Infectious diseases are major threats to human health worldwide, and
tremendous effort has gone into understanding various infectious agents
and their mechanisms of virulence. One theme that emerged early from
studies of bacterial pathogens is that many inject pathogenic factors (also
known as effectors or virulence factors) directly into host cells as part
of their pathogenic strategy, and these factors specifically target crucial
intracellular pathways in the host. This pathogenic strategy was shown
to be used by certain extracellular and intracellular pathogens. These
findings led to an exciting and dynamic crossover between related dis-
ciplines, including microbiology, cell biology, biochemistry and immu-
nology. And from such interdisciplinary studies arose the central tenet
that bacterial pathogens specifically attack key intracellular-signalling
and cytoskeletal pathways to alter host responses in a way that favours
the pathogen. Another important finding of these studies was that the
same pathways are often targeted by different bacterial effectors and that
these pathways can be attacked at several points by a single pathogen,
ensuring an override of important cellular functions.
In this review, we provide an overview of the current understanding of
how bacterial pathogens interact with host cells. How these microorgan-
isms exploit host cells is discussed in terms of both promoting the bacterial
life cycle and evading the host immune response. We focus on examples of
pathogenic bacteria that interact with mammalian intracellular-signalling,
vesicular-trafficking and cytoskeletal pathways, because these are some of
the best studied or most rapidly advancing areas.
Cell biology of bacterial infection
Bacterial pathogens use a range of effectors to subvert and control nor-
mal cellular functions. Effectors are usually specialized proteins that are
injected directly into the cytosol of the host cell by a type III secretion
system (T3SS) or a type IV secretion system (T4SS). These secretion sys-
tems consist of a structurally conserved proteinaceous apparatus that is
shaped like a needle1.
The concept of secreted proteins functioning as agents of microbial
virulence is not new. Toxins have been recognized in this capacity for
decades. But the ability to inject effectors directly into mammalian or
plant cells is repeatedly encountered when considering the cell biology of
the infectious process. Therefore, it is an important process in microbial
pathogenesis as it is understood at present, and an important interface
between pathogens and their hosts.
To promote their life cycle, bacterial pathogens can use host cells to aid
their own adherence, replication and/or dissemination. The initial step
in colonization by bacteria is adherence. Bacterial pathogens have a large
variety of cell-surface adhesins, including fimbriae and afimbrial adhesins
(see ref. 2 for a review), that enable them to attach to host cells. Some of
these adhesins also have a further role: they bind to their cognate receptors
on non-phagocytic cells, thereby allowing bacteria to be taken up by these
cells. Such adhesins with dual roles include the invasins of Yersinia spp.
and the internalins of Listeria spp.3. The internalization mechanisms of
intracellular pathogens differ on the basis of the effector involved, and
they require that complementary host intracellular-signalling pathways
The most commonly described cellular target of pathogens is the
cytoskeleton. Various intracellular microorganisms harness cytoskeletal
components to gain entry to, and to propel themselves within, host cells
(see ref. 4 for a review) (Fig. 1). The cytoskeleton of eukaryotic cells is
composed of actin filaments, microtubules and intermediate filaments.
In terms of bacterial pathogenesis, the most extensively studied of these
are actin filaments. Bacterial pathogens do not usually interact directly
with actin filaments themselves. Instead, they subvert and control the
polymerization of actin filaments by modulating cellular regulators of
this process, such as small Rho-like G proteins5, through the action
of delivered effectors (Fig. 1a).
Some bacterial pathogens remain in vacuoles after internalization,
and these microorganisms often use effectors to modulate vesicular traf-
ficking, providing a protective niche within host cells (Fig. 1c) (inclu-
ding in macrophages and neutrophils, which normally kill bacteria)6.
In addition, pathogens can interact with cell-death pathways (including
apoptosis), modulating host-cell death to facilitate pathogen survival
in the host7. Moreover, one of the key ways that pathogens evade or
subvert the host immune response (both the innate and the adaptive
immune mechanisms)8 is to secrete effector proteins (Fig. 1d). These
steps of the colonization and immune-evasion processes are discussed
in more detail later.
Manipulation of the cytoskeleton and membranous structures
How pathogenic bacteria exploit the host-cell cytoskeleton, membra-
nous structures and key signalling pathways to their advantage is dis-
cussed in this section, together with the strategies and rationales these
1The University of British Columbia, Michael Smith Laboratories, 301-2185 East Mall, Vancouver, British Columbia V6T 1Z4, Canada. 2Present address: Simon Fraser University, 8888 University
Drive, Department of Biological Sciences, Room B8276, Shrum Science Centre, Burnaby, British Columbia V5A 1S6, Canada.
Manipulation of host-cell pathways by
Amit P. Bhavsar1, Julian A. Guttman1,2 & B. Brett Finlay1
Bacterial pathogens operate by attacking crucial intracellular pathways in their hosts. These pathogens
usually target more than one intracellular pathway and often interact at several points in each of these
pathways to commandeer them fully. Although different bacterial pathogens tend to exploit similar pathway
components in the host, the way in which they ‘hijack’ host cells usually differs. Knowledge of how pathogens
target distinct cytoskeletal components and immune-cell signalling pathways is rapidly advancing, together
with the understanding of bacterial virulence at a molecular level. Studying how these bacterial pathogens
subvert host-cell pathways is central to understanding infectious disease.
NATURE|Vol 449|18 October 2007|doi:10.1038/nature06247
microorganisms use to invade, survive intracellularly and replicate in
Interactions of bacteria with the actin cytoskeleton
Bacterial pathogens manipulate the cytoskeleton to help invade a host
cell and/or to gain motility in the cell, as mentioned earlier. They often
interact with actin filaments in particular, and they do so by modulating
G proteins. This process is exemplified by the interaction of the invasive
bacterium Salmonella enterica with mammalian cells. During this pro-
cess, S. enterica delivers the T3SS effector proteins SopE and SopE2 into
the host cell. These effectors function as guanine-nucleotide-exchange
factors for G proteins, activating the G protein CDC42 and the RAC
family of G proteins in the target cell9–11. This G-protein activation, in
turn, induces the generation of actin-rich membrane ruffles that engulf
and internalize the bacteria (Fig. 1a). An interesting alternative strategy
has recently been reported12: bacterial effector proteins that contain a
Trp-X-X-X-Glu motif suppress the signalling of active G proteins and
mimic these active G proteins themselves, thereby obviating the need
for modulating the GTPase activity of G proteins (Box 1).
After invasion and escape from membrane-enclosed vesicles into the
cytosol, many pathogens also manipulate actin-filament dynamics so
that they can move within the infected host cell (see ref. 4 for a review).
They do so by recruiting actin to just one of their poles, through bac t-
erial-protein-mediated nucleation of actin. For example, the intracellular
motility of Shigella flexneri is mediated by the bacterial effector IcsA. IcsA
interacts directly with the host protein N-WASP (neural Wiskott–Aldrich
syndrome protein; also known as WASL), which in turn recruits a com-
plex known as the Arp2/3 complex (consisting of seven host proteins,
inclu ding actin-related protein 2 (ARP2) and ARP3).This complex polym-
erizes actin filaments behind the advancing bacterium13. By contrast, the
cytosolic motility of Listeria spp. is mediated by the bacterial protein ActA,
which binds directly to both the Arp2/3 complex and the actin-associated
protein VASP (vasodilator-stimulated phosphoprotein)14,15.
