REVIEWS

Scaffold proteins and immune-cell

signalling
Andrey S. Shaw* and Erin L. Filbert*

Abstract | Over the past 20 years great progress has been made in defining most of the key
signalling pathways that functionally regulate immune cells. Recently, it has become clear that
scaffold proteins have a crucial role in regulating many of these signalling cascades. By binding
two or more components of a signalling pathway, scaffold proteins can help to localize signalling
molecules to a specific part of the cell or to enhance the efficacy of a signalling pathway. Scaffold
proteins can also affect the thresholds and the dynamics of signalling reactions by coordinating
positive and negative feedback signals. In this Review, we focus on recent progress in the
understanding of the function of scaffold proteins in immune cells.
Immunological synapse
A large junctional structure
that is formed at the cell
surface between a T cell that is
interacting with an APC or a
target cell, which consists of
molecules that are required for
adhesion and signalling. This
structure is important for
establishing Tcell adhesion
and polarity, is influenced
by the cytoskeleton and
transduces highly controlled
secretory signals, thereby
allowing the directed release
of cytokines or lytic granules
towards the APC or target cell.

*Department of Pathology
and Immunology and

Howard Hughes Medical

Institute, Washington
University School of Medicine,
660 South Euclid, Saint Louis,
Missouri 63110, USA.
Correspondence to A.S.S.
e-mail: [email protected]
doi:10.1038/nri2473

During the past 20 years, it has become clear that signalling cascades are more than two-dimensional pathways
composed of proteins that can just be turned on or off.
Instead, the location of the proteins inside the cell and
the kinetics of their activation are important features of
signal-transduction pathways. How the signalling molecules are localized in the cell and how the strength and
quality of the signal is regulated is a largely unexplored
area of research, but increasing attention has been
focused on scaffold proteins as candidate molecules
that control these more complex aspects of signal transduction. The term scaffold has been used loosely by the
scientific community and, to our knowledge, there is
no strict definition. Here, we define a scaffold protein
as a molecule that binds to at least two other signalling
proteins. Inherent in this definition is the ability of a
scaffold protein to regulate signal transduction and, in
most cases, to localize signalling molecules at specific
areas of the cell, such as the plasma membrane, the
cytoplasm, the nucleus, the Golgi, the endosomes and
the mitochondria.
Receptor tyrosine kinases, such as the platelet-derived
growth factor receptor and the epidermal growth factor
receptors, were the first examples of signalling scaffold
proteins to be described1. Binding of a ligand to these
receptors results in tyrosine autophosphorylation of
the cytoplasmic domain of the receptor at sites where
SRC homology 2 (SH2)-domain-containing proteins
can bind. It is the combination of the signalling proteins that are recruited to the scaffold protein that then
determines the quality of the response that is generated.
Importantly, the cytoplasmic domain of these receptor
tyrosine kinases (which acts as the scaffold) also serves

to initially localize the signalling response to the plasma

membrane. In addition, the endocytosis of the activated
receptor and its subsequent intracellular trafficking are
important events in the localization of the signalling
response to specific areas of the cell2,3.
In the immune system, scaffold proteins that are
similar to receptor tyrosine kinases include immunoreceptor tyrosine-based activation motif (ITAM)containing receptors, such as the T-cell receptor
(TCR) and the B-cell receptor (BCR), and cytoplasmic scaffold proteins, such as linker for activation of
T cells (LAT), B-cell linker (BLNK) and SH2-domaincontaining leukocyte protein of 76 kDa (SLP76)46.
Each of these scaffold proteins can be phosphorylated
and can bind specific SH2-domain-containing signalling molecules. Similarly to receptor tyrosine kinases,
ITAM-containing receptors can localize the signalling
response to the plasma membrane and to the immuno
logical synapse. The general role of tyrosine-based scaffold proteins is well known and has been extensively
reviewed79, so in this Review we focus our discussion
on recently identified scaffold proteins, the function
of which is not based on tyrosine phosphorylation. We
hope that by reviewing these unfamiliar molecules and
their potential function in the regulation of signalling
reactions we will stimulate and aid future research on
signal-transduction pathways.

The function of scaffold proteins

Most of our understanding of the function of scaffold
molecules has come from mathematical modelling and
from engineered scaffold proteins. These approaches
have shown that scaffolds can function in at least four

NATuRE REvIEWS | Immunology

voLuME 9 | jANuARy 2009 | 47

2009 Macmillan Publishers Limited. All rights reserved

REVIEWS
ways: by acting as platforms on which signalling molecules can assemble, by localizing signalling molecules
at specific sites in a cell, by coordinating positive and
negative feedback signals to modify the signalling
pathways and by protecting activated signalling molecules from inactivation. These functions of scaffold
proteins can provide additional complexity to the
signalling cascade and create signalling thresholds or
regulate complex signalling behaviours, such as graded
or digital signalling, transient or sustained signalling,
and oscillatory signalling10 (FIG. 1).
Assembly of signalling components. The most basic
function of scaffold proteins is the assembly of the signalling components of a cascade. This could enhance
the efficiency of the signalling cascade by concentrating signalling components, as well as enhance signalling
specificity by preventing spurious interactions between
signalling proteins. For example, a scaffold might function to enhance the specificity of kinase phosphorylation by binding both the protein kinase and its substrate.
As some signalling proteins require multiple interactions for activation, scaffold proteins could facilitate the
a
Assembly

Scaffold
protein

Cell
membrane

Localization
Component of a
signalling cascade

Cytoplasm

Protection from phosphatases

Feedback

Phosphatase

Amplification

Inhibition

Analogue (graded)
signal

Digital (switch-like)
signal

Sustained or
transient signal

Oscillatory signal

Figure 1 | The function of scaffold proteins. a | Scaffold proteins can have at

least four functions: to assemble components of a signalling pathway, to localize
components of a signalling pathway to a specific intracellular compartment or
Nature
Reviewsactive
| Immunology
location, to regulate positive or negative feedback signals and
to protect
signalling intermediate proteins from deactivation by phosphatases. b | These
functions can lead to the generation of complex signalling behaviours, which include:
analogue (or graded) signalling, in which the signal output is proportional to the
input; digital (or switch-like) signalling, in which signalling only occurs after a certain
threshold is reached and the output is always maximal; sustained or transient
signalling; and oscillatory signalling10.

conversion of a distributive process (that is, multiple

independent interactions that result in many modifications) to a processive process (that is, a single interaction that results in many modifications)11,12. The
interaction of signalling proteins with the scaffold protein could also result in allosteric changes in the signalling protein that either enhance or inhibit the activation
process; therefore, some scaffolds can be referred to as
catalytic scaffolds13.
Localization of signalling molecules in the cell. Another
important function of scaffold proteins is to localize the
signalling reaction to a specific area in the cell (FIG. 1),
which might be important for the local production of
signalling intermediates. For example, A-kinase anchor
proteins (AKAPs) target cyclic AMP-dependent protein kinase (PKA; also known as PRKACA) to various
sites in the cell 14, which allows for the local regulation of PKA and the localized phosphorylation of its
substrates. Although the function of AKAPs in neurons
is well known, their function in immune cells is not
well understood. Numerous AKAPs are expressed
by T cells, in which they may be involved in T-cell
homeo stasis 15, but further investigation is needed
to fully understand the role of AKAP scaffolds in
immune cells.
Coordination of positive and negative feedback.
Scaffold proteins can potentially have other functions
that could allow for complicated signalling behaviours;
this hypothesis has been examined using engineered
scaffold proteins (BOX 1) and mathematical modelling.
Most research in immune cells has focused on the
scaffold proteins that are associated with the mitogenactivated protein kinase (MAPK) signalling cascade.
In this cascade, a small G-protein activates a MAPK
kinase kinase (MAPKKK) that in turn phosphorylates and activates a MAPK kinase (MAPKK), which
then activates and phosphorylates a MAPK (FIG. 2).
The best-studied MAPK pathway is the RASERK
(extracellular-signal-regulated kinase; also known as
MAPK1) pathway in which the small G-protein RAS
activates the MAPKKK RAF, which then activates the
MAPKK MAPK/ERK kinase 1 (MEK1; also known
as MAPKK1), which activates the MAPK ERK.
Initial ideas about the function of a three-kinase
signalling cascade proposed that a scaffold protein
could allow a weak initial stimulus to be amplified;
however, the discovery of scaffold proteins that bind
multiple components of the cascade has led to many
hypotheses about the function of a scaffold protein in
such a cascade. The scaffold protein could function to
enhance kinase specificity13 and, by binding all three
kinases at the same time, it might also limit amplification of the pathway by restricting the ability of a
kinase to phosphorylate more than one downstream
target13. Indeed, mathematical modelling of a MAPK
scaffold has suggested that the role of scaffold proteins is complex and that they can function to amplify
or inhibit a signalling cascade depending on many
variables11,16,17.

