2.20 Å3/Da and 45% solvent. These crystals diffracted to 1.70 Å and1.38 Å resolution, respectively (Table 1). Molecular replacement was performed using MOL-REP [38] and nanobody structures with high sequence similarity to nb25 (PDB: 4KRP) andnb02 (PDB: 4HEP) as starting models. Refinement was performed using autoBUSTER [39] al-ternated with rebuilding with COOT [40].
Accession numbersAtomic coordinates have been deposited in the Protein Data Bank as identifiers 4QLR (nb02)and 4QGY (nb25).
Surface Plasmon Resonance (SPR) measurementsThe interaction between TssMp and nanobodies was tested by SPR using a Biacore X100 (GEhealthcare). Nb02, nb25 and nb42 were covalently linked to a CM5 chip and different concen-trations of TssMp were injected in the microfluidic channel. Regeneration was achieved by theinjection of glycine-HCl (pH 2.5) buffer. The KD values were calculated by the multi-cycle ki-netics method (Biacore, GE-healthcare).
Biolayer interferometry (BLI)TssJ was first biotinylated using the EZ-Link NHS-PEG4-Biotin kit (Perbio Science, France).The reaction was stopped by removing the excess of the biotin using a Zeba Spin Desalting
Nanobody-Mediated T6SS Inhibition
PLOS ONE | DOI:10.1371/journal.pone.0122187 March 26, 2015 10 / 14
column (Perbio Science, France). BLI studies were performed in black 96-well plates (Greiner)at 25°C using an OctetRed96 (ForteBio, USA). Streptavidin biosensor tips (ForteBio, USA)were first hydrated with 0.2 ml Kinetic buffer (KB, ForteBio, USA) for 20 min and then loadedwith biotinylated TssJ (10 μg/ml in KB). The association of TssJ with TssM (300 nM) was mon-itored for 200 sec, as well as the dissociation in KB. For TssM/nanobody competition experi-ments, nb02 or nb05 nanobodies were used at a TssJ:nanobody ratio of 3:1.
Size Exclusion Chromatography (SEC)Size exclusion chromatography was performed on an Alliance 2695 HPLC system (Waters)using a pre-calibrated Superdex 75 10/300 GL gel filtration column run in 20 mM Tris-HCl(pH 8.0), 100 mM NaCl at 0.5 mL/min.
Hcp release assayThe expression of the nanobody constructs cloned into pHEN6 vector derivatives was inducedby the addition of 0.5 mM isopropyl-β-thio-galactoside (IPTG) for 45 min. Supernatant andcell fractions were separated as previously described [12, 13] with the addition of the iron che-lator 2,2’-dipyridyl (125 μM final concentration) to the culture medium 30 min before harvest-ing the cells to induce the expression of the sci-1 T6SS [41]. Briefly, 2 × 109 cells producing HAepitope-tagged Hcp (from plasmid pOK-HcpHA; [10]) and the 6×His-tagged nanobody con-structs were harvested and collected by centrifugation at 2,000 × g for 5 min. The supernatantfraction was then subjected to a second low-speed centrifugation and then at 16,000 × g for 15min. The supernatant was filtered on sterile polyester membranes with a pore size of 0.2 μm(membrex 25 PET, membraPure GmbH) before overnight precipitation with trichloroaceticacid (TCA) 15% on ice. Cells and precipitated supernatants were resuspended in loading bufferand analyzed by Sodium Dodecyl Sulfate-Polyacrylamide Gel Electrophoresis (SDS-PAGE)and immunoblotting with the anti-HA antibody. As control for cell lysis, Western blots wereprobed with antibodies raised against the periplasmic TolB protein. The production of thenanobodies was verified by immunoblotting with the anti-5His antibody.
Sci-1-dependent antibacterial assayThe antibacterial competition growth assay was performed as described for the studies on theCitrobacter rodentium and EAEC Sci-2 T6SSs [42, 43] with modifications. The wild-type E. colistrain W3110 bearing the pUA66-rrnB plasmid KanR [33] was used as prey in the competitionassay. The pUA66-rrnB plasmid provides a strong constitutive green fluorescent (GFP+) phe-notype. Attacker and prey cells were grown for 16 hours in LB medium, then diluted 100-foldin SIM. Once the culture reached an OD600nm = 0.8, the cells were harvested and resuspendedto an OD600nm of 0.5 into SIM. For the attacker cells, the expression of the VHH-encodinggenes was induced for 45min. with 0.5 mM IPTG prior to cell harvest. Attacker and prey cellswere mixed to a 4:1 ratio and 25-μl drops of the mixture were spotted in triplicate onto a pre-warmed dry SIM agar plate supplemented with 0.5 mM IPTG. After overnight incubation at30°C, the bacterial spots were cut off, and cells were resuspended in LB to an OD600nm of 0.5.Triplicates of 150 μl were transferred into wells of a black 96-well plate (Greiner), and the ab-sorbance at 600 nm and fluorescence (excitation, 485 nm; emission, 530 nm) were measuredwith a Tecan Infinite M200 microplate reader. The relative fluorescence was expressed as theintensity of fluorescence divided by the absorbance at 600 nm, after subtracting the values of ablank sample. For enumeration of viable prey cells, bacterial suspensions recovered from thespots were serially diluted and spotted onto selective LB agar plates supplemented with kana-mycin (for the E. coli prey cells).
Nanobody-Mediated T6SS Inhibition
PLOS ONE | DOI:10.1371/journal.pone.0122187 March 26, 2015 11 / 14
MiscellaneousSDS-Polyacrylamide gel electrophoresis was performed using standard protocols. For immu-nostaining, proteins were transferred onto nitrocellulose membranes, and immunoblots wereprobed with primary antibodies, and goat secondary antibodies coupled to alkaline phospha-tase, and developed in alkaline buffer in presence of 5-bromo-4-chloro-3-indolylphosphateand nitroblue tetrazolium. The anti-TolB polyclonal antibodies are from our laboratory collec-tion, while the anti-HA (3F10 clone, Roche), the anti-5His (Qiagen) monoclonal antibodiesand alkaline phosphatase-conjugated goat anti-rabbit, mouse, or rat secondary antibodies(Millipore) have been purchased as indicated.