The hijacking of actin-associated cytoskeletal components can also
occur during infection with extracellular pathogens. For example, the
attaching and effacing human pathogens enterohaemorrhagic Escherichia
coli (EHEC) and enteropathogenic E. coli (EPEC) have an elaborate actin-
recruiting process (Fig. 2). In this case, the bacterial effector protein Tir
mediates extensive modification of host-cell actin filaments beneath the
adherent microorganism. Tir is delivered into the target cell by the T3SS
and embeds itself in the plasma membrane, where it anchors the bac terium
firmly by binding to the bacterial outer-membrane protein intimin. In the
case of EPEC infection, Tir is tyrosine phosphorylated on the cytoplasmic
face of the host-cell plasma membrane and recruits the host adaptor pro-
tein Nck (non-catalytic region of tyrosine kinase)16. During EHEC infec-
tion, Nck is not recruited; instead, an additional bacterial effector, EspFu
(also known as TccP), is involved17,18. Downstream of this protein (Nck or
EspFu), N-WASP and the Arp2/3 complex are recruited, and they mediate
the polymerization of actin filaments beneath the adherent extracellular
bact erium16–18. This results in the formation of a ‘pedestal’ on the host-cell
EPEC or EHEC
MHC class II
Figure 1 | The cell biology of bacterial infections. Both extracellular
bacterial pathogens and intracellular bacterial pathogens initially
interact with the plasma membrane of host cells. These pathogens
commandeer various common structures and pathways. a, Extracellular
pathogens (for example, EPEC and EHEC) routinely manipulate the
actin cytoskeleton to generate actin-rich pedestal structures. By contrast,
the invasive intracellular pathogens Salmonella enterica and Shigella
spp. use this cytoskeletal component to invade the host cell. Another
invasive intracellular pathogen, Listeria monocytogenes, uses clathrin-
mediated endocytosis for invasion. b, During some bacterial infections,
microtubules, another component of the cytoskeleton, are disassembled
through the actions of particular effector proteins. Such proteins present
during EPEC infection (EspG) and Shigella flexneri infection (VirA)
disassemble microtubules in the vicinity of bacterial contact. c, After
invasive bacteria have entered the host cell, they occupy a vacuole. The
vacuole can offer protection against immune detection and can be a
replicative niche (in the case of S. enterica). Alternatively, the bacteria
can escape the vacuole and gain the ability to propel themselves through
the cytosol (in the case of L. monocytogenes and S. flexneri). Organellar
components can also be acquired: for example, Legionella pneumophila,
which is usually internalized by phagocytosis, interacts with endoplasmic-
reticulum-derived vesicles. d, Bacterial pathogens also have other
effector-protein-based mechanisms for evading both the innate immune
response and the adaptive immune response. Examples include subverting
the presentation of bacterial antigen at the cell surface (adaptive) and
interfering with the translocation of the pro-inflammatory transcriptional
activator NF-κB to the cell nucleus (innate). e, The nucleus is also the
destination of several T3SS effectors (for example, S. enterica SspH1,
Yersinia spp. YopM and S. flexneri OspF and IpaH9.8), and the effects of
this are beginning to come to light.
NATURE|Vol 449|18 October 2007
surface, where the pathogen resides. These pedestals can move (‘surf’)
on the cell surface19: the actin-disassembly proteins cofilin and gelsolin
have been identified in pedestals, and these proteins pres umably regu-
late actin-filament dynamics in conjunction with the actin-assembly
protein profilin. Numerous other actin-associated proteins — inclu-
ding cortactin20, GRB2 (growth-factor-receptor-bound protein 2), LPP
(lipoma-preferred partner), SHC (SRC-homology-2-domain-containing
transforming protein C), vinculin and zyxin21 — have been found in
pedestals, although their precise organization during pedestal genera-
tion has not yet been determined. The protein α-actinin has also been
identified in EPEC-induced pedestals, where it specifically interacts with
the amino terminus of Tir22. Surprisingly, the endocytosis-associated pro-
tein dynamin has been found in EPEC-induced pedestals23, as have the
intermediate-filament proteins cytokeratin 8 and cytokeratin 18 (ref. 24)
and the tight-junction component ZO1 (also known as TJP1)25. Therefore,
pedestals are useful sites at which to study the interplay of cytoskeletal
systems with components of signalling pathways, endocytosis pathways
and intercellular junctions.
Interactions of microtubules with effectors
Microtubules are also commonly targeted by microorganisms. These
polarized structures are normally used for structural support and as tracks
to guide and transport intracellular cargo, with the aid of microtubule-
associated molecular-motor proteins. During certain infections, both
the cargo transport and the microtubule assembly and/or disassembly
dynamics can be modified and controlled by the pathogen. For exam-
ple, on invasion by Shigella spp., the VirA protein interacts directly with
heterodimers of α-tubulin and β-tubulin, promoting destabilization
of the microtubules26 (Fig. 1b). This results in a localized absence of
microtubules near the invading bacteria, thus aiding invasion by Shigella
spp. A similar phenotype is seen during EPEC infections (Fig. 1b). In
this case, localized microtubule depolymerization depends on the bac-
terial effector EspG, which, similarly to VirA, interacts directly with
Whereas Shigella spp. and pathogenic E. coli disassemble microtubules,
a strain of Campylobacter jejuni has been shown to use microtubules and
their associated molecular motors to aid invasion. Microtubules are polar
structures that have distinct fast-growing (plus) and slow-growing (minus)
ends (Fig. 1b), allowing the directional transport of cargo in cells. There
are two general types of microtubule-based molecular motor found in
the host-cell cytosol: kinesins and cytoplasmic dynein. Members of the
kinesin superfamily generally transport cargo towards the plus ends of
microtubules. By contrast, cytoplasmic dynein is thought to be a minus-
end-directed motor. Microtubules, and particularly dynein, have been
implicated in invasion by a strain of C. jejuni28. Given that the minus ends
of microtubules in cultured, non-polarized cells are directed towards
the interior of the cell (Fig. 1b), this model seems plausible. However,
in polarized cells (such as those present in intestinal epithelial barriers),
microtubule polarity is reversed, so dynein would not transport C. jejuni
towards the cell interior. Therefore, further investigation into the uptake
of C. jejuni by host cells is required.
Life in a host cell
After entry to host cells, invasive pathogens are either localized in the
cytosol or sequestered in vesicular structures. Presumably, all intracellular
pathogens occupy a membrane-enclosed compartment at some point
of their intracellular phase, even if only transiently. The initial compart-
ments after internalization (vacuoles and modified phagosomes) are com-
posed of membranous host-cell components; therefore, internalization
often generates protected areas. Pathogens have adopted various strategies
to multiply in, or escape from, these structures before surviving in the
cytosol and then being disseminated throughout the host.
The ability to occupy a protected intracellular niche contributes to
the pathogenesis of both S. enterica and Legionella spp. (see refs 29 and 30
for reviews). On passive internalization by phagocytic cells, Legionella spp.
occupy a compartment known as a Legionella-containing vacuole (LCV).
This phagosome is modified by the Dot/Icm secretion system (a T4SS).