48 | jANuARy 2009 | voLuME 9

www.nature.com/reviews/immunol
2009 Macmillan Publishers Limited. All rights reserved

REVIEWS
All of the modelling studies have confirmed that
a simple scaffold protein that can bind a MAPKKK, a
MAPKK and a MAPK inhibits signal amplification by
constraining the ability of the kinase to phosphorylate
downstream substrates11,16,18. However, the general ability
of scaffold proteins to amplify or inhibit signalling may
be related to the expression levels of the signalling components11, their location in the cell, the stability of the
interactions between the scaffold and each of the kinase
components and the basal phosphatase activity in
the cell16. Indeed, it has been suggested that changes in the
expression level of the scaffold proteins could also affect
the signalling pathway11.
In addition to their use in mathematical modelling
studies, engineered scaffold proteins have been used
to further delineate the function of scaffold proteins
in a cell1921. In one study19, the yeast MAPK scaffold
protein Ste5 was engineered to have additional binding sites for either a positive or negative regulator. As
expected, the data showed that the recruitment of a
positive regulator to the scaffold protein enhanced the
signalling response, whereas the recruitment of a negative regulator dampened it. However, more complex
signalling behaviours could be induced by placing the
expression of the positive or negative regulator under
the control of a MAPK-dependent promoter or by using
MAPK-regulated expression of decoy molecules that
could compete with the regulators for binding to the
scaffold protein. using such a strategy, the authors
could engineer scaffold proteins that allowed for digital
(switch-like) responses (FIG. 1).
In addition to regulating thresholds for the initiation of a response (that is, requiring a stimulus to be
above a certain level), coordination of feedback loops
could allow signalling pathways to display an oscillatory
signalling behaviour (FIG. 1). In T cells, calcium fluxes
can oscillate repeatedly and have cycles that last a few
minutes22. Could scaffolds be modifiers of this signalling
pathway and mediate these complex behaviours?
Protecting activated signalling molecules from inactiv
ation. An important feature of all components of a
signalling pathway is their ability to be inactivated.
Inactivation could be mediated by enzymes that reverse
the activation state or induce the degradation of the

Box 1 | Using engineered scaffold proteins to study scaffold biology

Scaffold proteins physically organize proteins of a signalling pathway, but might also
optimize signal transduction by determining signalling thresholds or by controlling
positive and/or negative feedback. Engineered scaffolds are synthetic scaffold
proteins that have been designed to bind to proteins that they normally do not bind
to, using well-defined protein domains, such as leucine zippers, to facilitate the
interaction. This method can be used to recruit feedback molecules (such as kinases
or phosphatases), and studies using these proteins have shown that scaffold proteins
can indeed modulate both positive and negative feedback loops. Generation of
chimeric scaffolds has also contributed to the field of scaffold biology. Chimeric
scaffolds incorporate two pathways into a single scaffolding platform and have been
used to show that scaffold proteins can promote crosstalk between two distinct
pathways. Engineered scaffold proteins can test hypotheses on the function of
scaffolds and could act as new therapeutic agents to treat human diseases.

signalling proteins. Scaffold proteins might function to

protect activated signalling molecules from inactivation
and/or degradation.
Indeed, mathematical modelling predicts that when
phosphatase activity is high, scaffold proteins have an
important role in facilitating the activation of the cascade,
as in the absence of scaffold proteins kinases have a higher
probability of being inactivated by phosphatases before
they activate their downstream targets16. Conversely, scaffold proteins can inhibit signal transduction by kinases
when phosphatase activity is low by limiting the number
of downstream targets that can be phosphorylated.
In the remainder of this Review, we describe the role
of cytoplasmic scaffold proteins that have recently been
identified and implicated in the regulation of signalling
cascades in the immune system. Although the exact
function of most of these scaffold proteins has yet to
be fully described, an understanding of their potential
functions can help guide future research in this area.

MAPK pathway scaffolds proteins

The ERK pathway is one of the best defined signalling
pathways in biology. Because of this, several scaffold proteins that modify this pathway have been identified and
some of their functions have been characterized. Among
these are IQ-motif-containing GTPase activating protein 1, MAPK scaffold protein 1, SEF (similar expression
to FGF genes), paxillin, -arrestin and kinase suppressor
of RAS (KSR)23.
In addition to ERK, two other MAPK pathways are
crucial in immune cells: the juN N-terminal kinase
(jNK; also known as MAPK8) pathway and the p38
(also known as MAPK14) pathway, both of which sense
stress-induced responses, including those mediated by
pro-inflammatory cytokines. The scaffold proteins B-cell
lymphoma 10 (BCL-10) and MEK kinase 1 (MEKK1;
also known as MAPKKK1) are thought to be important modifiers of the jNK pathway in immune cells, and
recent evidence suggests that the scaffold protein discslarge homologue 1 (DLG1) has a role in p38 activation
(discussed later).
KSR. Because of intensive studies of the yeast MAPK scaffold protein Ste5 (reF. 24), efforts have been made to identify the mammalian equivalent of this protein. Although
sequence comparison studies indicated that mammalian
cells do not have a Ste5 homologue, the scaffold protein
KSR has emerged as the most probable equivalent of Ste5
in mammals25. KSR was originally described as a positive regulator of the RASMAPK signalling pathway in
Drosophila melanogaster and Caenorhabditis elegans2628.
It contains five conserved domains: a unique domain of
unknown function at the amino terminus, a proline-rich
domain, a cysteine-rich domain, a serine/threoninerich domain and a pseudokinase domain at the carboxyl terminus. KSR binds many proteins, including
all three components of the ERK signalling pathway
(RAF, MEK1 and ERK; FIG. 2). Binding of ERK to KSR
depends on the serine/threonine-rich region of KSR,
whereas binding to MEK1 and RAF occurs through the
pseudokinase domain25.

NATuRE REvIEWS | Immunology

voLuME 9 | jANuARy 2009 | 49

2009 Macmillan Publishers Limited. All rights reserved

REVIEWS
a

TCR
CD3

RAS
RAF

Unknown
function

Pseudokinase
domain

MEK1

Proline-rich
domain

Serine/threoninerich domain
ERK

Cysteine-rich
domain

PKC
P

CARMA1
P BCL-10 MALT1
BCL-10
TAK1
BCL-10
BCL-10 MAPKK7
BCL-10 JNK2

NF-B

KSR

Figure 2 | mitogen-activated protein kinase scaffold proteins in immune

signalling. a | The general mitogen-activated protein kinase (MAPK) signalling
cascade includes the activation of a MAPK kinase kinase (MAPKKK) that then
Nature Reviews | Immunology
activates a MAPKK, which activates the terminal MAPK. In the extracellular-signalregulated kinase (ERK) MAPK pathway, RAF acts as the MAPKKK, MAPK/ERK kinase 1
(MEK1) as the MAPKK and ERK as the MAPK. The scaffold protein kinase suppressor of
RAS (KSR) acts to assemble these components at the plasma membrane to enhance
ERK activation. b | In T cells, B-cell lymphoma 10 (BCL-10) oligomers form a scaffold
for the JUN N-terminal kinase (JNK) MAPK pathway, which assembles TGF-activated
kinase 1 (TAK1; a MAPKKK), MAPKK7 (a MAPKK) and JNK2 (a MAPK), thereby allowing
for signal specificity. CARMA1, CARDMAGUK protein 1; MALT1, mucosa-associatedlymphoid-tissue lymphoma-translocation gene 1; NF-B, nuclear factor-B; PKCq,
protein kinase Cq; TCR, T-cell receptor.

14-3-3 proteins
A family of conserved proteins
that is present in all eukaryotic
organisms and is involved in
diverse cellular processes, such
as apoptosis and stress, as well
as intracellular signalling and
cellcycle regulation. 1433
proteins function as scaffolds
in protein interactions and can
regulate protein localization
and enzymatic activity.
Approximately 100 binding
partners for the 1433
proteins have been reported.

KSR also binds 1433 proteins through two phosphorylation sites at serine 297 and serine 392 (reFs 29,30),
and this is thought to localize KSR to the cytoplasm.
Following cell stimulation, protein phosphatase 2A
(PP2A) dephosphorylates these serine residues and
releases KSR from 14-3-3 proteins in the cytoplasm, which
then allows KSR to bind to the cell membrane where it can
interact with activated RAS30. Because the localization of
KSR seems to be restricted to the plasma membrane during cell activation, it could have a role in the assembly of
components of the ERK pathway and in the localization of
activated ERK to the plasma membrane. Initiation of the
MAPK signalling cascade occurs following the activation
of KSR-associated RAF by the membrane-associated RAS.
It has been suggested that the scaffold protein connector enhancer of kinase suppressor of RAS 1 (CNKSR1)
facilitates the interaction between RAF and KSR31,32.
Additional proteins that interact with KSR include heatshock protein 90 (HSP90), HSP70, cell-division cycle 37,
PP2A and CDC25C-associated kinase 1 (reF. 25).
An interesting domain of KSR is its pseudokinase
domain. Pseudokinases, which make up ~10% of the
mammalian kinome33, are similar to bona fide kinases,
but lack one or more of the four highly conserved motifs
that are thought to confer catalytic activity and are therefore not thought to have kinase function. Recent studies, however, have shown that WNK and CASK (Ca2+/
calmodulin-dependent serine protein kinase), which are
two proteins that were originally thought to be pseudokinases, are in fact functional kinases34,35. It remains to be
determined whether KSR has kinase activity in addition
to its role as a scaffold protein.