Ethics statementPlease note that Eric Cascales is a PLOS ONE Academic Editor. This does not alter the authors’adherence to all the PLOS ONE policies on sharing data and material.
AcknowledgmentsWe thank the ERSF and Soleil Synchrotron radiation facilities for beamline allocation, NicolasFlaugnatti and Laure Journet for the development of the Sci-1-dependent antibacterial compe-tition assay, the members of the Cascales and Cambillau research groups for helpful discus-sions, Olivier Uderso, Isabelle Bringer and Annick Brun for technical assistance, and AmarDissoir for encouragements.
Author ContributionsConceived and designed the experiments: AR EC CC. Performed the experiments: VSN LL SSAD TTHL CK BD ED AR EC CC. Analyzed the data: VSN LL SS AD TTHL CK BD ED AR ECCC. Wrote the paper: VSN AD CK AR EC CC.
References1. Russell AB, Peterson SB, Mougous JD. Type VI secretion system effectors: poisons with a purpose.
Nat Rev Microbiol 2014; 12: 137–148. doi: 10.1038/nrmicro3185 PMID: 24384601
2. Durand E, Cambillau C, Cascales E, Journet L. VgrG, Tae, Tle, and beyond: the versatile arsenal ofType VI secretion effectors. Trends Microbiol 2014; 22: 498–507. doi: 10.1016/j.tim.2014.06.004PMID: 25042941
3. Cascales E, Cambillau C. Structural biology of type VI secretion systems. Philos Trans R Soc Lond BBiol Sci 2012; 367: 1102–1111. doi: 10.1098/rstb.2011.0209 PMID: 22411981
4. Coulthurst SJ. The Type VI secretion system—a widespread and versatile cell targeting system. ResMicrobiol 2013; 164: 640–654. doi: 10.1016/j.resmic.2013.03.017 PMID: 23542428
5. Ho BT, Dong TG, Mekalanos JJ. A View to a Kill: The Bacterial Type VI Secretion System. Cell Host Mi-crobe 2014; 15: 9–21. doi: 10.1016/j.chom.2013.11.008 PMID: 24332978
6. Zoued A, Brunet YR, Durand E, Aschtgen MS, Logger L, Douzi B, et al. Architecture and assembly ofthe Type VI secretion system. Biochim Biophys Acta 2014; 1843: 1664–1673. doi: 10.1016/j.bbamcr.2014.03.018 PMID: 24681160
7. Zoued A, Durand E, Bebeacua C, Brunet YR, Douzi B, Cambillau C, et al. TssK is a trimeric cytoplasmicprotein interacting with components of both phage-like and membrane anchoring complexes of theType VI secretion system. J Biol Chem 2014; 288: 27031–27041. doi: 10.1074/jbc.M113.499772PMID: 23921384
8. Aschtgen MS, Bernard CS, De Bentzmann S, Lloubes R, Cascales E. SciN is an outer membrane lipo-protein required for type VI secretion in enteroaggregative Escherichia coli. J Bacteriol 2008; 190:7523–7531. doi: 10.1128/JB.00945-08 PMID: 18805985
9. Ma LS, Lin JS, Lai EM. An IcmF family protein, ImpLM, is an integral inner membrane protein interactingwith ImpKL, and its walker a motif is required for type VI secretion system-mediated Hcp secretion in
Nanobody-Mediated T6SS Inhibition
PLOS ONE | DOI:10.1371/journal.pone.0122187 March 26, 2015 12 / 14
Agrobacterium tumefaciens. J Bacteriol 2009; 191: 4316–4329. doi: 10.1128/JB.00029-09 PMID:19395482
10. Aschtgen MS, Gavioli M, Dessen A, Lloubes R, Cascales E. The SciZ protein anchors the enteroaggre-gative Escherichia coli Type VI secretion system to the cell wall. Mol Microbiol. 2010; 75: 886–899. doi:10.1111/j.1365-2958.2009.07028.x PMID: 20487285
11. Felisberto-Rodrigues C, Durand E, Aschtgen MS, Blangy S, Ortiz-Lombardia M, Douzi B, et al. To-wards a structural comprehension of bacterial type VI secretion systems: characterization of the TssJ-TssM complex of an Escherichia coli pathovar. PLoS Pathog 2011; 7: e1002386. doi: 10.1371/journal.ppat.1002386 PMID: 22102820
12. Durand E, Zoued A, Spinelli S, Watson PJ, Aschtgen MS, Journet L, et al. Structural characterizationand oligomerization of the TssL protein, a component shared by bacterial type VI and type IVb secretionsystems. J Biol Chem 2012; 287: 14157–14168. doi: 10.1074/jbc.M111.338731 PMID: 22371492
13. AschtgenMS, Zoued A, Lloubes R, Journet L, Cascales E. The C-tail anchored TssL subunit, an essen-tial protein of the enteroaggregative Escherichia coli Sci-1 Type VI secretion system, is inserted byYidC. Microbiologyopen 2012; 1: 71–82. doi: 10.1002/mbo3.9 PMID: 22950014
14. Bingle LE, Bailey CM, Pallen MJ. Type VI secretion: a beginner's guide. Curr Opin Microbiol 2008; 11:3–8. doi: 10.1016/j.mib.2008.01.006 PMID: 18289922
15. Cascales E. The type VI secretion toolkit. EMBO Rep 2008; 9: 735–741. doi: 10.1038/embor.2008.131PMID: 18617888
16. Hamers-Casterman C, Atarhouch T, Muyldermans S, Robinson G, Hamers C, Songa EB, et al. Natural-ly occurring antibodies devoid of light chains. Nature 1993; 363: 446–448. PMID: 8502296
17. Muyldermans S, Cambillau C, Wyns L. Recognition of antigens by single-domain antibody fragments:the superfluous luxury of paired domains. Trends Biochem Sci 2001; 26: 230–235. PMID: 11295555
18. Muyldermans S. Nanobodies: Natural Single-Domain Antibodies. Ann Rev Bioch, 2013; 82: 775–797.doi: 10.1146/annurev-biochem-063011-092449 PMID: 23495938
19. Pardon E, Laeremans T, Triest S, Rasmussen SG, Wohlkonig A, et al. A general protocol for the gener-ation of Nanobodies for structural biology. Nat Protoc 2014; 9: 674–693. doi: 10.1038/nprot.2014.039PMID: 24577359
20. Muyldermans S, Baral TN, Retarnozzo VC, De Baetselier P, De Genst E, Kinne J, et al. Camelid immu-noglobulins and nanobody technology. Vet Immunol Immunopath 2009; 128: 178–183. doi: 10.1016/j.vetimm.2008.10.299 PMID: 19026455
21. Steyaert J. Structural investigation of GPCR transmembrane signaling by use of nanobodies. Eur Bio-phys J. 2011; 40: 223–223.