T4SS effectors enable LCVs both to evade common phagocytic-degrada-
tion pathways (by preventing the acidification of vacuoles and the associa-
tion of proteins found in late endosomes and lysosomes with LCVs) and to
acquire components commonly found in secretory pathways. One of these
acquired components is the GTPase ADP-ribosylation factor 1 (ARF1), the
function of which is mediated by the bacterial effector RalF31. Legionella
spp. further modify the phagosome by using the T4SS effector SidJ to
recruit small endoplasmic-reticulum-derived vesicles to the phagosomal
membrane, and then mediate fusion with these vesicles32, potentially pro-
viding a nutrient-rich resource for the bacteria (Fig. 1c). In addition, the
GTPase RAB1 is found at the LCV membrane and has a role in the fusion
of endoplasmic-reticulum-derived vesicles with the LCV. The function of
RAB1 is controlled by the bacterial effector DrrA (also known as SidM),
One of the central themes of bacterial pathogenesis is the manipulation
of host-cell cytoskeletal components by injected effector proteins,
which mediate this effect by subverting host G-protein signalling
through their guanine-nucleotide-exchange and GTPase-activating
activities. However, new data show that bacterial effector proteins
can also catalyse novel, diverse and ingenious biochemical reactions
that contribute to pathogenesis. For example, several effector proteins
contain a Trp-X-X-X-Glu amino-acid motif, which enables them to
‘mimic’ the function of active (GTP-bound) G proteins, allowing
downstream signalling and cytoskeletal remodelling12. This novel
biochemical activity, the mechanism of which remains unknown,
bypasses the requirement for G proteins.
Insights into the biochemistry of effector proteins that target host-cell
ubiquitylation pathways are also emerging. The Shigella flexneri effector
IpaH9.8 and the Salmonella enterica effector SspH1 were recently
shown to have E3 ubiquitin–protein ligase activity66. IpaH9.8 catalyses
the transfer of ubiquitin to the yeast mitogen-activated protein kinase
(MAPK)-signalling-cascade member Ste7, presumably inducing its
degradation and abrogating MAPK signalling. SspH1 catalyses the
transfer of ubiquitin to the mammalian protein kinase PKN1, although
the effect on PKN1-mediated signalling is unclear. By contrast, the
S. enterica effector SseL was recently shown to catalyse the removal
of polyubiquitin chains that had been attached to host proteins during
infection67. It is intriguing to consider that the opposing biochemical
activities of SspH1 and SseL might reflect the necessity of coordinating
effector functions (discussed later).
MAPK-signalling pathways have an important role in immunity
of the host to bacterial pathogens. Consequently, these canonical
phosphorylation cascades are subject to attack by multiple bacterial
effector proteins. Recently, two novel and diverse biochemical activities
were identified for effectors targeting this pathway. Proteomic analyses
indicate that the Yersinia spp. effector YopP/J is an acetylase68. The
authors of this study propose that the transfer of acetyl moieties
to key residues on MAPK substrates competes effectively with
phosphorylation at these sites, thereby blocking signal transduction
(Fig. 3). By contrast, the S. flexneri effector OspF has been shown to
irreversibly dephosphorylate specific MAPKs — extracellular-signal-
regulated kinase 2 (ERK2; also known as MAPK1), p38 MAPK (also
known as MAPK14) and Jun amino-terminal kinase (JNK; also known as
MAPK8) — by an elimination reaction that chemically modifies the key
threonine residue of the substrate so that this MAPK cannot function
in the signalling pathway69 (Fig. 3). This enzymatic activity is called
phosphothreonine-lyase activity, and it also seems to be present in
other pathogenic bacteria. Interestingly, the authors of this study found
that OspF had no phosphotyrosine-phosphatase activity, in contrast to
another report that claimed OspF was a dual-specificity phosphatase48.
The dephosphorylation of phosphotyrosine by an elimination reaction
is a highly improbable mechanism, so to determine whether OspF is a
dual-specificity phosphatase requires further study. Nevertheless, the
identification of irreversible phosphothreonine-lyase activity opens
the door to debate about the benefits for the infecting bacteria of
irreversible modification of host-cell biology compared with those of
Box 1 | Novel biochemical activities of translocated bacterial effectors
NATURE|Vol 449|18 October 2007
which has guanine-nucleotide-exchange-factor activity for RAB1 during
infection with Legionella spp.33. As the infection progresses, these vesicles
disappear (as assessed by morphological characteristics), and ribosomes
are found to interact with the LCV membrane, thus placing the bacteria
within a rough-endoplasmic-reticulum-like vacuole.
Similarly, S. enterica modifies its phagosome-like vacuole, by using
a set of T3SS effectors, to provide a protective niche where the bac t-
eria survive and replicate30 (Fig. 1c). It accomplishes this by selectively
interacting with components of the endocytic machinery of the host
cell, thereby acquiring molecules such as early endosome antigen 1
and lysosomal-associated membrane protein 1 (refs 34, 35). Although
it has long been accepted that lysosomes are inhibited from fusion with
Salmonella-containing vacuoles (SCVs), recent advances have shown
that lysosomes can readily fuse with SCVs during S. enterica infection36,
raising numerous questions about the exact mechanisms used to evade
destruction by the host.
Active invasion by S. flexneri produces a vacuole around the invading
microorganism. However, during invasion of epithelial cells, S. flexneri
occupies this vacuole only briefly. This escape from the vacuole allows
the bacterium to replicate in the host-cell cytosol and, eventually, to
spread from cell to cell37. Listeria monocytogenes, a pathogen that is inter-
nalized through clathrin-mediated endocytosis38, is also initially found
in a vacuole (Fig. 1c). In a similar manner to invasion by S. flexneri, these
vacuoles are short lived. L. monocytogenes uses the membrane-pore-
forming toxin listeriolysin O, as well as the enzymes PlcA and PlcB, to
destroy the surrounding membrane, thereby allowing escape from the
vacuole and, subsequently, replication within the host-cell cytosol and
actin-mediated spreading from cell to cell39.
Interactions of bacterial pathogens with signalling pathways
A beneficial strategy used by many pathogens is to interfere with the
phosphorylation cascades in the intracellular-signalling pathways of
the host cell. Phosphorylation states are usually controlled by protein
kinases and protein phosphatases, and the functions of these enzymes
are mimicked by certain bacterial effector proteins. Evidence for this
comes from the study of T3SS effectors from S. enterica and Yersinia
spp. The S. enterica effector SigD (also known as SopB) functions as a
phosphoinositide phosphatase that catalyses the dephosphorylation
of host phosphatidylinositol-4,5-bisphosphate and phosphatidyl-
inositol-3,4,5-trisphosphate40. As a result, membrane-fission dynam-
ics are altered and probably affect SCV formation. The effector YpkA
(and its homologue YopO), produced by Yersinia spp., has structural
and functional similarities to serine/threonine kinases. This effector is
secreted by the bacteria in an inactive form and is autophosphorylated,
and thereby activated, in the host cell, where it modulates the actin
cytoskeleton through a direct interaction with the GTPase RAC1
(ref. 41). An interesting variation of manipulating host intracellular-
signalling pathways involves another effector produced by Yersinia spp.,
YopM, which simultaneously binds to (and thereby activates) two host
protein kinases42. However, the functional significance of the formation
of this complex remains poorly understood (Box 2).