Two isoforms of KSR are expressed by C. elegans

and mammals: KSR1 and KSR2 (reFs 36,37). Deletion
of both isoforms of KSR in C. elegans severely compromises the activation of the ERK signalling pathway
downstream of many receptors36. In mammals, KSR1
is highly expressed in the brain, thymus and spleen,
and analysis of T cells from KSR1-deficient mice confirmed that KSR1 is required for efficient ERK activation38. The defect in ERK activation that was observed
in KSR1-deficient mice resulted in attenuated cytokine
production and defective T-cell proliferation in the
periphery. Surprisingly, despite significant expression
of KSR1 in the thymus and a known role for ERK in
thymopoiesis39, thymocyte development seemed to be
normal in KSR1-deficient mice38.
KSR1 is also expressed by neutrophils and macrophages, and has been shown to be required for ERK activation in response to pro-inflammatory cytokines40. For
example, KSR1-deficient macrophages have a defect in
ERK activation following exposure to tumour-necrosis
factor (TNF), interleukin-1 (IL-1) or osmotic shock.
Interestingly, ERK activation downstream of proinflammatory stimuli is independent of RAF activation,
suggesting that KSR1 may function to link ERK with
the MAPKKKs that are normally associated with jNK
or p38 signalling. Indeed, KSR has been shown to bind
to the pro-inflammatory MAPKKK tumour-progression
locus 2 (TPL2)37. In addition, KSR1-deficient mice are
resistant to serum-induced arthritis40, suggesting a
physiological role for this interaction.
Other MAPK scaffold proteins. jNK is a stress-activated
MAPK that has an important role in the regulation of
lymphocyte activation and development41,42, and, along
with p38, is an important sensor of cellular stress, such as
that induced by pro-inflammatory cytokines, ultraviolet
radiation and osmotic shock. jNK is triggered by two
different MAPKKs, MAPKK4 and MAPKK7, which are
activated by various different MAPKKKs. jNK-interacting
proteins (jIPs) are thought to act as scaffolds in the jNK
signalling pathway, as they bind to all three components
of the jNK signalling cascade43,44. In the nervous system,
jIPs also bind to the kinesin microtubule motor proteins
and are therefore thought to link the kinase activation of
jNK with the transport of vesicles along the nerve axon45.
However, the function of these scaffold proteins in immunity has been largely unexplored, probably because they
are only weakly or not expressed by immune cells.
Another important scaffold protein for the jNK signalling pathway is the MAPKKK MEKK1, which can
activate the jNK, ERK and p38 pathways (reFs 18,46).
Because of its large size (196 kDa) and its ability to
bind jNK, MAPKK4 and MAPKK7, as well as the
upstream activating kinase NF-B-inducing kinase
(NIK), it has long been suspected that MEKK1 can
function as both a scaffold protein for the jNK pathway and as a MAPKKK 47. More recently, MEKK1
has been shown to have an important scaffold function downstream of the TNF family receptor CD40
(reF. 48). This study showed that the signalling complex
in the CD40-induced signalling cascade is scaffolded

50 | jANuARy 2009 | voLuME 9

www.nature.com/reviews/immunol
2009 Macmillan Publishers Limited. All rights reserved

REVIEWS
Caspase-recruitment
domain
A domain that is found in
certain initiator caspases
(for example, mammalian
caspase 9) and their adaptor
proteins (for example, APAF1).
This domain mediates
proteinprotein interactions.

Nuclear export signal

A short aminoacid sequence
of 56 hydrophobic residues
that targets a protein for
export from the cell nucleus
to the cytoplasm.

Leucine zipper
A common dimerization
domain found in some proteins
that are involved in regulating
gene expression. Leucine
zipper refers to the secondary
structure of two parallel
helices found in the protein.

by MEKK1 and contains MAPKK4 and MAPKK7, as

well as various other signalling molecules. The release
of this complex into the cytoplasm was shown to be
required for signal propagation and CD40-dependent
jNK activation46, providing evidence for a role for scaffold proteins in the localization of kinases to specific
cellular locations.
Recent data suggest that BCL-10, a protein involved
in nuclear factor-B (NF-B) activation, can also function as a jIP-like scaffold protein and is required for
the selective activation of jNK2 downstream of TCR
signalling49. BCL-10 contains an N-terminal caspase
recruitment domain (CARD) and a serine/threonine-rich
domain near the C-terminus. It functions in a complex
with mucosa-associated-lymphoid-tissue lymphomatranslocation gene 1 (MALT1) and CARDMAGuK
protein 1 (CARMA1; also known as CARD11) to link
the activation of protein kinase Cq (PKCq) with NF-B
downstream of antigen receptors50. In addition, the
scaffold function of BCL-10 can be regulated by TCR
signalling-induced oligomerization49. The formation
of BCL-10 oligomers allows it to bind to the MAPKKK
TGF-activated kinase 1 (TAK1), MAPKK7 and jNK2,
which are all components of the jNK signalling cascade
(FIG. 2). once activated, jNK2 releases the transcription
factor juN to which it is bound, allowing juN to accumulate and become activated by jNK1. Interestingly,
activation of jNK1 does not require BCL-10, suggesting
Calcium
channel

TCR
CD3

ZAP70

PtdIns(4,5)P2
PLC1
AHNAK
Ca2+

InsP3

Ca2+

Calmodulin

Ca2+
InsP3

Proper localization

Ca2+

Calcineurin

NFAT HOMER

STIM1

InsP3R

NFAT
Nucleus

Ca2+
NFAT

Ca2+
ER

Figure 3 | Scaffold proteins of the calcium signalling pathway. T-cell receptor

(TCR) stimulation leads to the generation of inositol-1,4,5-trisphosphate (InsP3), which
binds to the InsP3 receptor (InsP3R) on the endoplasmic reticulum (ER), thereby
releasing Ca2+ found in the ER. The release of the intracellular Ca2+ stores causes
calcium channels in the plasma membrane to open and allow an influx of Ca2+ from the
Nature
Reviews1| (STIM1).
Immunology
extracellular space through the action of stromal interaction
molecule
This calcium flux allows for the activation of calcineurin, which dephosphorylates
NFAT (nuclear factor of activated T cells), thereby allowing NFAT to translocate to
the nucleus and transcribe target genes. The scaffold protein AHNAK1 is thought
to function in mediating the proper localization of Ca2+ channels in the plasma
membrane. Further downstream, a HOMER family scaffold protein acts as a negative
regulator of NFAT activation by competing with calcineurin for NFAT binding. HOMER
has also been shown to bind InsP3R, which indicates that it might have a central role in
the regulation of calcium signalling. PtdIns(4,5)P2; phosphatidylinositol-4,5-bisphosphate;
ZAP70, -chain-associated protein kinase of 70 kDa.

that BCL-10 acts as a scaffold protein specifically for

jNK2 signalling49. However, it will be necessary to generate a mutant of BCL-10 that cannot oligomerize but can
still activate NF-B to determine the exact physiological
role of BCL-10 as a scaffold protein for jNK2 signalling.
It will also be important to determine whether BCL-10
localizes jNK activation to a specific area of the cell and
whether it coordinates positive and/or negative feedback
signalling pathways.