22. Rasmussen SGF, Choi HJ, Fung JJ, Pardon E, Casarosa P, Chae PS, et al. Structure of a nanobody-stabilized active state of the beta(2) adrenoceptor. Nature 2011; 469: 175–180. doi: 10.1038/nature09648 PMID: 21228869
23. Ring AM, Manglik A, Kruse AC, Enos MD, Weis WI, Garcia KC, et al. Adrenaline-activated structure ofb2-adrenoceptor stabilized by an engineered nanobody. Nature 2013; 502:575–579. doi: 10.1038/nature12572 PMID: 24056936
24. Hassaine G, Deluz C, Grasso L, Wyss R, Tol MB, Hovius R, et al. X-ray structure of the mouse seroto-nin 5-HT3 receptor. Nature 2014; 512: 276–281. doi: 10.1038/nature13552 PMID: 25119048
25. Lam AY, Pardon E, Korotkov KV, Hol WGJ, Steyaert J. Nanobody-aided structure determination of theEpsI:EpsJ pseudopilin heterodimer from Vibrio vulnificus. J Struct Biol 2009; 166: 8–15. doi: 10.1016/j.jsb.2008.11.008 PMID: 19118632
26. Desmyter A, Spinelli S, Roussel A, Cambillau C. Camelid nanobodies: killing two birds with one stone.Curr Opin Struct Biol 2015; 32C: 1–8. doi: 10.1016/j.sbi.2015.01.001 PMID: 25614146
27. Desmyter A, Transue TR, Ghahroudi MA, Thi MHD, Poortmans F, Hamers R, et al. Crystal structure ofa camel single-domain V-H antibody fragment in complex with lysozyme. Nat Struct Biol 1996; 3:803–811. PMID: 8784355
28. Alzogaray V, DanquahW, Aguirre A, Urrutia M, Berguer P, Garcia Vescovi E, et al. Single-domainllama antibodies as specific intracellular inhibitors of SpvB, the actin ADP-ribosylating toxin of Salmo-nella typhimurium. Faseb J. 2011; 25: 526–534. doi: 10.1096/fj.10-162958 PMID: 20940265
29. Adams H, HorrevoetsWM, Adema SM, Carr HEV, vanWoerden RE, Koster M, et al. Inhibition of biofilmformation by Camelid single-domain antibodies against the flagellum of Pseudomonas aeruginosa. JBiotech 2014; 186: 66–73. doi: 10.1016/j.jbiotec.2014.06.029 PMID: 24997356
30. Nguyen VS, Spinelli S, Desmyter A, Le TTH, Kellenberger C, Cascales E, et al. Production, crystalliza-tion and X-ray diffraction analysis of a complex between a fragment of the TssM T6SS protein and acamelid nanobody. Acta Crystallogr F. 2015; 71:in press
Nanobody-Mediated T6SS Inhibition
PLOS ONE | DOI:10.1371/journal.pone.0122187 March 26, 2015 13 / 14
31. Korotkov KV, Pardon E, Steyaert J, Hol WGJ. Crystal Structure of the N-Terminal Domain of the Secre-tin GspD from ETEC Determined with the Assistance of a Nanobody. Structure 2009; 17: 255–265.doi: 10.1016/j.str.2008.11.011 PMID: 19217396
32. Hussack G, Riazi A, Ryan S, van Faassen H, MacKenzie R, Tanha J, et al. Protease-resistant single-domain antibodies inhibit Campylobacter jejuni motility. Protein Eng Des Select 2014; 27: 191–198.doi: 10.1093/protein/gzu011 PMID: 24742504
33. Zaslaver A, Bren A, Ronen M, Itzkovitz S, Kikoin I, Shavit S, et al. A comprehensive library of fluores-cent transcriptional reporters for Escherichia coli. Nat Methods 2006; 3: 623–628. PMID: 16862137
34. Ghahroudi MA, Desmyter A, Wyns L, Hamers R, Muyldermans S. Selection and identification of singledomain antibody fragments from camel heavy-chain antibodies. Febs Letters 1997; 414: 521–526.PMID: 9323027
35. Desmyter A, Farenc C, Mahony J, Spinelli S, Bebeacua C, Blangy S, et al. Viral infection modulationand neutralization by camelid nanobodies. Proc Natl Acad Sci U S A 2013; 110: 1371–1379.
36. Conrath K, Pereira AS, Martins CE, Timoteo CG, Tavares P, Spinelli S, et al. Camelid nanobodiesraised against an integral membrane enzyme, nitric oxide reductase. Prot Sci 2009; 18: 619–628.