It is therefore clear that bacterial pathogens use diverse mechanisms
to accomplish a similar goal — to interact with and, potentially, alter
the host cell. However, bacterial pathogens must also ensure their own
preservation in the host: they need to evade the immune response, and
this facilitates replication and spread, which are essential for the suc-
cess of any pathogen. The following examples highlight the incredibly
diverse mechanisms that bacterial pathogens use to evade both innate
immune responses and adaptive immune responses.
Inflammation and nuclear factor-κB
A cornerstone of innate immunity is the expression of genes that are
responsive to the transcription factor nuclear factor-κB (NF-κB)43
(Fig. 3). This process is induced after bacterial pathogen-associated
molecular patterns (PAMPs) are detected by pattern-recognition
receptors (PRRs), including Toll-like receptors and NOD (nucleotide-
binding oligomerization-domain protein)-like receptors (NLRs) (see
ref. 44 for a review and see page 819). NF-κB-responsive genes include
those that encode pro-inflammatory cytokines, anti-apoptotic factors
(such as Bcl-2) and defensins (a class of antimicrobial peptide). Before
these genes can be transcribed, NF-κB needs to be activated, and this
occurs when its cytoplasmic binding partner, inhibitor of NF-κB (IκB),
is degraded, enabling NF-κB to translocate to the nucleus. The degrada-
tion of IκB occurs after it is phosphorylated by the protein IκB kinase
EPEC or EHEC
Figure 2 | Generation of pedestals by EPEC and EHEC. During infection with
the extracellular bacterium EPEC, the intimin receptor (Tir) translocates
into the host cell and inserts itself into the host-cell plasma membrane
(a process mediated by the T3SS). This receptor interacts with intimin on
the bacterial surface, thereby firmly anchoring the bacterium to the host
cell. The carboxy terminus of EPEC Tir becomes phosphorylated on the
tyrosine residue at position 474 by at least two host protein kinases, Fyn
and Abl, resulting in host adaptor protein Nck being recruited and binding
directly to Tir. During infection with EHEC, by contrast, the tyrosine-
phosphorylation event is subverted by the EHEC effector EspFu, so Nck is
not required. During EPEC or EHEC infection, N-WASP and the Arp2/3
complex (which consists of seven host proteins) are recruited downstream of
the Tir-interacting protein (Nck or EspFu), leading to the generation of actin
filaments beneath the attached bacteria and the formation of the pedestal
structure. Numerous proteins are found in EPEC pedestals (some of which
are listed in the shaded box); however, the precise organization of these
proteins in EPEC- and EHEC-induced pedestal generation has not been
clearly shown. It has been demonstrated that the tight-junction-associated
protein ZO1 localizes to the distal portion of the actin filaments of the EPEC
pedestal. In addition, the actin-disassembly protein cofilin has been shown
to localize to pedestals and presumably, together with the actin-assembly
protein profilin, regulates the actin-filament dynamics in pedestals. Also,
the amino terminus of Tir has been shown to bind directly to α-actinin,
but the effect of this interaction is unknown.
NATURE|Vol 449|18 October 2007
(IKK), the activity of which is stimulated by PRRs. Phosphorylated IκB
is then modified with ubiquitin and undergoes proteolytic degradation
(see ref. 45 for a review).
Pathogenic microorganisms have come to ‘understand’ the NF-κB-
activation pathway and have developed strategies to circumvent it (Fig. 3).
For example, both S. flexneri and Yersinia spp. can prevent IκB from being
ubiquitylated and therefore prevent its degradation, causing NF-κB to
remain inactive in the cell cytoplasm46. These bacteria effect this through
the T3SS effector proteins OspG (from S. flexneri) and YopP/J (YopP
and YopJ being orthologous proteins from different species of Yersinia).
OspG binds to the ubiquitylated form of the E2 ubiquitin-conjugating
enzyme UBCH5B (also known as UBE2D2) and prevents the trans-
fer of ubiquitin to IκB by an E3 ubiquitin–protein ligase, even though
IκB phosphorylation still occurs47. By contrast, until recently, it was
known that YopP/J inhibits NF-κB signalling, but it was unclear whether
this results from the inhibition of IκB phosphorylation or from the
de-ubiquitylation of IκB. Intriguing new biochemical data indicate that
inhibition of phosphorylation is the mechanism of action (Box 1).
Interestingly, S. flexneri ensures evasion of innate immune responses
by altering the NF-κB-activation pathway at several points. Recent work
has shown that S. flexneri uses the T3SS effector OspF to manipulate
the physical and spatial context of DNA encoding NF-κB-responsive
genes48 (Fig. 3). Epigenetic regulation through DNA modifications such
as methylation can have marked effects on gene expression49. OspF func-
tions as a unique phosphatase (Box 1). It dephosphorylates the mitogen-
activated protein kinase (MAPK) ERK2 in the nucleus (Box 2), so ERK2
cannot then activate mitogen- and stress-activated kinase 1 (MSK1) and
MSK2 (ref. 50). This, in effect, prevents histone phosphorylation, which
is a prerequisite for NF-κB-dependent transcription48. Therefore, genes
that are usually transcriptionally activated by NF-κB in response to the
detection of S. flexneri remain silent.
Altering antigen presentation
The adaptive immune response functions together with the innate
immune response but is pathogen specific. It therefore initially requires
precise pathogen identification, and it culminates in robust induction
of immunity. Bacteria are recognized and internalized by specialized
cells known as antigen-presenting cells (APCs) (Fig. 4). The micro-
organisms are then degraded by proteases that are sequestered in
specialized membrane-enclosed vesicular compartments. The degraded
microbial components (peptides) then bind to host proteins known as
major histocompatibility complex (MHC) class II molecules, and these
antigen complexes are transported to the host-cell surface, where they
are presented to other immune cells.
Several microbial pathogens subvert the initiation of adaptive immune
responses. Although the mechanisms that these pathogens use have not
been defined completely at the molecular level, it is clear that these mech-
anisms are diverse. For example, S. enterica can block antigen presentation
by dendritic cells, an important class of APC, through an incompletely
defined mechanism (Fig. 4). This is accomplished by inducing a decrease
in the number of peptide-bound MHC class II molecules on the surface of
infected dendritic cells51, resulting in the activation (and proliferation)
of fewer T cells, an important class of adaptive immune cell.
Yersinia spp. also alter antigen presentation by dendritic cells (Fig. 4),
and the mechanism for this process is better understood than that used
by S. enterica. Yersinia enterocolitica uses the T3SS effector YopP/J, which
inhibits adaptive immune responses by affecting the canonical MAPK-
signalling pathway (see ref. 52 for a review of MAPK-signalling path-
ways). YopP/J prevents phosphorylation of the MAPKs JNK and p38
MAPK, leading dendritic cells to take up less antigen by clathrin-medi-
ated endocytosis53. Presumably, this interference in the MAPK-signalling
cascade results from the acetyltransferase activity of YopP/J, but this has
yet to be determined. Nevertheless, by reducing antigen uptake, Yersinia
spp. elicit a similar response to that induced by S. enterica: they restrict the
proliferation of T cells and thereby limit the adaptive immune response.
These findings that YopP/J has more than one effect raise the intriguing
idea that an effector protein can have several functions in the host cell.
Given the large number of effector proteins that are now being identified
to be produced by bacterial pathogens (see the section Coordination of the
attack), the existence of multifunctional effectors is noteworthy.