Calcium signalling scaffolds

AHNAK1. In addition to the MAPK signalling cascades,
calcium signalling is also integral for the proper functioning of immune cells. In T cells, the influx of calcium
is required for the nuclear translocation of the transcription factor NFAT (nuclear factor of activated T cells),
which is a crucial step in T-cell activation and proliferation51. The scaffold protein AHNAK (meaning giant in
Hebrew) is a large (700 kDa) protein with three structural regions: a C-terminal region that contains putative
regulatory elements such as a nuclear export signal and
a leucine zipper, a large central region that consists of
glycine/proline-rich repeats and an N-terminal domain
of unknown function. In non-immune cells, AHNAK
acts as a scaffolding protein in the calcium signalling pathway by binding calcium channels along with
phospholipase C (PLC) and PKC5254. As the entry of
calcium into the cytoplasm is rapidly quenched by high
expression levels of calcium-binding proteins, the ability to link calcium influx with two important calcium
effector molecules might have an important role in
controlling the signals that are induced by transient and
weak calcium influx54.
A recent study has shown that AHNAK1 is important
for efficient calcium signalling and NFAT activation in
T cells because it properly localizes calcium channels
at the plasma membrane55 (FIG. 3). Consistent with this,
T cells that were deficient for AHNAK1 had defective
calcium influx, which led to a defect in the nuclear translocation of NFAT and consequently resulted in decreased
T-cell proliferation and cytokine production in response
to TCR stimulation. Despite in vitro evidence indicating
that AHNAK1 can bind and activate PLC (reF. 56), PLC
activation, as well as calcium release from intracellular
stores, was normal in AHNAK1-deficient T cells. It will
be of interest to further define the role of AHNAK1 in
T cells and to determine whether there are additional
proteins (including PKC) that form a complex with the
AHNAK scaffold protein.
HOMeR. The HoMER family of scaffold proteins contains three members that exist in numerous isoforms. All
isoforms contain a similar domain structure that consists
of an EvH1 domain, which binds to proline-rich regions
in target proteins, and a coiled-coil domain, which is
involved in multimerization of HoMER proteins57.
Similarly to many other scaffold proteins, members of
the HoMER family were first described in neuronal
cells58, in which they function to aggregate receptors,
such as the metabotropic glutamate receptors, and signalling proteins at the neurological synapse. Interestingly,

NATuRE REvIEWS | Immunology

voLuME 9 | jANuARy 2009 | 51

2009 Macmillan Publishers Limited. All rights reserved

REVIEWS
HoMER proteins can also modulate calcium signalling
downstream of the glutamate receptor in neuronal cells
by linking it with inositol-1,4,5-trisphosphate receptors
in the endoplasmic reticulum59. A more recent study
extended the function of HoMER scaffold proteins in
calcium-signalling pathways to muscle differentiation60,
which indicates that they might have a role in nonneuronal tissues. Indeed, a recent study has defined a
role for HoMER scaffold proteins in T cells61.
In contrast to the positive regulatory role of HoMER
scaffold proteins in calcium signalling in neurons,
Huang et al.61 showed that HoMER2 and HoMER3
compete with calcineurin to bind to the N terminus
of NFAT in activated T cells. Interestingly, binding of
HoMER proteins to NFAT seemed to be attenuated by
CD28 co-stimulation and depended on the activation
of the serine/threonine kinase AKT. AKT might prevent the interaction of NFAT with HoMER proteins
by directly phosphorylating HoMER, although this
hypothesis needs to be confirmed experimentally. By
inhibiting the dephosphorylation of NFAT by calcineurin, HoMER proteins were shown to reduce NFAT
activation and thereby reduce the production of the
pro-survival cytokine IL-2 (reF. 61) (FIG. 3). Indeed,
HoMER-deficient T cells have increased NFAT activation and produce higher levels of IL-2. In addition,
HoMER-deficient mice have higher numbers of effector
memory T cells accumulating in the periphery than wildtype mice and develop slightly more severe autoimmunity in a pneumonitis-disease model61. This study
establishes that the HoMER scaffold proteins have a
role in immune signalling and that this role is distinct
from their role in neuronal cells. Further studies on the
function of HoMER proteins during immune responses
might identify new functions for these proteins and
potentially explain the differential function of HoMER
proteins in lymphoid and neuronal cells.

Effector memory T cell

A terminally differentiated
T cell that lacks lymphnode
homing receptors but
expresses receptors that
enable it to home to inflamed
tissues. effector memory cells
can exert immediate effector
functions without the need for
further differentiation.

Inflammasome
A molecular complex of several
proteins that, once assembled,
cleaves proIL1, thereby
producing active IL1.

Vitiligo
A depigmenting disorder of the
skin caused by the destruction
of melanocytes, which produce
cutaneous pigments.

Scaffold proteins in innate immune signalling

The Pellino family. Although numerous examples of
scaffold proteins in T-cell signalling exist, much less
is known about the role of scaffold proteins in signal
transduction in innate immune cells. one of the most
important signalling pathways in innate immune cells
is the Toll-like receptor (TLR) pathway. Initiation of
this signalling pathway involves the recruitment
of the adaptor protein myeloid differentiation primaryresponse gene 88 (MyD88) to the cytoplasmic domain
of the TLR following ligand binding. Subsequent activation of IL-1 receptor-associated kinase 4 (IRAK4)
and IRAK1 initiates the formation of a protein complex that includes TNFR-associated factor 6 (TRAF6),
TRAF-interacting protein with forkhead-associated
domain (TIFA), TAK1 and TAK1-binding protein 1
(TAB1). Activation of TAK1 by this complex leads
to the activation of p38, jNK and NF-B62. Although
the specific domain interactions between each of
these proteins have been described, it is unclear what
orchestrates the formation of the large multiprotein
complex that is necessary for TLR signalling. Data from
research studies over the past 5 years have implicated

Pellino proteins in this process, and evidence now

exists that they might function as scaffold proteins in
the TLR signalling pathway63.
The Pellino family is highly conserved throughout evolution and was originally identified in D. melanogaster as
an interacting partner of Pelle, which is an IRAK1 homologue64. Three Pellino proteins exist in humans and mice,
all of which contain a RING domain at the C terminus.
Although the function of Pellino proteins is still unclear,
they can associate with IRAK1, TRAF6 and TAK1 following IL-1 receptor (IL-1R) activation65,66. There are also data
indicating that some Pellino proteins can bind NIK67 and
BCL-10 (reF. 68), although the significance of this remains
to be determined, as NIK is not thought to be induced
downstream of IL-1R or TLRs. Because of these associations, we speculate that the Pellino proteins function as
scaffolds that assemble and localize crucial components of
the TLR signalling pathway close to the receptor. However,
these studies should be interpreted with caution, as most
of them have relied mainly on protein overexpression and
the function of scaffold proteins can change with their
concentration in the cell13. Additional studies using small
interfering RNA or knockout mice should provide more
information on the function of the Pellino proteins in TLR
signalling pathways.
The NLRP family. A second large and highly conserved
family of innate immune receptors is the nucleotidebinding domain and leucine-rich-repeat-containing
(NLR) family69. These cytoplasmic receptors sense various pathogenic and chemical stimuli, such as toxins,
bacterial DNA and uric acid crystals, and initiate the
formation of inflammasomes, which leads to the secretion
of pro-inflammatory cytokines, such as IL-1 and IL-18
(reF. 70). A subgroup of the NLR family is the NLRP
(NLR family, pyrin-domain-containing) family of receptors70, which function as scaffold proteins that mediate
the assembly of the inflammasome71. The importance
of these scaffold proteins in immunity is emphasized by
the association between mutations in the NLRP family
and human autoimmune diseases, including vitiligo72,73. A
member of the NLRP family (NLRP3) has also been suggested to be the sensor for aluminium adjuvants, which
are used in most human vaccines74.
The NLRP family consists of 14 proteins that all
contain four conserved domains: an N-terminal pyrin
domain, a NACHT domain (domain present in NAIP,
CIITA, HET-E and TP1) that promotes oligomerization,
a NACHT-associated domain (NAD) and a leucine-rich
repeat region that acts as the ligand sensor. In addition,
NLRP1 also contains a FIIND (function-to-find domain)
and CARD, making it the largest member of the NLRP
family. The NLRP scaffold proteins can bind adaptor
proteins and signalling proteins, such as ASC (apoptosis-associated speck-like protein containing a CARD),
HSP90, SuGT1 (suppressor of the G2 allele of skp1) and
caspase 5 (reFs 7577). Ligand binding by NLRP1 leads
to its association with the adaptor protein ASC, which
allows for the recruitment of pro-caspase 1. Association
of pro-caspase 1 with the NLRP1ASC complex leads
to pro-caspase 1 processing and the release of active

52 | jANuARy 2009 | voLuME 9

www.nature.com/reviews/immunol
2009 Macmillan Publishers Limited. All rights reserved

REVIEWS

PDZ-domain-containing scaffold proteins

Numerous parallels can be drawn between immune cells,
in particular lymphocytes and antigen-presenting cells,
and neurons, as all three cell types are highly polarized
and form complex structures known as synapses79,80.
In neurons, synapses are formed when neurons contact other neurons or muscle cells, and are the area
where activated neurons secrete neurotransmitters.
These neurotransmitters are detected by receptors in
the membrane of the receiving cell that are clustered
together by scaffold proteins which organize the receptors and downstream signalling molecules81. The most
important of these scaffold proteins in neurons are the
membrane-associated guanylate kinase (MAGuK) proteins82. MAGuK proteins are defined by the presence
of three functional domains: a PDZ (PSD95, DLGA
and Zo1 homology) domain, an SH3 domain and a
guanylate kinase domain at the C terminus (FIG. 4).
At the apex of their N termini, MAGuK proteins are
diverse and contain additional PDZ domains or other
domains, such as CARDs, WW domains (domain that
contains two conserved tryptophan residues) and kinase
domains. Members of the MAGuK family include
DLG1DLG4, zona occludens 1 (Zo1), CASK, CARMA
and membrane-associated guanylate kinase-, WW- and
PDZ-domain-containing 1 (MAGI1) protein.
PDZ domains are short (90100) amino-acid domains
that generally bind to specific sequences at the C termini
of transmembrane proteins. In the nervous system, the
PDZ domains of MAGuK proteins bind to glutamate
receptors, N-methyl-d-aspartic acid (NMDA) receptors and voltage-gated potassium channels83. The SH3
domain is thought to link MAGuK proteins with the
actin cytoskeleton and signalling proteins. Although guanylate kinase domains usually function to convert GMP to
GDP, it is unclear whether the guanylate kinase domains
of MAGuK proteins are enzymatically active or whether
they have some other, perhaps binding, function.
It is well known that MAGuK scaffold proteins have
an important role in the development and stability of
neuronal synapses84. Given that many of these proteins
are also expressed by immune cells, it is possible that
they might also be involved in regulating immunological synapses. The best-studied MAGuK protein in
T cells is DLG1.