37. Sulzenbacher G, Gruez A, Roig-Zamboni V, Spinelli S, Valencia C, Pagot F, et al. A medium-through-put crystallization approach. Acta Crystallogr D Biol Crystallogr 2002; 58: 2109–2115. PMID:12454472
38. Vagin A, Teplyakov A. Molecular replacement with MOLREP. Acta Crystallogr D Biol Crystallogr 2010;66: 22–25. doi: 10.1107/S0907444909042589 PMID: 20057045
39. Blanc E, Roversi P, Vonrhein C, Flensburg C, Lea SM, Bricogne G. Refinement of severely incompletestructures with maximum likelihood in BUSTER-TNT. Acta Crystallogr D Biol Crystallogr 2004; 60:2210–2221. PMID: 15572774
40. Emsley P, Cowtan K. Coot: model-building tools for molecular graphics. Acta Crystallogr D Biol Crystal-logr 2004; 60: 2126–2132. PMID: 15572765
41. Brunet YR, Bernard CS, Gavioli M, Lloubes R, Cascales E. An epigenetic switch involving overlappingfur and DNAmethylation optimizes expression of a type VI secretion gene cluster. PLoS Genet 2011;7: e1002205. doi: 10.1371/journal.pgen.1002205 PMID: 21829382
42. Gueguen E, Cascales E. Promoter swapping unveils the role of the Citrobacter rodentium CTS1 typeVI secretion system in interbacterial competition. Appl Environ Microbiol 2013; 79: 32–38. doi: 10.1128/AEM.02504-12 PMID: 23064344
43. Brunet YR, Espinosa L, Harchouni S, Mignot T, Cascales E. Imaging type VI secretion-mediated bacte-rial killing. Cell Rep 2013; 3: 36–41. doi: 10.1016/j.celrep.2012.11.027 PMID: 23291094
Nanobody-Mediated T6SS Inhibition
PLOS ONE | DOI:10.1371/journal.pone.0122187 March 26, 2015 14 / 14
research communications
266 doi:10.1107/S2053230X15000709 Acta Cryst. (2015). F71, 266–271
Received 30 November 2014
Accepted 13 January 2015
Keywords: type VI secretion system; TssM;
nb25.
Production, crystallization and X-ray diffractionanalysis of a complex between a fragment of theTssM T6SS protein and a camelid nanobody
Van Son Nguyen,a,b Silvia Spinelli,a,b Aline Desmyter,a,b Thi Thu Hang Le,a,b
Christine Kellenberger,a,b Eric Cascales,c Christian Cambillaua,b* and Alain
Roussela,b*
aArchitecture et Fonction des Macromolecules Biologiques, CNRS, Campus de Luminy, Case 932, 13288 Marseille,
France, bArchitecture et Fonction des Macromolecules Biologiques, Aix-Marseille Universite, Campus de Luminy,
Case 932, 13288 Marseille, France, and cLaboratoire d’Ingenierie des Systemes Macromoleculaires, Institut de
Microbiologie de la Mediterranee, CNRS and Aix-Marseille Universite, 31 Chemin Joseph Aiguier, 13402 Marseille,
France. *Correspondence e-mail: [email protected], [email protected]
The type VI secretion system (T6SS) is a machine evolved by Gram-negative
bacteria to deliver toxin effectors into target bacterial or eukaryotic cells.
The T6SS is functionally and structurally similar to the contractile tail of the
Myoviridae family of bacteriophages and can be viewed as a syringe anchored to
the bacterial membrane by a transenvelope complex. The membrane complex is
composed of three proteins: the TssM and TssL inner membrane components
and the TssJ outer membrane lipoprotein. The TssM protein is central as it
interacts with both TssL and TssJ, therefore linking the membranes. Using
controlled trypsinolysis, a 32.4 kDa C-terminal fragment of enteroaggregative
Escherichia coli TssM (TssM32Ct) was purified. A nanobody obtained from llama
immunization, nb25, exhibited subnanomolar affinity for TssM32Ct. Crystals of
the TssM32Ct–nb25 complex were obtained and diffracted to 1.9 A resolution.
The crystals belonged to space group P64, with unit-cell parameters a = b = 95.23,
c = 172.95 A. Molecular replacement with a model nanobody indicated the
presence of a dimer of TssM32Ct–nb25 in the asymmetric unit.
1. Introduction
The type VI secretion system (T6SS) is one of the key players
during the intense warfare for nutrients that bacteria
encounter in their ecological niche (Russell et al., 2014). This
large multi-protein complex is widespread in Gram-negative
bacteria and is dedicated to the delivery of enzymatic effectors
directly into bacterial or eukaryotic prey cells (Russell et al.,
2014; Durand et al., 2014). Among the effectors that have
already been identified and characterized, peptidoglycan
hydrolases, DNases and phospholipases are the most common
(Russell et al., 2014; Durand et al., 2014). Although T6SS
phospholipase and DNase effectors might also affect the
integrity of eukaryotic cells, eukaryotic specific effectors have
been described such as the VgrG-borne actin cross-linking
domain in Vibrio cholerae (Pukatzki et al., 2007; Satchell, 2009;
Durand, Derrez et al., 2012).