Yersinia spp. and other microbial pathogens have in common another
clever strategy to counteract adaptive immune responses. They eradicate
APCs by inducing apoptosis. YopP/J is also a key component of this
process during infection with Yersinia spp. This effect could result from
abrogation of the potent anti-apoptotic signalling stimulus provided by
NF-κB signalling. For example, NF-κB induces the production of the
anti-apoptotic regulator Bcl-2, and, as previously mentioned, YopP/J
can disrupt NF-κB signalling effectively. Then, with no APCs present
to activate T cells, the proliferation of these immune cells is abrogated,
halting the adaptive immune response.
Interestingly, Shigella spp. and S. enterica induce the death of macro-
phages, by an unknown NF-κB-independent mechanism that relies
on signalling through the key pro-inflammatory activator caspase 1
(refs 7, 54) (Fig. 4). It was thought initially that these bacteria use
T3SS effectors to induce cell death. However, recent evidence indicates
that caspase 1 activation is mediated through the detection of bacte-
rial flagellin by IPAF (also known as NLRC4), a cytosolic PRR55. Intri-
guingly, the secretion of flagellin into the host-cell cytosol by S. enterica
depends on a functional T3SS. The mechanism of caspase-1-mediated
macrophage death (now termed pyroptosis) is emerging, but the reason
The Shigella flexneri effector OspF is noteworthy not only for its
biochemistry but also for its localization. OspF dephosphorylates
ERK2 in the host-cell nucleus48 (Fig. 3). OspF is one of the few bacterial
effectors that are known to translocate to the nucleus48. Others include
the Salmonella enterica effector SspH1, the Yersinia spp. effector YopM,
and another S. flexneri effector, IpaH9.8 (Fig. 1e).
Recently, the nuclear localization of these effectors has been
considered together with their functions, shedding light on the unique
targeting of these effectors. For example, IpaH9.8 lacks a conventional
nuclear-localization signal (NLS), but it is targeted to the nucleus70.
IpaH9.8 has now been shown to bind to the host splicing factor U2AF35,
impairing the U2AF35-mediated processing of messenger RNAs that
encode products important for pro-inflammatory signalling pathways
(for example, interleukin-8)71. The interaction of IpaH9.8 and U2AF35
provides a plausible mechanism for the nuclear translocation of
IpaH9.8, given that U2AF35 shuttles between the cytosol and nucleus.
Similarly, SspH1 (from S. enterica) is translocated to the host-cell
nucleus, where it decreases the effects of interleukin-8-mediated
signalling, by inhibiting NF-κB-dependent gene transcription72. The
recent finding that the host protein kinase PKN1, which can translocate
to the nucleus and decrease NF-κB signalling, binds SspH1 provides
a plausible mechanism for both the targeting and the function of this
bacterial effector73. However, given that PKN1 is a substrate of SspH1
(that is, PKN1 is modified by the E3 ubiquitin–protein ligase activity
of SspH1; see Box 1), the interaction between these two proteins is
By contrast, YopM (from Yersinia spp.) is targeted to the nucleus by
two unconventional NLSs that are present within the coding sequence74.
Recently, this effector was identified to form a tripartite complex with
two host protein kinases: protein-kinase-C-like 2 (PRK2; also known
as PKN2) and ribosomal protein S6 kinase 1 (ref. 42). Although this
complex could be found in the nucleus, it was much more abundant in
the cytosol, and the functional importance of the nuclear targeting of
YopM remains in question.
Perhaps the most interesting of the T3SS effectors that translocate
to the nucleus are the AvrBs3-like proteins, which are secreted by
the plant pathogens Xanthomonas spp. These effectors translocate to
the nucleus as a result of a carboxy-terminal NLS that is proximal to
a transcriptional-activation domain75. It is proposed that the AvrBs3-
like proteins alter the transcription of host genes directly, although
promoters modulated by these proteins have yet to be identified.
Effector proteins from mammalian pathogens have not been reported
to have a similar activity.
Box 2 | Nuclear translocation of bacterial effectors
NATURE|Vol 449|18 October 2007
for its existence is enigmatic. Indeed, the induction of cell death as a bac-
terial immune-evasion strategy is controversial. However, the outcome
of pathogen-induced host-cell death is context specific and therefore
might either harm or benefit the host (see ref. 56 for a review).
The above examples illustrate that bacterial pathogens have evolved
different strategies to ensure their preservation when they interact with
host cells, and these most commonly involve the manipulation of host-
cell intracellular-signalling pathways by secreted bacterial effectors.
Considering the coverage of discrete steps targeted in a given signalling
pathway, microbial ‘understanding’ of eukaryotic signalling pathways
is truly astounding.
Despite progress towards understanding the molecular mechanisms
that underlie microbial pathogenesis, much remains to be learned about
the bacterial effector proteins that are central to this process. In this
section, we outline some of the important unanswered questions about
effector acquisition, evolution and coordination.
Selection and evolution of virulence factors
It is clear that microbial pathogens have diverse mechanisms for inter-
acting with, and manipulating, host cells and for evading host immune
responses. Presumably, bacterial pathogens have evolved these processes
because they provide a selective advantage. Curiously, pathogens often
encode several related copies of effectors: for example, EHEC produces
at least 14 variants of the T3SS effector NleG, and L. monocytogenes
produces several internalin proteins. Although, in some cases, these
proteins might target related variants of host-cell components, they are
presumably redundant in other cases. Why bacterial pathogens main-
tain such a vast collection of seemingly redundant effectors is controver-
sial, although recent evidence indicates that bacterial effector proteins
are also crucial for successful transmission between hosts57.
It is now understood that the genetic transfer of effector-encoding
genes between bacteria (by bacteriophages, conjugation or transforma-
tion) has a key role in generating diversity in pathogens (and in gen-
erating new diseases). For example, it has recently been shown that a
bacterial nucleoid protein, H-NS, can silence genes when they are ini-
tially acquired by horizontal transfer. These genes are then integrated
into various regulatory pathways, including virulence regulons (in which
the silencing is presumably removed), allowing the encoded molecules
to participate in virulence pathways58.
Coordination of the attack
Pathogens with a T3SS commonly have a large repertoire of effec-
tors (often tens to hundreds). For example, recent studies have shown
that EHEC has at least 40 T3SS effectors, whereas the plant pathogen
Pseudomonas syringae has 190 T3SS effectors59,60. Similar numbers of
effectors are also present in pathogens with a T4SS, such as Legionella
pneumophila. This raises an important question: how are the expres-
sion, secretion and functional delivery of these effectors regulated and,
perhaps more crucially, coordinated? Virulence-factor coordination was
recently reported in the Gram-positive pathogen Staphylococcus aureus:
the production of its pore-forming toxin Panton–Valentine leukocidin
was shown to increase the production of Spa, a known S. aureus viru-
lence factor61. The authors of this study suggested that Panton–Valen-
tine leukocidin and Spa function together to exacerbate pathogenesis,
although how they do so is unclear. Such effector coordination is also
likely to be found in other pathogens with a T3SS or T4SS.
There is no doubt that specialized secretion systems need to coordinate
the delivery of effectors to maintain an overall virulence strategy; however,
this has not been documented. It also has not been documented, although
it is presumed, that adherence precedes effector delivery. In theory, patho-
gens could deliver different specialized sets of effectors to different types of
host cell (for example, a macrophage and an epithelial cell) or to different
tissue sites in a host, and different sets of effectors could also be used to
target different host species. Moreover, it is conceivable that secretion sys-
tems could work in reverse: that is, they could acquire host-cell molecules,
Figure 3 | Subversion of NF-κB-mediated signalling. The transcription
factor NF-κB initiates the expression of genes that encode many innate
immune factors. The NF-κB-signalling pathway is therefore crucial to
the host. Various pathogens can subvert this pathway at different points.