DLG. Although D. melanogaster contains only a single Dlg

gene, mammals contain four genes, DLG1 (also known
as SAP97), DLG2 (also known as PSD93), DLG3 (also
known as SAP102) and DLG4 (also known as PSD95).
Studies of DLG proteins in immune cells have focused
on DLG1 because it is the most highly expressed DLG
protein in lymphocytes as determined by gene arrays (see
Genomic Institute of the Novartis Research Foundation
web site). Although DLG1 is expressed throughout
T-cell development, thymocyte maturation is normal
in DLG1-deficient mice85; however, it is important for
T-cell activation in the periphery. DLG1 is recruited to
the immunological synapse and is reported to bind the
-chain of the TCR and to -chain-associated protein
kinase 70 kDa (ZAP70), LCK, vAv1, Casitas B-lineage
lymphoma (CBL), WiskottAldrich syndrome protein
(WASP) and p38 (reFs 8689) (FIG. 4). These data suggest
that DLG1 may link the TCR signalling machinery with
regulators of the cytoskeleton, but more work is needed
to clarify the importance of these protein interactions.
Moreover, it is also unclear whether DLG1 acts as
a positive or negative regulator of T-cell activation. In
one study85, DLG1-deficient T cells had no defects in
T-cell polarization, migration and cytokine secretion,
and TCR-induced calcium flux and nuclear translocation of NFAT were normal. However, following TCR
engagement, DLG1-deficient T cells were hyperproliferative compared with wild-type T cells, which suggests
that endogenous DLG1 might act as a negative regulator of lymphocyte proliferation. The mechanism of
this negative regulation remains unclear; however, the
same study suggested that the negative regulatory effect
is mediated by a decrease in the expression of MyC,
which is an important transcriptional regulator that is
induced at the onset of T-cell proliferative responses.

TCR
CD3

GK
SH3
p38
PDZ
?
NFAT

ZAP70

caspase 1, which can then cleave the precursor proteins of

IL-1 and IL-18 (reF. 78). Interestingly, NLRP1 also binds
the anti-apoptotic molecules BCL-2 and BCL-XL75. These
binding events suppress the NLRP1-dependent activation
of caspase 1, which has been implicated in apoptosis in
addition to maturation of pro-inflammatory cytokines75.
This negative regulation of NLRP1 by anti-apoptotic
molecules indicates that cells have evolved complex pathways to allow for the coordination of host defence and
cell death. Although we have a basic understanding of
the role of NLRP proteins as scaffold proteins, a detailed
understanding of their exact function in regulating the
inflammasome has yet to be fully described, and it will
be important to determine whether they provide positive
or negative feedback.

DLG1

LCK
VAV1

WASP
Cytoskeletal
remodelling

Figure 4 | Dlg1 in T cells. Discs-large homologue 1

(DLG1) is a membrane-associated guanylate kinase
(MAGUK) family protein that contains three PDZ (PSD95,
Nature Reviews | Immunology
DLGA and ZO1 homology) domains, a SRC homology 3
(SH3) domain and a guanylate kinase (GK) domain. It
functions to assemble several T-cell receptor (TCR)-proximal
signalling molecules at the TCR, thereby linking TCR
activation with cytoskeletal reorganization through VAV1
and WiskottAldrich syndrome protein (WASP). DLG1 also
binds to the mitogen-activated protein kinase p38 and
might enhance its activation by the alternative activation
pathway in T cells. ZAP70, -chain-associated protein
kinase of 70 kDa. NFAT, nuclear factor of activated T cells.

NATuRE REvIEWS | Immunology

voLuME 9 | jANuARy 2009 | 53

2009 Macmillan Publishers Limited. All rights reserved

REVIEWS
Short hairpin RNA
One of the two most common
forms of short (~21 base pairs)
doublestranded rNAs that are
used for gene silencing. The
other form is small interfering
rNA.

Uropod
The posterior tail of migrating
amoeboid cells. It is rich in
filamentous actin, microtubules
and cytoskeletal adaptor
proteins (such as ezrin and
moesin), as well as adhesion
molecules (such as CD43 and
CD44) and lipid rafts.

Another study86 used short hairpin rNA that was specific

for DLG1 in jurkat cells and showed that NFAT activity was increased in cells that lacked DLG1. By contrast,
two studies by Miceli and co-workers88,89 used smallinterfering-RNA-mediated knockdown of DLG1 in
TCR-transgenic (oT-1) CD8+ T cells and found that
inhibition of DLG1 expression impaired both NFAT
activation and cytokine secretion. This finding was supported by the observation that overexpression of DLG1
in oT-1 CD8+ T cells resulted in enhanced NFAT activation89. Interestingly, the study by Miceli and co-workers
found that p38 was crucial for NFAT activation and that
DLG1 was required for activation of p38 (reF. 89) through
an alternative activation pathway that is thought to exist
in T cells and to involve LCK and/or ZAP70 (reF. 90). In
support of this, p38 was shown to associate with DLG1
and phosphorylation of p38 was decreased in DLG1knockdown T cells. Because DLG1 can bind LCK, ZAP70
and p38, these data suggest that DLG1 can act as a scaffold protein for the alternative activation of p38, resulting
in NFAT activation. In addition, other studies in T cells
suggest a positive role for p38 in NFAT activation9193.
The reasons for the disparate results on the role of
DLG1 in T-cell activation are not clear but could be due
to differences in the cell types (CD4+ or CD8+ T cells)
and/or the type of stimulation (antibody or antigen) that
was used in each study. It is possible that the phenotype

of DLG1-deficient T cells is caused by changes in the morphology or efficiency of the formation of the immunological synapse, as the immunological synapse can
function to enhance the downregulation of TCR expression on the cell surface94 and DLG1 is thought to be
involved in this process88. Nevertheless, these studies all
support the fact that the function of DLG1 in T cells is
complex, and a complete understanding of its function
may require different experimental approaches, including
mathematical modelling.
DLG proteins and the PDZ-domain-containing scaffold protein SCRIB have also been implicated in lymphocyte polarity95. DLG and SCRIB are initially recruited
to the immunological synapse and are subsequently
redistributed to the uropod of the T cell96. Knockdown
of SCRIB expression in T cells, or inhibition of DLG4
function, resulted in decreased uropod formation96. The
role of both proteins in T-cell polarity is an exciting and
emerging area of research that should provide us with a
better understanding of these MAGuK proteins in T-cell
activation in the future.
Spinophilin. Another PDZ-domain-containing scaffold
protein, spinophilin (encoded by PPP1R9B in humans),
is unrelated to the MAGuK proteins and has been implicated in dendritic cell (DC) function. Spinophilin was
first characterized as a modulator of PP1 (reF. 97) but

Table 1 | The properties of immune scaffold proteins

Scaffold
proteins

Domains

KSR

Proline-rich, cysteine-rich, serine/ RAF, MEK1, ERK, 14-3-3 proteins, HSP70, T cells, macrophages
threonine-rich and pseudokinase HSP90, CDC37, CTAK1 and PP2A
and neutrophils