The T6SS comprises a set of 13 conserved structural
components that participate in the architecture and dynamics
of the secretion system (Cascales & Cambillau, 2012; Coult-
hurst, 2013; Ho et al., 2014; Zoued et al., 2014). Architecturally,
the T6SS can be seen as a syringe-like module tethered to the
cell envelope through contacts with a membrane-associated
complex composed of the TssJ, TssL and TssM proteins
ISSN 2053-230X
# 2015 International Union of Crystallography
electronic reprint
(Cascales & Cambillau, 2012; Zoued et al., 2014). In several
instances such as in enteroaggregative Escherichia coli
(EAEC), an additional component, TagL, mediates anchoring
of this complex to the peptidoglycan layer (Aschtgen, Gavioli
et al., 2010; Aschtgen, Thomas et al., 2010). TagL, TssL and
TssM are all inner membrane proteins, whereas TssJ is an
outer membrane lipoprotein (Aschtgen et al., 2008, 2012;
Ma et al., 2009; Aschtgen, Thomas et al., 2010; Felisberto-
Rodrigues et al., 2011; Durand, Zoued et al., 2012). In EAEC,
TssM (accession No. EC042_4539; GenBank gi:284924260) is a
large, 1129-amino-acid protein bearing a �750-residue peri-
plasmic domain. In addition to making contacts with the TssL
subunit in the inner membrane, the C-terminal region of the
TssM protein interacts with the TssJ outer membrane lipo-
protein (Felisberto-Rodrigues et al., 2011). Three-dimensional
structures of the soluble domains of TagL, TssJ and TssL have
recently been reported (Aschtgen, Gavioli et al., 2010; Felis-
berto-Rodrigues et al., 2011; Rao et al., 2011; Durand, Zoued et
al., 2012; Robb et al., 2012). However, although TssM is of the
utmost importance as it constitutes the scaffold of the T6SS
membrane-associated complex, structural information on this
component is still lacking. To facilitate the crystallization of
TssM, we have raised camelid antibodies against the peri-
plasmic domain of EAEC TssM (amino acids 386–1129;
hereafter called TssMp). Camelid antibodies differ from
conventional antibodies as they harbour unique variable
heavy, antigen-binding domains called VHH domains or
nanobodies (Hamers-Casterman et al., 1993; Muyldermans,
2013). These single-chain antibodies are easy to produce and
have demonstrated their efficiency in improving the crystal-
lization of awkward proteins (Pardon et al., 2014). In parallel,
we have subjected TssMp to controlled enzymatic digestion
to obtain stable structural domains that are more prone to
crystallization. We report here that a combination of these two
approaches is successful: we obtained crystals of a C-terminal
fragment of TssMp in complex with a nanobody at a resolution
sufficient for structure determination
2. Materials and methods
2.1. Production, controlled trypsinolysis and purification ofTssM32Ct
The periplasmic domain of the enteroaggregative E. coli
TssM protein (EAEC_042_4539; gi:284924260), TssMp (resi-
dues 386–1129), was produced and purified as described
previously (Felisberto-Rodrigues et al., 2011). The purified
recombinant TssMp was digested with trypsin [1000:1(m:m)]
at room temperature for 24 h. The reaction was quenched by
the addition of 1 mM phenylmethylsulfonyl fluoride (PMSF)
and insoluble TssMp fragments were discarded by centrifu-
gation at 20 000g for 30 min. A proteolysis-resistant fragment
of apparent size �32 kDa (hereafter called TssM32Ct) was
further purified by consecutive ion-exchange (Mono Q 5/50
GL column; GE Healthcare) and size-exclusion (Superdex 75
16/600 HL column) chromatography using an AKTA system
(GE Healthcare). The purified fragment was subjected to
N-terminal Edman sequencing. A PVDF membrane was
rinsed three times with a water/ethanol mixture (10:90) and
inserted into the A cartridge of a Procise 494A sequencer.
After five cycles of Edman degradation, the sequence DYGSL
was identified, indicating that cleavage after Arg834 generated
a C-terminal fragment of theoretical mass 32 398 Da, in
agreement with the 32 kDa band observed on SDS–PAGE.
2.2. Generation of nanobodies against TssM
Four injections of 1 mg purified TssMp (in 20 mM Tris–HCl
pH 8.0, 150 mM NaCl) were performed subcutaneously at
two-week intervals followed by a fifth injection one month
later in two llamas (Lama glama; Capralogics Inc., Hardwick,
Masachusetts, USA). Lymphocytes were isolated from blood
samples obtained one week after the last immunization. The
cDNA was synthesized from purified total RNA by reverse
transcription and was used as a template for PCR amplifica-
tion to amplify the sequences corresponding to the variable
domains of the heavy-chain antibodies. PCR fragments were
then cloned into the phagemid vector pHEN4 (Pardon et al.,
2014) to create a nanobody phage display library. Selection
and screening of nanobodies were performed as described
previously (Desmyter et al., 2013). Three rounds of panning
resulted in the isolation of TssMp-specific binders. Nano-
bodies nb02 and nb25 were selected, sequenced and cloned
into the pHEN6 expression vector downstream of the pelB
signal peptide and fused to a C-terminal 6�His tag (Conrath et
al., 2009).
2.3. Production and purification of nanobodies
E. coli WK6 cells carrying the pHEN6 derivatives were
grown at 37�C in Terrific Broth medium containing 0.1%
glucose and ampicillin to an optical density (600 nm) of �0.8.
Expression of VHH was induced by the addition of 1 mM
isopropyl �-d-1-thiogalactopyranoside (IPTG) and incubation
for 16 h at 28�C. The periplasmic fraction containing the
nanobodies was prepared by osmotic shock. The His-tail-
containing fusion proteins were purified by immobilized
metal-affinity chromatography on a 5 ml Ni–NTA column
equilibrated in 20 mM Tris–HCl pH 8.0, 300 mM NaCl, 10 mM
imidazole. The fractions eluted in 250 mM imidazole were
concentrated on an Amicon Ultra 10 kDa cutoff concentrator
prior to loading onto a HiLoad 16/60 Superdex 75 gel-
filtration column equilibrated in 20 mM Tris–HCl pH 8.0,
150 mM NaCl buffer.
2.4. Bio-layer interferometry (BLI)
Nanobody nb25 was first biotinylated using the EZ-Link
NHS-PEG4-Biotin kit (Perbio Science, France). The reaction
was stopped by removing excess biotin using a Zeba Spin
Desalting column (Perbio Science, France). BLI studies were
performed in black 96-well plates (Greiner) at 25�C using
an OctetRed96 (ForteBio, USA). Streptavidin biosensor tips
(ForteBio, USA) were hydrated with 0.2 ml kinetic buffer
(KB; ForteBio, USA) for 20 min and then coupled to bioti-
nylated nb25 (10 mg ml�1 in KB). The association of nb25 with
research communications
Acta Cryst. (2015). F71, 266–271 Nguyen et al. � TssM fragment–nanobody complex 267electronic reprint
various concentrations of TssM32Ct (0.13, 0.4, 1.22, 3.66, 11
and 33 nM) was recorded for a period of 600 s followed by
monitoring the kinetics of dissociation in KB for 900 s.