After pathogenic bacteria are detected by PRRs (TLRs and/or NLRs),
a signalling cascade is triggered, resulting in the phosphorylation of
the protein-kinase complex IKK. Activated IKK then catalyses the
phosphorylation of the inhibitor of NF-κB, IκB. Ubiquitin, carried by an
E2 ubiquitin-conjugating enzyme, is attached to phosphorylated IκB by
an E3 ubiquitin–protein ligase, marking IκB for degradation in the cytosol
and releasing NF-κB to translocate to the nucleus. The induction of gene
expression by NF-κB requires remodelling of chromatin through the
phosphorylation of histones. This is mediated by activated MSK, a protein
kinase that is activated by the MAPK ERK2. The points at which protein
effectors secreted by pathogenic bacteria interfere with NF-κB-mediated
gene expression are indicated in red. YopP/J, produced by Yersinia spp.,
disarms IKK by competitively acetylating key amino-acid residues,
thereby preventing their phosphorylation. Shigella flexneri subverts
this signalling pathway at two points, in the cytosol and in the nucleus.
The S. flexneri effector OspG ‘derails’ IκB ubiquitylation by binding
the ubiquitylated E2 molecule. By contrast, another S. flexneri effector,
OspF, prevents chromatin remodelling, through dephosphorylating
NATURE|Vol 449|18 October 2007
signals or even energy and nutrients for the bacteria, although there is no
evidence to support this idea at present.
A recent bioinformatic study59 shows that the complexity of these viru-
lence systems might be even greater than has been thought. The authors of
this study proposed that pathogenic microorganisms can instan taneously
produce novel chimaeric hybrids of T3SS effectors through a pro cess
referred to as terminal reassortment59. By fusing new protein-coding
sequences to sequences that control the expression and secretion of T3SS
effectors, microorganisms can ‘sample’ new combinations of secreted
effectors. This seemingly hastened effector evolution, coupled with the
propensity of the genes encoding these ‘shuffled’ effectors to be located on
mobile genetic elements and to confer strong selective pressure through
either virulence or transmission, suggests that pathogenic micro organisms
can coordinate the induced host-cell biology so that pathogenesis is
optimized for the benefit of the microorganism. How induced functional
responses are coordinated in the host cell is poorly understood.
Pathogenesis in the bigger picture
Pathogenic bacteria have the arduous task of interacting with host cells
and reprogramming the complex molecular and cellular networks of
these cells to allow bacterial replication and spread, while countering
host-defence strategies. Evolution and transmission have shaped this
bacterial pursuit through the accumulation of (sometimes) vast arse-
nals of genes that encode effector proteins, which are probably subject
to complex regulation. Piecemeal study of these arsenals might help to
define them, but a deeper understanding of their mechanisms and of
potential intervention points would best be achieved by considering the
delivery, coordination and mechanistic functions of these arsenals as a
whole. The current challenge is to assemble a cross-disciplinary toolbox
that will enable pathogenesis to be studied at the ‘systems’ level.
The field of bacterial pathogenesis is a rapidly evolving and expand-
ing one. As the vastness of effector functions is being realized, it is
a considerable challenge to integrate the numerous host-cell targets
and to translate this knowledge into an accurate understanding of the
mechanisms by which effector proteins cause disease.
Moving from studying cultured cells to relevant animal disease models
is crucial for understanding disease, yet such studies are often neglected,
because cell-culture-based systems are easier to manipulate. However,
the opportunity to study pathogenesis in relevant animal models is now
within reach, because the current understanding of the mechanistic
details of the host–pathogen interface, some of which have been outlined
in this article, allows a directed approach to the problem. An elegant
example is the recent re-engineering of an L. monocytogenes internalin
protein to extend the host range of this bacterium to include mice62.
Conversely, the host can now be engineered such that it is susceptible
to infection63. These two studies present a glorious opportunity to probe
the host–pathogen interface during disease. Similarly, genomic stud-
ies have led to the identification of mutations in humans that alter the
outcome of bacterial infections64. The recent realization that the host
microbiota has a crucial role in mediating the outcome of disease adds
another layer of complexity65.
Recognizing that pathogens can overrun crucial host-cell pathways
by using a myriad of mechanisms has led to an increased understanding
of microbiology, cell biology, biochemistry and immunology. However,
this knowledge now needs to be advanced to the point at which it can be
translated into a true understanding of disease. This remains the crucial
challenge to all who are involved in this field. Only then will it be pos-
sible to target these effector mechanisms rationally as a preventive or
1. Galan, J. E. & Wolf-Watz, H. Protein delivery into eukaryotic cells by type III secretion
machines. Nature 444, 567–573 (2006).
Pizarro-Cerda, J. & Cossart, P. Bacterial adhesion and entry into host cells. Cell 124,
Cossart, P. & Sansonetti, P. J. Bacterial invasion: the paradigms of enteroinvasive
pathogens. Science 304, 242–248 (2004).
Stevens, J. M., Galyov, E. E. & Stevens, M. P. Actin-dependent movement of bacterial
pathogens. Nature Rev. Microbiol. 4, 91–101 (2006).
Finlay, B. B. Bacterial virulence strategies that utilize Rho GTPases. Curr. Top. Microbiol.
Immunol. 291, 1–10 (2005).
Meresse, S. et al. Controlling the maturation of pathogen-containing vacuoles: a matter of
life and death. Nature Cell Biol. 1, E183–E188 (1999).
Gao, L. & Abu Kwaik, Y. Hijacking of apoptotic pathways by bacterial pathogens. Microbes
Infect. 2, 1705–1719 (2000).
Finlay, B. B. & McFadden, G. Anti-immunology: evasion of the host immune system by
bacterial and viral pathogens. Cell 124, 767–782 (2006).
Hardt, W. D., Chen, L. M., Schuebel, K. E., Bustelo, X. R. & Galan, J. E. S. typhimurium
encodes an activator of Rho GTPases that induces membrane ruffling and nuclear
responses in host cells. Cell 93, 815–826 (1998).
10. Stender, S. et al. Identification of SopE2 from Salmonella typhimurium, a conserved guanine
nucleotide exchange factor for Cdc42 of the host cell. Mol. Microbiol. 36, 1206–1221
MHC class II
Figure 4 | Prevention of antigen presentation by APCs.
Components of the adaptive immune system are activated
after they recognize fragments of pathogens that are
presented at the surface of infected host cells. Antigens
from pathogenic bacteria are presented in this way after the
bacteria have been taken up into endocytic vesicles that then
fusewith cellular-protease-containing vesicles, in which the
bacterial pathogens are degraded. Proteolytically processed
bacterial components bind to MHC class II molecules, and
the vesicles containing these complexes are transported to
the cell surface, where the bacterial peptides are displayed to
immune cells. Bacterial pathogens interfere with this antigen
processing and presentation pathway at several points,
which are indicated in red. Yersinia enterocolitica produces
YopP/J, which subverts antigen presentation by inhibiting
the MAPK-signalling pathway and thereby preventing
clathrin-mediated endocytosis. Salmonella enterica secretes
an unidentified effector that prevents antigen presentation,
although the targeted step also remains unidentified (dashed
blocking arrows). In addition, S. enterica can prevent antigen
presentation by inducing cell death. This is thought to occur
through the T3SS-mediated delivery of flagellin into the
cytosol. Flagellin is subsequently detected by the PRR known
as IPAF, which initiates a signalling cascade that results in the
activation of caspase 1. Caspase 1 then mediates macrophage
death by the newly defined mechanism called pyroptosis.