Assembly and localization of the

ERK signalling pathway

MEKK1

Proline-rich and pleckstrin

homology

JNK, MAPKK4, MAPKK7, NIK, GSK3

and STAT3

B cells, T cells, DCs

and macrophages

Assembly and localization of the

death-receptor signalosome

BCL-10

CARD and serine/threonine-rich

MALT1, CARMA1, TAK1, MAPKK7 and

JNK2

B cells and T cells

Assembly and specificity of JNK2

signalling

AHNAK1

NES, leucine zipper and

glycine/proline-rich repeats

PLC, S100A and caveolin 1 channels

T cells

Assembly and localization of

calcium channels in the plasma
membrane

HOMER

EVH1 and coiled-coil

InsP3R, other HOMER proteins and

group 1 metabolic receptors

T cells

Inhibition of NFAT activation

Pellino

RING

IRAK1, IRAK4, TRAF6, TAK1, NIK,

BCL-10, MyD88 and SMAD6

Macrophages

Assembly of the TLR signalosome

NLRP

Pyrin, NACHT, NAD and

leucine-rich repeats

ASC, SUGT1, HSP90, caspase 5, BCL-2

and BCL-XL

Macrophages

Assembly of the inflammasome

DLG1

PDZ, SH3 and guanylate kinase

ZAP70, LCK, VAV1, CBL, WASP and p38

B cells and T cells

Assembly and localization of TCR

proximal signalling molecules;
alternative activation of p38

F-actin, PP1 and RGS proteins

DCs

Possibly assembly of DC
immunological-synapse proteins

Spinophilin F-actin binding, PP1-binding,

PDZ and coiled-coil

Binding partners

Expression pattern Potential functions

ASC, apoptosis-associated speck-like protein containing a CARD; BCL-10, B-cell lymphoma 10; CARD, caspase-recruitment domain; CARMA1, CARDMAGUK
protein 1; CBL, Casitas B-lineage lymphoma; CDC37, cell-division cycle 37; CTAK1, CDC25C-associated kinase 1; DC, dendritic cell; DLG1, discs-large homologue 1;
ERK, extracellular-signal-regulated kinase; GSK3, glycogen-synthase kinase 3; HSP, heat-shock protein; InsP3R, inositol-1,4,5-trisphosphate receptor; IRAK,
interleukin-1-receptor-associated kinase; JNK, JUN N-terminal kinase; KSR, kinase suppressor of RAS; MALT1, mucosa-associated-lymphoid-tissue lymphoma-translocation
gene 1; MAPKK, mitogen-activated protein kinase kinase; MEK1, MAPK/ERK kinase 1; MEKK1, MEK kinase 1; MyD88, myeloid differentiation primary-response gene 88;
NACHT, domain present in NAIP, CIITA, HET-E and TP1; NAD, NACHT-associated domain; NES, nuclear export signal; NFAT, nuclear factor of activated T cells; NF-B,
nuclear factor-B; NIK, NF-B-inducing kinase; NLR, nucleotide-binding domain and leucine-rich-repeat-containing receptor; NLRP, NLR family, pyrin-domain-containing; PDZ, PSD95, DLGA and ZO1 homology; PLC, phospholipase C; PP, protein phosphatase; RSG, regulator of G-protein signalling; S100A, S100 calcium-binding
protein A; SH3, SRC homology 3; SMAD6, mothers against decapentaplegic homologue 6; STAT3, signal transducer and activator of transcription 3; SUGT1, suppressor
of the G2 allele of skp1; TAK1, TGF-activated kinase; TCR, T-cell receptor; TGF, transforming growth factor-; TLR, Toll-like receptor; TNFR, tumour-necrosis factor
receptor; TRAF6, TNFR-associated factor 6; WASP, WiskottAldrich syndrome protein; ZAP70, -chain-associated protein kinase of 70 kDa.

54 | jANuARy 2009 | voLuME 9

www.nature.com/reviews/immunol
2009 Macmillan Publishers Limited. All rights reserved

REVIEWS
has also been shown to bind F-actin10. This interaction with F-actin may allow spinophilin to regulate the
actin cytoskeleton and may help target spinophilin to
areas that are enriched with actin, such as the immunological synapse98. In addition to its F-actin- and
PP1-binding domains, spinophilin contains coiled-coil
and PDZ domains, as well as a domain that binds to
seven-transmembrane-spanning receptors, such as the
D2-dopamine and 2-adrenergic receptors99.
Spinophilin is thought to have a role in the formation of immunological synapses in DCs100. During DC
maturation, spinophilin moves from the cytoplasm to
the plasma membrane and dendrites, and is recruited
to the immunological synapse following contact with
a T cell. This recruitment seems to have functional
consequences, as spinophilin-deficient DCs cannot
activate naive antigen-specific CD4+ T cells in vitro or
in vivo100. Although this study provides intriguing evidence for a role for spinophilin in antigen presentation by
DCs, more work is needed to elucidate a mechanism
by which spinophilin facilitates antigen presentation, as
it does not seem to affect antigen uptake or the expression of MHC molecules by DCs100. It is possible that
spinophilin is involved in regulating the duration of
cell contact through the immunological synapse or

1.

2.

3.

4.
5.
6.
7.
8.
9.
10.
11.

12.
13.
14.
15.

Cooper, J. A., Bowen-Pope, D. F., Raines, E., Ross, R.

& Hunter, T. Similar effects of platelet-derived
growth factor and epidermal growth factor on the
phosphorylation of tyrosine in cellular proteins. Cell
31, 263273 (1982).
Wang, Y., Pennock, S., Chen, X. & Wang, Z.
Internalization of inactive EGF receptor into
endosomes and the subsequent activation of
endosome-associated EGF receptors. Epidermal
growth factor. Sci. STKE 2002, PL17 (2002).
Wang, Y., Pennock, S. D., Chen, X., Kazlauskas, A. &
Wang, Z. Platelet-derived growth factor receptormediated signal transduction from endosomes.
J. Biol. Chem. 279, 80388046 (2004).
Kane, L. P., Lin, J. & Weiss, A. Signal transduction by
the TCR for antigen. Curr. Opin. Immunol. 12,
242249 (2000).
Takeda, K., Kaisho, T. & Akira, S. Toll-like receptors.
Annu. Rev. Immunol. 21, 335376 (2003).
Aggarwal, B. B. Signalling pathways of the TNF
superfamily: a double-edged sword. Nature Rev.
Immunol. 3, 745756 (2003).
Aguado, E., Martinez-Florensa, M. & Aparicio, P.
Activation of T lymphocytes and the role of the
adapter LAT. Transpl. Immunol. 17, 2326 (2006).
Wu, J. N. & Koretzky, G. A. The SLP-76 family of
adapter proteins. Semin. Immunol. 16, 379393
(2004).
Burack, W. R., Cheng, A. M. & Shaw, A. S. Scaffolds,
adaptors and linkers of TCR signaling: theory and
practice. Curr. Opin. Immunol. 14, 312316 (2002).
Nakanishi, H. et al. Neurabin: a novel neural tissuespecific actin filament-binding protein involved in
neurite formation. J. Cell Biol. 139, 951961 (1997).
Levchenko, A., Bruck, J. & Sternberg, P. W. Scaffold
proteins may biphasically affect the levels of mitogenactivated protein kinase signaling and reduce its
threshold properties. Proc. Natl Acad. Sci. USA 97,
58185823 (2000).
Ferrell, J. E. Jr. What do scaffold proteins really do?
Sci. STKE 2000, PE1 (2000).
Burack, W. R. & Shaw, A. S. Signal transduction:
hanging on a scaffold. Curr. Opin. Cell Biol. 12,
211216 (2000).
Wong, W. & Scott, J. D. AKAP signalling complexes:
focal points in space and time. Nature Rev. Mol. Cell
Biol. 5, 959970 (2004).
Williams, R. O. Cutting Edge: A-kinase anchor proteins
are involved in maintaining resting T cells in an
inactive state. J. Immunol. 168, 53925396 (2002).

in regulating the recycling of MHC molecules or other

co-stimulatory molecules in the cell. Spinophilin is also
expressed by B and T cells, and it will be of interest to
determine whether it has a role in immunological-synapse
formation in these cells.

Concluding remarks
Although mathematical studies and engineered scaffold proteins have implicated scaffold proteins in the
regulation of immune signalling responses, the specific
mechanisms that are involved are still being investigated
(TABLe 1). Most of the scaffold proteins that are crucial
for immune-cell signalling have probably not been
identified. At this point, investigations of scaffold proteins have focused on protein interactions and on their
intracellular localization. It is probable that the function
and regulation of scaffold proteins are complex, and a
detailed understanding of their biology will require the
use of new experimental tools, such as computational
modelling and single-cell analysis, in addition to classic biochemical approaches. Although neuroscientists
and cell biologists might have initiated the field of scaffold biology, immunologists are now poised to reveal
additional insights into the function and importance of
scaffold proteins.