2.5. Crystallization, data collection and molecularreplacement
Crystallization trials were initiated with the TssM32Ct frag-
ment and the nb25 nanobody. The purified His-tagged nano-
body (10 mg ml�1) was mixed with TssM32Ct (1 mg ml�1) in a
1:1 ratio and the TssM32Ct–nb25 complex was purified by gel
filtration on a Superdex 75 16/60 HL column using an AKTA
purifier (GE Healthcare). The complex was concentrated to
11.5 mg ml�1 using a 10 kDa cutoff Centricon in 20 mM Tris–
HCl pH 8.0, 150 mM NaCl and subjected to crystallization
screening with a Mosquito robot (TTP Labtech) using various
commercial kits. Drops were prepared by mixing different
volumes (100, 200 and 300 nl) of the TssM32Ct–nb25 complex
solution and 100 nl precipitant solution, and were equilibrated
against a 150 ml reservoir volume in Greiner plates. Promising
hits were obtained with 20%(m/v) PEG 4000, 0.2 M ammo-
nium sulfate. These crystals were subjected to grid optimiza-
tion using 10–30%(m/v) PEG 4000, 0.1–0.3 M ammonium
sulfate. The best conditions consisted of 18–20% PEG 4000,
0.25 M ammonium sulfate.
TssM32Ct–nb25 crystals were briefly soaked in crystal-
lization solution supplemented with 20%(v/v) glycerol before
being flash-cooled in a nitrogen-gas stream at 100 K. An X-ray
data set was collected from a single crystal of TssM32Ct–nb25
on an in-house rotating-anode generator (Brucker MicroStar)
with a MAR345 image-plate detector. An additional data set
was collected on the PROXIMA1 beamline at SOLEIL, Saint
Aubin, France with a PILATUS 6M detector (Table 1). 1000
images were collected with an oscillation step of 0.20� and 0.4 s
exposure time. The data were integrated, scaled and merged
using the XDS package (Kabsch, 2010). The merged inten-
sities were converted to structure-factor amplitudes using
TRUNCATE from the CCP4 package (Winn et al., 2011).
Molecular replacement was performed using nanobody
structures with high sequence similarity to nb25. As starting
models for MOLREP (Vagin & Teplyakov, 2010), we succes-
sively used three nanobodies with more than 70% sequence
similarity to nb25 (PDB entries 3stb, 4nc2 and 4hem; Park et
al., 2012; Murase et al., 2014; Desmyter et al., 2013).
3. Results and discussion
TssM is a central component of the membrane-anchoring
complex of the type VI secretion system. Topology studies
of TssM showed that this protein is composed of a short
N-terminal cytoplasmic domain followed by three transmem-
brane segments and a large periplasmic domain (Ma et al.,
2009). The C-terminal fragment of this periplasmic domain
contacts the TssJ outer membrane lipoprotein (Felisberto-
Rodrigues et al., 2011) and therefore TssM bridges the inner
and outer membranes. Although the TssM periplasmic
domain (TssMp) readily purified and behaved as a monomer
(Felisberto-Rodrigues et al., 2011), crystallization screening
was unsuccessful. Here, we have subjected TssMp to limited
proteolysis with the goal of selecting stable structural domains
that are more prone to crystallization. Interestingly, SDS–
PAGE analyses showed that a stable fragment with an
apparent molecular mass of �32 kDa resisted digestion with
trypsin [1000:1(m:m)] at room temperature for 24 h (Fig. 1).
This fragment, TssM32Ct, starts at residue Asp835 as deter-
mined by Edman sequencing and corresponds to the
C-terminal region of TssMp. To further facilitate the crystal-
lization process, we selected TssMp-specific camelid anti-
bodies after the immunization of llamas with TssMp. Among
research communications
268 Nguyen et al. � TssM fragment–nanobody complex Acta Cryst. (2015). F71, 266–271
Table 1Data-collection statistics.
Values in parentheses are for the outer shell.
Beamline PROXIMA1, SOLEILWavelength (A) 0.931Detector PILATUS 6MSpace group P64
Unit-cell parameters (A) a = b = 95.2, c = 172.9Resolution range (A) 50.0–1.92 (1.97–1.92)Rmerge 0.079 (1.08)CC1/2 99.9 (84)Total No. of reflections 771552 (55929)No. of unique reflections 67557 (4957)Completeness (%) 100.0 (100.0)Multiplicity 11.4 (11.3)hI/�(I)i 18.0 (2.0)B factor from Wilson plot (A2) 37.1MOLREP contrast with 3stb† 7.1, 6.5 [3.1]MOLREP contrast with 4nc2† 6.9, 6.2 [3.2]MOLREP contrast with 4hem† 6.5, 5.9 [3.0]
† The contrast of the next solution is given in square brackets.
Figure 1Isolation of the TssM32Ct fragment by controlled trypsinolysis. Fragmentsof purified TssMp (lane 1) subjected to trypsin digestion at roomtemperature for 2, 5, 10, 20, 40, 80 min and 20 h (lanes 2–8, respectively)were analysed by SDS–PAGE with Coomassie Blue staining. Themolecular-weight markers are shown in lane 9 and are labelled in kDa.The full-length TssM periplasmic domain (TssMp) and the trypsinolysedfragment TssM32Ct are indicated.
electronic reprint
the four anti-TssMp nanobodies, nanobody nb25 binds to
TssM32Ct as shown by size-exclusion chromatography (Fig. 2).
The binding constant was determined by bio-layer inter-
ferometry (BLI) using an OctetRed96 apparatus (ForteBio).
Analysis of BLI curves indicated that nb25 binds to TssM32Ct
with a Kd value of 0.88 � 0.005 nM (Fig. 3).
research communications
Acta Cryst. (2015). F71, 266–271 Nguyen et al. � TssM fragment–nanobody complex 269
Figure 2TssM32Ct–nb25 complex formation. (a) Amino-acid sequences of TssM32Ct and nb25. The cleavage site in TssM is indicated by a triangle. The CDRs ofnb25 are coloured blue, green and red for CDRs 1, 2 and 3, respectively. The nb25 cysteines are coloured orange. (b) Gel-filtration analysis of TssM32Ct
(red) and the TssM32Ct–nb25 complex (blue). The volume of elution is indicated above each peak. (c) Selected fractions were analyzed by SDS–PAGE.Molecular-weight markers are shown in lane 1 (labelled in kDa). Lanes 2 and 3 correspond to the insoluble fraction (pellet) and the soluble fraction(supernatant) obtained after the digestion of TssMp with trypsin, respectively, lanes 7–10 correspond to fractions at 50–63 ml and lane 13 corresponds tothe fractions centred at 81.6 ml. The peak at 56.66 ml therefore corresponds to the TssM32Ct–nb25 complex, while the peak at 81.63 ml corresponds toexcess nb25.