Shigella spp. are also thought to induce macrophage death in
the same manner.
NATURE|Vol 449|18 October 2007
11. Zhou, D., Chen, L. M., Hernandez, L., Shears, S. B. & Galan, J. E. A Salmonella inositol Download full-text
polyphosphatase acts in conjunction with other bacterial effectors to promote host
cell actin cytoskeleton rearrangements and bacterial internalization. Mol. Microbiol. 39,
12. Alto, N. M. et al. Identification of a bacterial type III effector family with G protein mimicry
functions. Cell 124, 133–145 (2006).
13. Egile, C. et al. Activation of the CDC42 effector N-WASP by the Shigella flexneri IcsA
protein promotes actin nucleation by Arp2/3 complex and bacterial actin-based motility.
J. Cell Biol. 146, 1319–1332 (1999).
14. Chakraborty, T. et al. A focal adhesion factor directly linking intracellularly motile Listeria
monocytogenes and Listeria ivanovii to the actin-based cytoskeleton of mammalian cells.
EMBO J. 14, 1314–1321 (1995).
15. Welch, M. D., Iwamatsu, A. & Mitchison, T. J. Actin polymerization is induced by Arp2/3
protein complex at the surface of Listeria monocytogenes. Nature 385, 265–269 (1997).
16. Gruenheid, S. et al. Enteropathogenic E. coli Tir binds Nck to initiate actin pedestal
formation in host cells. Nature Cell Biol. 3, 856–859 (2001).
17. Campellone, K. G., Robbins, D. & Leong, J. M. EspFU is a translocated EHEC effector that
interacts with Tir and N-WASP and promotes Nck-independent actin assembly. Dev. Cell
7, 217–228 (2004).
18. Garmendia, J. et al. TccP is an enterohaemorrhagic Escherichia coli O157:H7 type III effector
protein that couples Tir to the actin-cytoskeleton. Cell. Microbiol. 6, 1167–1183 (2004).
19. Shaner, N. C., Sanger, J. W. & Sanger, J. M. Actin and α-actinin dynamics in the adhesion
and motility of EPEC and EHEC on host cells. Cell Motil. Cytoskeleton 60, 104–120 (2005).
20. Cantarelli, V. V. et al. Cortactin is necessary for F-actin accumulation in pedestal structures
induced by enteropathogenic Escherichia coli infection. Infect. Immun. 70, 2206–2209
21. Goosney, D. L., DeVinney, R. & Finlay, B. B. Recruitment of cytoskeletal and signaling
proteins to enteropathogenic and enterohemorrhagic Escherichia coli pedestals. Infect.
Immun. 69, 3315–3322 (2001).
22. Goosney, D. L. et al. Enteropathogenic E. coli translocated intimin receptor, Tir, interacts
directly with α-actinin. Curr. Biol. 10, 735–738 (2000).
23. Unsworth, K. E. et al. Dynamin is required for F-actin assembly and pedestal formation by
enteropathogenic Escherichia coli (EPEC). Cell. Microbiol. 9, 438–449 (2007).
24. Batchelor, M. et al. Involvement of the intermediate filament protein cytokeratin-18 in actin
pedestal formation during EPEC infection. EMBO Rep. 5, 104–110 (2004).
25. Hanajima-Ozawa, M. et al. Enteropathogenic Escherichia coli, Shigella flexneri, and Listeria
monocytogenes recruit a junctional protein, zonula occludens-1, to actin tails and pedestals.
Infect. Immun. 75, 565–573 (2007).
26. Yoshida, S. et al. Shigella deliver an effector protein to trigger host microtubule
destabilization, which promotes Rac1 activity and efficient bacterial internalization.
EMBO J. 21, 2923–2935 (2002).
27. Hardwidge, P. R. et al. Modulation of host cytoskeleton function by the enteropathogenic
Escherichia coli and Citrobacter rodentium effector protein EspG. Infect. Immun. 73,
28. Hu, L. & Kopecko, D. J. Campylobacter jejuni 81-176 associates with microtubules and
dynein during invasion of human intestinal cells. Infect. Immun. 67, 4171–4182 (1999).
29. Roy, C. R. & Tilney, L. G. The road less traveled: transport of Legionella to the endoplasmic
reticulum. J. Cell Biol. 158, 415–419 (2002).
30. Knodler, L. A. & Steele-Mortimer, O. Taking possession: biogenesis of the Salmonella-
containing vacuole. Traffic 4, 587–599 (2003).
31. Nagai, H. et al. A C-terminal translocation signal required for Dot/Icm-dependent delivery
of the Legionella RalF protein to host cells. Proc. Natl Acad. Sci. USA 102, 826–831 (2005).
32. Robinson, C. G. & Roy, C. R. Attachment and fusion of endoplasmic reticulum with
vacuoles containing Legionella pneumophila. Cell. Microbiol. 8, 793–805 (2006).
33. Murata, T. et al. The Legionella pneumophila effector protein DrrA is a Rab1 guanine
nucleotide-exchange factor. Nature Cell Biol. 8, 971–977 (2006).
34. Steele-Mortimer, O., Meresse, S., Gorvel, J. P., Toh, B. H. & Finlay, B. B. Biogenesis of
Salmonella typhimurium-containing vacuoles in epithelial cells involves interactions with
the early endocytic pathway. Cell. Microbiol. 1, 33–49 (1999).
35. Cuellar-Mata, P. et al. Nramp1 modifies the fusion of Salmonella typhimurium-containing
vacuoles with cellular endomembranes in macrophages. J. Biol. Chem. 277, 2258–2265
36. Drecktrah, D., Knodler, L. A., Howe, D. & Steele-Mortimer, O. Salmonella trafficking is
defined by continuous dynamic interactions with the endolysosomal system. Traffic
8, 212–225 (2007).
37. Sansonetti, P. J., Ryter, A., Clerc, P., Maurelli, A. T. & Mounier, J. Multiplication of Shigella
flexneri within HeLa cells: lysis of the phagocytic vacuole and plasmid-mediated contact
hemolysis. Infect. Immun. 51, 461–469 (1986).
38. Veiga, E. & Cossart, P. Listeria hijacks the clathrin-dependent endocytic machinery to
invade mammalian cells. Nature Cell Biol. 7, 894–900 (2005).
39. Shaughnessy, L. M., Hoppe, A. D., Christensen, K. A. & Swanson, J. A. Membrane
perforations inhibit lysosome fusion by altering pH and calcium in Listeria monocytogenes
vacuoles. Cell. Microbiol. 8, 781–792 (2006).
40. Terebiznik, M. R. et al. Elimination of host cell PtdIns(4,5)P2 by bacterial SigD promotes
membrane fission during invasion by Salmonella. Nature Cell Biol. 4, 766–773 (2002).