16. Locasale, J. W., Shaw, A. S. & Chakraborty, A. K.

Scaffold proteins confer diverse regulatory properties
to protein kinase cascades. Proc. Natl Acad. Sci. USA
104, 1330713312 (2007).
This study uses mathematical modelling to examine
possible functions of scaffold proteins and identifies
important variables that might alter these functions.
17. Pincet, F. Membrane recruitment of scaffold proteins
drives specific signaling. PLoS ONE 2, e977 (2007).
18. Uhlik, M. T., Abell, A. N., Cuevas, B. D., Nakamura, K.
& Johnson, G. L. Wiring diagrams of MAPK regulation
by MEKK1, 2, and 3. Biochem. Cell Biol. 82,
658663 (2004).
19. Bashor, C. J., Helman, N. C., Yan, S. & Lim, W. A.
Using engineered scaffold interactions to reshape
MAP kinase pathway signaling dynamics. Science
319, 15391543 (2008).
Using engineered scaffold proteins to recruit
pathway modulators to a signalling cascade, this
study identifies a range of signalling behaviours
that can be induced by scaffold proteins.
20. Park, S. H., Zarrinpar, A. & Lim, W. A. Rewiring MAP
kinase pathways using alternative scaffold assembly
mechanisms. Science 299, 10611064 (2003).
21. Dueber, J. E., Mirsky, E. A. & Lim, W. A. Engineering
synthetic signaling proteins with ultrasensitive input/
output control. Nature Biotechnol. 25, 660662
(2007).
22. Lewis, R. S. Calcium signaling mechanisms in
T lymphocytes. Annu. Rev. Immunol. 19,
497521 (2001).
23. Kolch, W. Coordinating ERK/MAPK signalling through
scaffolds and inhibitors. Nature Rev. Mol. Cell Biol. 6,
827837 (2005).
24. Elion, E. A. The Ste5p scaffold. J. Cell Sci. 114,
39673978 (2001).
25. Claperon, A. & Therrien, M. KSR and CNK: two
scaffolds regulating RAS-mediated RAF activation.
Oncogene 26, 31433158 (2007).
26. Therrien, M. et al. KSR, a novel protein kinase
required for RAS signal transduction. Cell 83,
879888 (1995).
27. Kornfeld, K., Hom, D. B. & Horvitz, H. R. The ksr-1
gene encodes a novel protein kinase involved in Rasmediated signaling in C. elegans. Cell 83, 903913
(1995).
28. Sundaram, M. & Han, M. The C. elegans ksr-1 gene
encodes a novel Raf-related kinase involved in Rasmediated signal transduction. Cell 83, 889901
(1995).

NATuRE REvIEWS | Immunology

29. Cacace, A. M. et al. Identification of constitutive and

Ras-inducible phosphorylation sites of KSR:
implications for 14-3-3 binding, mitogen-activated
protein kinase binding, and KSR overexpression.
Mol. Cell. Biol. 19, 229240 (1999).
30. Muller, J., Ory, S., Copeland, T., Piwnica-Worms, H. &
Morrison, D. K. C-TAK1 regulates Ras signaling by
phosphorylating the MAPK scaffold, KSR1. Mol. Cell
8, 983993 (2001).
31. Douziech, M., Sahmi, M., Laberge, G. & Therrien, M.
A KSR/CNK complex mediated by HYP, a novel SAM
domain-containing protein, regulates RAS-dependent
RAF activation in Drosophila. Genes Dev. 20,
807819 (2006).
32. Therrien, M., Wong, A. M. & Rubin, G. M. CNK, a
RAF-binding multidomain protein required for RAS
signaling. Cell 95, 343353 (1998).
33. Boudeau, J., Miranda-Saavedra, D., Barton, G. J. &
Alessi, D. R. Emerging roles of pseudokinases.
Trends Cell Biol. 16, 443452 (2006).
34. Huang, C. L., Cha, S. K., Wang, H. R., Xie, J. & Cobb,
M. H. WNKs: protein kinases with a unique kinase
domain. Exp. Mol. Med. 39, 565573 (2007).
35. Mukherjee, K. et al. CASK functions as a Mg2+
independent neurexin kinase. Cell 133, 328339
(2008).
36. Ohmachi, M. et al. C. elegans ksr-1 and ksr-2 have
both unique and redundant functions and are required
for MPK-1 ERK phosphorylation. Curr. Biol. 12,
427433 (2002).
37. Channavajhala, P. L. et al. Identification of a novel
human kinase supporter of Ras (hKSR-2) that
functions as a negative regulator of Cot (Tpl2)
signaling. J. Biol. Chem. 278, 4708947097
(2003).
38. Nguyen, A. et al. Kinase suppressor of Ras (KSR) is a
scaffold which facilitates mitogen-activated protein
kinase activation in vivo. Mol. Cell. Biol. 22,
30353045 (2002).
39. Fischer, A. M., Katayama, C. D., Pages, G., Pouyssegur,
J. & Hedrick, S. M. The role of Erk1
and Erk2 in multiple stages of T cell development.
Immunity 23, 431443 (2005).
40. Fusello, A. M. et al. The MAPK scaffold kinase
suppressor of Ras is involved in ERK activation by
stress and proinflammatory cytokines and induction
of arthritis. J. Immunol. 177, 61526158 (2006).
41. Rincon, M. & Pedraza-Alva, G. JNK and p38 MAP
kinases in CD4+ and CD8+ T cells. Immunol. Rev. 192,
131142 (2003).

voLuME 9 | jANuARy 2009 | 55

2009 Macmillan Publishers Limited. All rights reserved

REVIEWS
42. Dong, C., Davis, R. J. & Flavell, R. A. MAP kinases
in the immune response. Annu. Rev. Immunol. 20,
5572 (2002).
43. Yasuda, J., Whitmarsh, A. J., Cavanagh, J., Sharma, M.
& Davis, R. J. The JIP group of mitogen-activated
protein kinase scaffold proteins. Mol. Cell. Biol. 19,
72457254 (1999).
44. Willoughby, E. A., Perkins, G. R., Collins, M. K. &
Whitmarsh, A. J. The JNK-interacting protein-1
scaffold protein targets MAPK phosphatase-7
to dephosphorylate JNK. J. Biol. Chem. 278,
1073110736 (2003).
45. Verhey, K. J. et al. Cargo of kinesin identified as JIP
scaffolding proteins and associated signaling
molecules. J. Cell Biol. 152, 959970 (2001).
46. Gallagher, E. et al. Kinase MEKK1 is required for
CD40-dependent activation of the kinases Jnk and
p38, germinal center formation, B cell proliferation
and antibody production. Nature Immunol. 8, 5763
(2007).
47. Su, Y. C., Han, J., Xu, S., Cobb, M. & Skolnik, E. Y.
NIK is a new Ste20-related kinase that binds NCK and
MEKK1 and activates the SAPK/JNK cascade via a
conserved regulatory domain. EMBO J. 16,
12791290 (1997).
48. Matsuzawa, A. et al. Essential cytoplasmic
translocation of a cytokine receptor-assembled
signaling complex. Science 321, 663668 (2008).
This paper proposes a two-step mechanism for
signalling by CD40 in which a CD40-associated
signalling complex that is scaffolded by MEKK1
must be released from the receptor into the
cytoplasm for signal propagation.
49. Blonska, M. et al. The CARMA1Bcl10 signaling
complex selectively regulates JNK2 kinase in the T cell
receptor-signaling pathway. Immunity 26, 5566
(2007).
50. Lin, X. & Wang, D. The roles of CARMA1, Bcl10, and
MALT1 in antigen receptor signaling. Semin. Immunol.
16, 429435 (2004).
51. Feske, S. Calcium signalling in lymphocyte activation
and disease. Nature Rev. Immunol. 7, 690702
(2007).
52. Haase, H. et al. Ahnak is critical for cardiac Ca(v)1.2
calcium channel function and its -adrenergic
regulation. FASEB J. 19, 19691977 (2005).
53. Lee, I. H. et al. Ahnak protein activates protein kinase
C (PKC) through dissociation of the PKC-protein
phosphatase 2A complex. J. Biol. Chem. 283,
63126320 (2008).
54. Lee, I. H. et al. AHNAK-mediated activation of
phospholipase C-1 through protein kinase C.
J. Biol. Chem. 279, 2664526653 (2004).
55. Matza, D. et al. A scaffold protein, AHNAK1, is
required for calcium signaling during T cell activation.
Immunity 28, 6474 (2008).
This study establishes a role for AHNAK1 in
calcium signalling in T cells.
56. Sekiya, F., Bae, Y. S., Jhon, D. Y., Hwang, S. C. &
Rhee, S. G. AHNAK, a protein that binds and
activates phospholipase C-1 in the presence of
arachidonic acid. J. Biol. Chem. 274, 1390013907
(1999).
57. Shiraishi-Yamaguchi, Y. & Furuichi, T. The Homer
family proteins. Genome Biol. 8, 206 (2007).
58. Thomas, U. Modulation of synaptic signalling
complexes by Homer proteins. J. Neurochem.
81, 407413 (2002).
59. Xiao, B., Tu, J. C. & Worley, P. F. Homer: a link between
neural activity and glutamate receptor function.
Curr. Opin. Neurobiol. 10, 370374 (2000).
60. Stiber, J. A. et al. Homer modulates NFAT-dependent
signaling during muscle differentiation. Dev. Biol. 287,
213224 (2005).
61. Huang, G. N. et al. NFAT binding and regulation of
T cell activation by the cytoplasmic scaffolding Homer
proteins. Science 319, 476481 (2008).
This is the first study to show a role for HoMEr
proteins in calcium signalling in T cells.
62. Ninomiya-Tsuji, J. et al. The kinase TAK1 can activate
the NIK-IB as well as the MAP kinase cascade in the

63.
64.