Figure 3Bio-layer interferometry analysis of the TssM32Ct–nb25 interaction. The chip was covered with biotinylated nb25 nanobody and the responses afterinjection of various concentrations of TssM32Ct (0.13, 0.4, 1.22, 3.66, 11, 33, 100 and 300 nM) were recorded (blue curves). The response (in nanometres)is plotted versus the time (in seconds). The kon and koff analysis (red curves) excluded the outlier values obtained at 100 and 300 nM. The values of kon
and koff are 3.26 � 105 and 2.4 � 10�4, respectively, and Kd is 0.88 � 0.005 nM.
electronic reprint
Crystals of the TssM32Ct–nb25 complex grew after 20 d from
a protein solution concentrated to 11.5 mg ml�1 in 20 mM
Tris–HCl pH 8.0, 150 mM NaCl with precipitation conditions
consisting of 20% PEG 4000, 0.25 M ammonium sulfate. The
dimensions of the crystals were 0.04 � 0.04 � 0.06 mm (Fig. 4).
The crystals belonged to space group P64, with unit-cell
parameters a = b = 95.2, c = 172.9 A. With a complex
molecular weight of 47.1 kDa, a Matthews coefficient of
2.41 A3 Da�1 (49% solvent) corresponds to the presence of
two molecules in the asymmetric unit.
A complete data set was collected using a PILATUS 6M
detector on the PROXIMA1 beamline at SOLEIL, Saint-
Aubin, France, yielding a resolution of 1.9 A and excellent
statistics (Table 1). Owing to the uncertainty in the asymmetric
unit content, we performed molecular-replacement assays
with MOLREP (Vagin & Teplyakov, 2010) using three
different nanobodies sharing high sequence homology with
nb25 (PDB entries 3stb, 4nc2 and 4hem). All three molecular-
replacement attempts returned a common outcome with two
nanobodies in the asymmetric unit (Table 1). The crystal
packing (Fig. 5) is compatible with the presence of two
TssM32Ct molecules bound to the two nanobodies. Further
phasing experiments with SeMet labelling are under way to
solve the structure of the complex. This information will be of
great interest in order to better understand the T6SS archi-
tecture, as TssM32Ct corresponds to a conserved domain that
mediates interaction with the TssJ outer membrane lipo-
protein.
Acknowledgements
We thank the SOLEIL synchrotron for data-collection time
and the PROXIMA1 beamline staff for assistance. This work
was supported by the French Infrastructure for Integrated
Structural Biology (FRISBI) ANR-10-INSB-05-01.
References
Aschtgen, M.-S., Bernard, C. S., De Bentzmann, S., Lloubes, R. &Cascales, E. (2008). J. Bacteriol. 190, 7523–7531.
Aschtgen, M.-S., Gavioli, M., Dessen, A., Lloubes, R. & Cascales, E.(2010). Mol. Microbiol. 75, 886–889.
Aschtgen, M.-S., Thomas, M. S. & Cascales, E. (2010). Virulence, 1,535–540.
Aschtgen, M.-S., Zoued, A., Lloubes, R., Journet, L. & Cascales, E.(2012). MicrobiologyOpen, 1, 71–82.
Cascales, E. & Cambillau, C. (2012). Philos. Trans. R. Soc. Lond. BBiol. Sci. 367, 1102–1111.
Conrath, K., Pereira, A. S., Martins, C. E., Timoteo, C. G., Tavares, P.,Spinelli, S., Kinne, J., Flaudrops, C., Cambillau, C., Muyldermans,S., Moura, I., Moura, J. J. G., Tegoni, M. & Desmyter, A. (2009).Protein Sci. 18, 619–628.
Coulthurst, S. J. (2013). Res. Microbiol. 164, 640–654.Desmyter, A., Farenc, C., Mahony, J., Spinelli, S., Bebeacua, C.,
Blangy, S., Veesler, D., van Sinderen, D. & Cambillau, C. (2013).Proc. Natl Acad. Sci. USA, 110, E1371–E1379.
Durand, E., Cambillau, C., Cascales, E. & Journet, L. (2014). TrendsMicrobiol. 22, 498–507.
Durand, E., Derrez, E., Audoly, G., Spinelli, S., Ortiz-Lombardia, M.,Raoult, D., Cascales, E. & Cambillau, C. (2012). J. Biol. Chem. 287,38190–38199.
research communications
270 Nguyen et al. � TssM fragment–nanobody complex Acta Cryst. (2015). F71, 266–271
Figure 5Crystal packing of nb25 in the TssM32Ct–nb25 complex after molecularreplacement. Two orientations rotated by 90� from each other arepresented. The two nanobodies in the asymmetric unit are identified bywhite arrows and are rainbow-coloured (views were generated usingCoot; Emsley & Cowtan, 2004).
Figure 4Representative crystal of the TssM32Ct–nb25 complex.
electronic reprint
Durand, E., Zoued, A., Spinelli, S., Watson, P. J., Aschtgen, M. S.,Journet, L., Cambillau, C. & Cascales, E. (2012). J. Biol. Chem. 287,14157–14168.
Emsley, P. & Cowtan, K. (2004). Acta Cryst. D60, 2126–2132.Felisberto-Rodrigues, C., Durand, E., Aschtgen, M. S., Blangy, S.,
Ortiz-Lombardia, M., Douzi, B., Cambillau, C. & Cascales, E.(2011). PLoS Pathog. 7, e1002386.
Hamers-Casterman, C., Atarhouch, T., Muyldermans, S., Robinson,G., Hammers, C., Songa, E. B., Bendahman, N. & Hammers, R.(1993). Nature (London), 363, 446–448.
Ho, B. T., Dong, T. G. & Mekalanos, J. J. (2014). Cell Host Microbe, 15,9–21.