41. Prehna, G., Ivanov, M. I., Bliska, J. B. & Stebbins, C. E. Yersinia virulence depends on mimicry
of host Rho-family nucleotide dissociation inhibitors. Cell 126, 869–880 (2006).
42. McDonald, C., Vacratsis, P. O., Bliska, J. B. & Dixon, J. E. The Yersinia virulence factor YopM
forms a novel protein complex with two cellular kinases. J. Biol. Chem. 278, 18514–18523
43. Hayden, M. S., West, A. P. & Ghosh, S. NF-κB and the immune response. Oncogene 25,
44. Akira, S., Uematsu, S. & Takeuchi, O. Pathogen recognition and innate immunity. Cell 124,
45. Perkins, N. D. Post-translational modifications regulating the activity and function of the
nuclear factor κB pathway. Oncogene 25, 6717–6730 (2006).
46. Angot, A., Vergunst, A., Genin, S. & Peeters, N. Exploitation of eukaryotic ubiquitin
signaling pathways by effectors translocated by bacterial type III and type IV secretion
systems. PLoS Pathog. 3, e3 (2007).
47. Kim, D. W. et al. The Shigella flexneri effector OspG interferes with innate immune
responses by targeting ubiquitin-conjugating enzymes. Proc. Natl Acad. Sci. USA 102,
48. Arbibe, L. et al. An injected bacterial effector targets chromatin access for transcription
factor NF-κB to alter transcription of host genes involved in immune responses. Nature
Immunol. 8, 47–56 (2007).
49. Kouzarides, T. Chromatin modifications and their function. Cell 128, 693–705
50. Soloaga, A. et al. MSK2 and MSK1 mediate the mitogen- and stress-induced
phosphorylation of histone H3 and HMG-14. EMBO J. 22, 2788–2797 (2003).
51. Cheminay, C., Mohlenbrink, A. & Hensel, M. Intracellular Salmonella inhibit antigen
presentation by dendritic cells. J. Immunol. 174, 2892–2899 (2005).
52. Ashwell, J. D. The many paths to p38 mitogen-activated protein kinase activation in the
immune system. Nature Rev. Immunol. 6, 532–540 (2006).
53. Autenrieth, S. E. et al. Yersinia enterocolitica YopP inhibits MAP kinase-mediated antigen
uptake in dendritic cells. Cell. Microbiol. 9, 425–437 (2007).
54. Scott, A. M. & Saleh, M. The inflammatory caspases: guardians against infections and
sepsis. Cell Death Differ. 14, 23–31 (2007).
55. Miao, E. A. et al. Cytoplasmic flagellin activates caspase-1 and secretion of interleukin-1β
via Ipaf. Nature Immunol. 7, 569–575 (2006).
56. DeLeo, F. R. Modulation of phagocyte apoptosis by bacterial pathogens. Apoptosis 9,
57. Wickham, M. E., Brown, N. F., Boyle, E. C., Coombes, B. K. & Finlay, B. B. Virulence is
positively selected by transmission success between mammalian hosts. Curr. Biol. 17,
58. Navarre, W. W. et al. Selective silencing of foreign DNA with low GC content by the H-NS
protein in Salmonella. Science 313, 236–238 (2006).
59. Stavrinides, J., Ma, W. & Guttman, D. S. Terminal reassortment drives the quantum
evolution of type III effectors in bacterial pathogens. PLoS Pathog. 2, e104 (2006).
60. Tobe, T. et al. An extensive repertoire of type III secretion effectors in Escherichia coli O157
and the role of lambdoid phages in their dissemination. Proc. Natl Acad. Sci. USA 103,
61. Labandeira-Rey, M. et al. Staphylococcus aureus Panton–Valentine leukocidin causes
necrotizing pneumonia. Science 315, 1130–1133 (2007).
62. Wollert, T. et al. Extending the host range of Listeria monocytogenes by rational protein
design. Cell 129, 891–902 (2007).
63. Lecuit, M. et al. A transgenic model for listeriosis: role of internalin in crossing the intestinal
barrier. Science 292, 1722–1725 (2001).
64. Casanova, J. L. & Abel, L. Human genetics of infectious diseases: a unified theory. EMBO J.
26, 915–922 (2007).
65. Lupp, C. et al. Host-mediated inflammation disrupts the intestinal microbiota and
promotes the overgrowth of Enterobacteriaceae. Cell Host Microbe 2, 119–129 (2007).
66. Rohde, J. R., Breitkreutz, A., Chenal, A., Sansonetti, P. J. & Parsot, C. Type III secretion
effectors of the IpaH family are E3 ubiquitin ligases. Cell Host Microbe 1, 77–83 (2007).
67. Rytkonen, A. et al. SseL, a Salmonella deubiquitinase required for macrophage killing and
virulence. Proc. Natl Acad. Sci. USA 104, 3502–3507 (2007).
68. Mukherjee, S. et al. Yersinia YopJ acetylates and inhibits kinase activation by blocking
phosphorylation. Science 312, 1211-1214 (2006).
69. Li, H. et al. The phosphothreonine lyase activity of a bacterial type III effector family. Science
315, 1000–1003 (2007).
70. Toyotome, T. et al. Shigella protein IpaH9.8 is secreted from bacteria within mammalian cells
and transported to the nucleus. J. Biol. Chem. 276, 32071–32079 (2001).
71. Okuda, J. et al. Shigella effector IpaH9.8 binds to a splicing factor U2AF35 to modulate host
immune responses. Biochem. Biophys. Res. Commun. 333, 531–539 (2005).
72. Haraga, A. & Miller, S. I. A Salmonella enterica serovar Typhimurium translocated leucine-
rich repeat effector protein inhibits NF-κB-dependent gene expression. Infect. Immun. 71,
73. Haraga, A. & Miller, S. I. A Salmonella type III secretion effector interacts with the
mammalian serine/threonine protein kinase PKN1. Cell. Microbiol. 8, 837–846 (2006).
74. Benabdillah, R., Mota, L. J., Lutzelschwab, S., Demoinet, E. & Cornelis, G. R. Identification of
a nuclear targeting signal in YopM from Yersinia spp. Microb. Pathog. 36, 247–261
75. Schornack, S., Meyer, A., Romer, P., Jordan, T. & Lahaye, T. Gene-for-gene-mediated
recognition of nuclear-targeted AvrBs3-like bacterial effector proteins. J. Plant Physiol. 163,
Acknowledgements We thank members of B.B.F.’s laboratory for helpful
discussions and critical reading of the manuscript. We gratefully acknowledge
F. Ness for assistance with the preparation of figures. We apologize to authors
whose work could not be cited as a result of space restrictions. Work in B.B.F.’s
laboratory is supported by grants from the Canadian Institutes of Health Research
(CIHR), the Howard Hughes Medical Institute (HHMI), the Foundation for
the National Institutes of Health, and Genome Canada. A.P.B. is supported by
fellowships from the CIHR and the Michael Smith Foundation for Health Research
(MSFHR). J.A.G. is supported by a Canadian Association for Gastroenterology/
CIHR/AstraZeneca fellowship and a fellowship from the MSFHR. B.B.F. is a CIHR
Distinguished Investigator, an HHMI International Research Scholar, and the Peter
Wall Distinguished Professor, at the University of British Columbia.
Author Information Reprints and permissions information is available at
npg.nature.com/reprints. Correspondence should be addressed to B.B.F.
NATURE|Vol 449|18 October 2007