65.

66.
67.

68.
69.
70.

71.

72.

73.
74.

75.

76.

77.

78.
79.
80.
81.
82.
83.

IL-1 signalling pathway. Nature 398, 252256

(1999).
Schauvliege, R., Janssens, S. & Beyaert, R. Pellino
proteins: novel players in TLR and IL-1R signalling.
J. Cell. Mol. Med. 11, 453461 (2007).
Grosshans, J., Schnorrer, F. & Nusslein-Volhard, C.
Oligomerisation of Tube and Pelle leads to nuclear
localisation of dorsal. Mech. Dev. 81, 127138
(1999).
Jiang, Z. et al. Pellino 1 is required for interleukin-1
(IL-1)-mediated signaling through its interaction with
the IL-1 receptor-associated kinase 4 (IRAK4)IRAK
tumor necrosis factor receptor-associated factor 6
(TRAF6) complex. J. Biol. Chem. 278, 1095210956
(2003).
Yu, K. Y. et al. Cutting Edge: mouse pellino-2
modulates IL-1 and lipopolysaccharide signaling.
J. Immunol. 169, 40754078 (2002).
Jensen, L. E. & Whitehead, A. S. Pellino3, a novel
member of the Pellino protein family, promotes
activation of c-Jun and Elk-1 and may act as a
scaffolding protein. J. Immunol. 171, 15001506
(2003).
Liu, Y. et al. BCL10 mediates lipopolysaccharide/
toll-like receptor-4 signaling through interaction with
Pellino2. J. Biol. Chem. 279, 3743637444 (2004).
Ting, J. P., Willingham, S. B. & Bergstralh, D. T.
NLRs at the intersection of cell death and immunity.
Nature Rev. Immunol. 8, 372379 (2008).
Martinon, F., Gaide, O., Petrilli, V., Mayor, A. &
Tschopp, J. NALP inflammasomes: a central role in
innate immunity. Semin. Immunopathol. 29, 213229
(2007).
Petrilli, V., Dostert, C., Muruve, D. A. & Tschopp, J.
The inflammasome: a danger sensing complex
triggering innate immunity. Curr. Opin. Immunol. 19,
615622 (2007).
Aganna, E. et al. Association of mutations in the
NALP3/CIAS1/PYPAF1 gene with a broad phenotype
including recurrent fever, cold sensitivity,
sensorineural deafness, and AA amyloidosis.
Arthritis Rheum. 46, 24452452 (2002).
Jin, Y. et al. NALP1 in vitiligo-associated multiple
autoimmune disease. N. Engl. J. Med. 356,
12161225 (2007).
Eisenbarth, S. C., Colegio, O. R., OConnor, W.,
Sutterwala, F. S. & Flavell, R. A. Crucial role for the
Nalp3 inflammasome in the immunostimulatory
properties of aluminium adjuvants. Nature 453,
11221126 (2008).
Bruey, J. M. et al. Bcl-2 and Bcl-XL regulate
proinflammatory caspase-1 activation by interaction
with NALP1. Cell 129, 4556 (2007).
This interesting study connects the NLrP1
inflammasome with the cell-death pathway and
shows that anti-apoptotic molecules can inhibit the
function of NLrP1.
Mayor, A., Martinon, F., De Smedt, T., Petrilli, V. &
Tschopp, J. A crucial function of SGT1 and HSP90 in
inflammasome activity links mammalian and plant
innate immune responses. Nature Immunol. 8,
497503 (2007).
Martinon, F., Burns, K. & Tschopp, J. The
inflammasome: a molecular platform triggering
activation of inflammatory caspases and processing
of proIL-. Mol. Cell 10, 417426 (2002).
Faustin, B. et al. Reconstituted NALP1 inflammasome
reveals two-step mechanism of caspase-1 activation.
Mol. Cell 25, 713724 (2007).
Hall, Z. W. & Sanes, J. R. Synaptic structure and
development: the neuromuscular junction. Cell 72,
S99S121 (1993).
Cemerski, S. & Shaw, A. Immune synapses in T-cell
activation. Curr. Opin. Immunol. 18, 298304 (2006).
Kennedy, M. B. Signal-processing machines at the
postsynaptic density. Science 290, 750754 (2000).
Gonzalez-Mariscal, L., Betanzos, A. & Avila-Flores, A.
MAGUK proteins: structure and role in the tight
junction. Semin. Cell Dev. Biol. 11, 315324 (2000).
Kim, E. & Sheng, M. PDZ domain proteins of
synapses. Nature Rev. Neurosci. 5, 771781 (2004).

56 | jANuARy 2009 | voLuME 9

84. Montgomery, J. M., Zamorano, P. L. & Garner, C. C.

MAGUKs in synapse assembly and function: an
emerging view. Cell. Mol. Life Sci. 61, 911929 (2004).
85. Stephenson, L. M. et al. DLGH1 is a negative
regulator of T-lymphocyte proliferation. Mol. Cell. Biol.
27, 75747581 (2007).
86. Xavier, R. et al. Discs large (Dlg1) complexes in
lymphocyte activation. J. Cell Biol. 166, 173178
(2004).
87. Hanada, T., Lin, L., Chandy, K. G., Oh, S. S. &
Chishti, A. H. Human homologue of the Drosophila
discs large tumor suppressor binds to p56lck tyrosine
kinase and Shaker type Kv1.3 potassium channel in
T lymphocytes. J. Biol. Chem. 272, 2689926904
(1997).
88. Round, J. L. et al. Dlgh1 coordinates actin
polymerization, synaptic T cell receptor and lipid raft
aggregation, and effector function in T cells. J. Exp.
Med. 201, 419430 (2005).
89. Round, J. L. et al. Scaffold protein Dlgh1 coordinates
alternative p38 kinase activation, directing T cell
receptor signals toward NFAT but not NF-B
transcription factors. Nature Immunol. 8, 154161
(2007).
90. Ashwell, J. D. The many paths to p38 mitogenactivated protein kinase activation in the immune
system. Nature Rev. Immunol. 6, 532540 (2006).
91. Yang, T. T., Xiong, Q., Enslen, H., Davis, R. J. &
Chow, C. W. Phosphorylation of NFATc4 by p38
mitogen-activated protein kinases. Mol. Cell. Biol.
22, 38923904 (2002).
92. Gomez del Arco, P., Martinez-Martinez, S.,
Maldonado, J. L., Ortega-Perez, I. & Redondo, J. M.
A role for the p38 MAP kinase pathway in the nuclear
shuttling of NFATp. J. Biol. Chem. 275,
1387213878 (2000).
93. Wu, C. C., Hsu, S. C., Shih, H. M. & Lai, M. Z. Nuclear
factor of activated T cells c is a target of p38 mitogenactivated protein kinase in T cells. Mol. Cell. Biol. 23,
64426454 (2003).
94. Cemerski, S. et al. The stimulatory potency of T cell
antigens is influenced by the formation of the
immunological synapse. Immunity 26, 345355
(2007).
95. Krummel, M. F. & Macara, I. Maintenance and
modulation of T cell polarity. Nature Immunol. 7,
11431149 (2006).
96. Ludford-Menting, M. J. et al. A network of PDZcontaining proteins regulates T cell polarity and
morphology during migration and immunological
synapse formation. Immunity 22, 737748 (2005).
97. Allen, P. B., Ouimet, C. C. & Greengard, P. Spinophilin,
a novel protein phosphatase 1 binding protein
localized to dendritic spines. Proc. Natl Acad. Sci. USA
94, 99569961 (1997).
98. Zito, K., Knott, G., Shepherd, G. M., Shenolikar, S. &
Svoboda, K. Induction of spine growth and synapse
formation by regulation of the spine actin
cytoskeleton. Neuron 44, 321334 (2004).
99. Sarrouilhe, D., di Tommaso, A., Metaye, T. &
Ladeveze, V. Spinophilin: from partners to functions.
Biochimie 88, 10991113 (2006).
100. Bloom, O. et al. Spinophilin participates in information
transfer at immunological synapses. J. Cell Biol. 181,
203211 (2008).

DATABASES
Entrez Gene: http://www.ncbi.nlm.nih.gov/entrez/query.
fcgi?db=gene
AHNAK | BCL-10 | DLG1 | ERK | HOMER2 | HOMER3 | JNK | KSR |
MAPKK4 | MAPKK7 | MEK1 | MEKK1 | p38 | Pellino | PPP1R9B |
SCRIB

FURTHER INFORMATION
Andrey S. Shaws homepage: http://pathbox.wustl.
edu/~shawlab
Genomic Institute of the Novartis Research Foundation:
http://symatlas.gnf.org/SymAtlas
All lInkS ArE AcTIvE In ThE onlInE PDf

www.nature.com/reviews/immunol
2009 Macmillan Publishers Limited. All rights reserved