Kabsch, W. (2010). Acta Cryst. D66, 125–132.Ma, L.-S., Lin, J.-S. & Lai, E.-M. (2009). J. Bacteriol. 191, 4316–
4329.Murase, T., Eugenio, L., Schorr, M., Hussack, G., Tanha, J., Kitova,
E. N., Klassen, J. S. & Ng, K. K. S. (2014). J. Biol. Chem. 289, 2331–2343.
Muyldermans, S. (2013). Annu. Rev. Biochem. 82, 775–797.
Pardon, E., Laeremans, T., Triest, S., Rasmussen, S. G., Wohlkonig,A., Ruf, A., Muyldermans, S., Hol, W. G. J., Kobilka, B. K. &Steyaert, J. (2014). Nature Protoc. 9, 674–693.
Park, Y.-J., Pardon, E., Wu, M., Steyaert, J. & Hol, W. G. J. (2012).Nucleic Acids Res. 40, 1828–1840.
Pukatzki, S., Ma, A. T., Revel, A. T., Sturtevant, D. & Mekalanos, J. J.(2007). Proc. Natl Acad. Sci. USA, 104, 15508–15513.
Rao, V. A., Shepherd, S. M., English, G., Coulthurst, S. J. & Hunter,W. N. (2011). Acta Cryst. D67, 1065–1072.
Robb, C. S., Nano, F. E. & Boraston, A. B. (2012). J. Mol. Biol. 419,277–283.
Russell, A. B., Peterson, S. B. & Mougous, J. D. (2014). Nature Rev.Microbiol. 12, 137–148.
Satchell, K. J. (2009). Toxins, 1, 123–133.Vagin, A. & Teplyakov, A. (2010). Acta Cryst. D66, 22–25.Winn, M. D. et al. (2011). Acta Cryst. D67, 235–242.Zoued, A., Brunet, Y. R., Durand, E., Aschtgen, M.-S., Logger, L.,
Douzi, B., Journet, L., Cambillau, C. & Cascales, E. (2014).Biochim. Biophys. Acta, 1843, 1664–1673.
research communications
Acta Cryst. (2015). F71, 266–271 Nguyen et al. � TssM fragment–nanobody complex 271electronic reprint
RÉSUMÉ Le système de sécrétion de type VI (T6SS) est une machine multi-protéines qui injecte des effecteurs protéiques toxiques pour tuer les cellules proies procaryotes ou eucaryotes, à des fins de compétition interbactérienne et de virulence. Le mécanisme d'action du T6SS nécessite la contraction d'une structure en forme de gaine qui propulse un tube interne coiffé par une pointe vers les cellules cibles, permettant l’injection des effecteurs protéiques. En 2006 le système de sécrétion de type VI (T6SS) a été décrit , pour la première fois, dans deux genres bactériens, Vibrio cholerae et Pseudomonas aeruginosa. Plus tard, ce système a été trouvé dans diverses bactéries, dont Burkholderia mallei, Burkholderia cenocepacia, Edwardsiella tarda, Serratia marcescens, Escherichia coli, Agrobacterium tumefaciens, Aeromonas hydrophila, Helicobacter, Campylobacter ainsi que d'autres organismes. L'analyse génomique suggère que les T6SS se retrouvent dans au-moins 25% de toutes les bactéries Gram-négatives séquencées, faisant du T6SS le système de sécrétion spécialisé le plus répandu. Mon travail a porté sur la caractérisation des effecteurs toxiques et protéines d’immunité du T6SS Sci-1 d’Escherichia coli Entero-agrégatif, éléments de la lutte inter-bactérienne. Nous avons identifié en outre Tle1, un effecteur de toxine codé par ce groupe et montré que Tle1 possède des activités de phospholipase A1 et A2 requises pour détruire la cellule proie dans la compétition interbactérienne. L'auto-protection de la cellule attaquante est assurée par une lipoprotéine de membrane externe, Tli1, qui lie Tle1 dans un rapport stoechiométrique 1: 1 avec une affinité nanomolaire et inhibe son activité phospholipase. Tle1 est délivré dans le périplasme des cellules proie en utilisant la protéine VgrG1 comme support. D'autres analyses démontrent que le domaine d'extension C-terminal de VgrG1, incluant un domaine de type transthyrétine, est responsable de l'interaction avec Tle1 et de sa distribution ultérieure dans des cellules cibles. Sur la base de ces résultats, il a été proposé un mécanisme supplémentaire de transport des effecteurs T6SS dans lequel des effecteurs apparentés sont sélectionnés par des motifs spécifiques situés dans les régions C-terminales de protéines VgrG. Un groupe pathogène de E. coli, appelé E. coli invasive/envahissantes (AIEC), a été largement impliqué dans la maladie de Crohn humaine (CD) et est actuellement l'un des éléments des plus importants dans l'histoire de ce pathogène. Il existe au moins deux groupes de gènes dans le génome AIEC qui codent pour les composants du T6SS appelés AIEC LF82 T6SS1 et AIEC LF82 T6SS2. Il a été prédit que la protéine 435 provenant à partir d'un groupe de gènes T6SS1 de l'agent pathogène AIEC LF82 est une phospholipase de la famille d'effecteurs Tle3 avec une activité PLA1. Sa toxicité peut être neutralisée par la protéine d'immunité cognate 434 qui est un Tli3 putatif, en formant le complexe de protéine Tle3 - Tli3. Les deux protéines séparées et leur complexe ont ensuite été appelées protéines complexes Tle3AIEC, Tli3AIEC et Tle3AIEC - Tli3AIEC, respectivement. Afin d'étudier plus en détail le mécanisme de Tle3-AIEC et de Tli3-AIEC, nous avons réalisé l'expression, la purification, la caractérisation, la cristallisation des deux protéines et des études cristallographiques de rayons X préliminaires du complexe Tle3-AIEC/Tli3-AIEC afin de comprendre comment la protéine Tle3-AIEC reconnaît et se lie à son effecteur apparenté Tli3-AIEC et inhibe son activité. Les données préliminaires de diffraction des rayons X ont été recueillies à partir de cristaux Tle3AIEC-SeMet/Tli3AIEC à une résolution de 3,8 Å.
