MOLECULAR AND CELLULAR BIOLOGY, Apr. 2009, p. 2082–2091
Copyright © 2009, American Society for Microbiology. All Rights Reserved.
Vol. 29, No. 8
KSR1 Modulates the Sensitivity of Mitogen-Activated Protein Kinase
Pathway Activation in T Cells without Altering
Fundamental System Outputs?
Joseph Lin,1*† Angus Harding,2*† Emanuele Giurisato,1and Andrey S. Shaw1
Department of Pathology and Immunology, Howard Hughes Medical Institute, Washington University School of Medicine,
660 S. Euclid, Box 8118, St. Louis, Missouri 63110,1and Queensland Brain Institute, University of Queensland,
QBI Building 79, St. Lucia, Queensland 4072, Australia2
Received 20 October 2008/Returned for modification 2 December 2008/Accepted 22 January 2009
Mitogen-activated protein kinase (MAPK) cascades are evolutionarily conserved signaling pathways that
regulate cell fate decisions. They generate a wide range of signal outputs, including graded and digital
responses. In T cells, MAPK activation is digital in response to T-cell-receptor stimulation; however, whether
other receptors on T cells that lead to MAPK activation are graded or digital is unknown. Here we evaluate
MAPK activation in T cells at the single-cell level. We show that T cells responded digitally to stimulation with
superantigen-loaded antigen-presenting cells, whereas they responded in a graded manner to the chemokine
SDF-1, demonstrating that the system output of the MAPK module is highly plastic and determined by
components upstream of the MAPK module. These findings also confirm that different MAPK system outputs
are used by T cells to control discrete biological functions. Scaffold proteins are essential for proper MAPK
signaling and function as they physically assemble multiple components and regulators of MAPK cascades. We
found that the scaffold protein KSR1 regulated the threshold required for MAPK activation in T cells without
affecting the nature of the response. We conclude that KSR1 plays a central role in determining the sensitivity
of T-cell responses and is thus well positioned as a key control point.
T cells are a central component of the adaptive immune
response. They are activated through the recognition of ligands
by their T-cell receptors (TCR) leading to proliferation; how-
ever, T cells also have a multitude of other receptors that
mediate different functions, including chemotaxis and dif-
ferentiation. One of the essential signaling pathways down-
stream of not only the TCR but also other receptors on the
surface of T cells is the canonical mitogen-activated protein
kinase (MAPK) cascade, composed of Raf (MAP3K), MEK
(MAP2K), and ERK (MAPK) (22). When activated, MAPK
modules can generate both graded and digital outputs in vivo
(11, 14) (Fig. 1). In a graded system, the pathway transmits
continuous information that is proportional to the input stim-
ulus. In contrast, the all-or-none digital output can switch be-
tween two steady states but cannot rest in intermediate states,
thereby functioning as a digital switch with the low and high
steady states representing “off” and “on,” respectively. These
different signal outputs can be used to drive discrete cell fate
decisions within a single cell, a principle that is strikingly illus-
trated in PC12 cells, in which a graded MAPK output drives
cells to proliferate, whereas a digital MAPK output directs
The MAPK pathway is thought to regulate both positive and
negative selection during T-cell development, with activation
of the MAPK module from the Golgi correlating with positive
selection and activation from the plasma membrane with neg-
ative selection (7). Recent results have demonstrated that ac-
tivation of the MAPK module from the Golgi generates a
graded output, whereas ERK activation from the plasma mem-
brane is digital (16). In addition, mature T cells use a digital
MAPK signal output downstream of TCR engagement to
commit to T-cell activation (2). Taken together, these re-
sults suggest that T cells utilize different system outputs
from the MAPK module to regulate multiple biological re-
sponses in vivo. However, whether T cells can generate
multiple outputs from the MAPK cascade has not been
It is currently unclear how diverse MAPK system dynamics,
such as graded and digital outputs, as well as different sensi-
tivity and durations of signaling, are generated within the same
cell. The use of scaffold proteins as signal-processing hubs may
provide a solution to this question. Scaffold proteins act as
docking platforms that bind to two or more components of the
MAPK module together in a protein complex (4). There are at
least two ways scaffolds can modulate the system output of
MAPK cascades. First, scaffolds could set the sensitivity of the
system by bringing the three kinases of the MAPK module into
close proximity to increase the efficiency of signal transfer, a
hypothesis supported by in silico modeling studies (24, 25).
Second, scaffolds could change the fundamental system output
of the MAPK module. In an elegant series of experiments, Lim
and coworkers engineered scaffold-specific feedback loops by
regulating recruitment of positive and negative regulators to
the yeast MAPK scaffold protein Ste5 (3). The resulting syn-
* Corresponding author. Mailing address for Joseph Lin: Depart-
ment of Pathology and Immunology, Howard Hughes Medical Insti-
tute, Washington University School of Medicine, 660 S. Euclid, Box
8118, St. Louis, MO 63110. Phone: (314) 362-4601. Fax: (314) 362-
8888. E-mail: firstname.lastname@example.org. Mailing address for Angus
Harding: Queensland Brain Institute, University of Queensland, QBI
Building 79, St. Lucia, Queensland 4072, Australia. Phone: (61) 7-3346-
6300. Fax: (61) 7-3346-6301. E-mail: email@example.com.
† J.L. and A.H. contributed equally to this study.
?Published ahead of print on 2 February 2009.
thetic circuits yielded diverse MAPK outputs, providing the
first critical proof-of-principle experiments that scaffold pro-
teins themselves can be used to rewire MAPK modules to
generate different outputs (3). Recent data revealed that the
yeast scaffold protein Ste5 converts the inherent switch-like
signal output of the yeast mating MAPK module into a graded
output, confirming that scaffolds are indeed used to modulate
MAPK system output in vivo (42). Since Ste5 is expressed only
in yeast, it is important determine whether mammalian MAPK
scaffolds perform a similar role given the lack of homology
between Ste5 and mammalian scaffold proteins.
The best-characterized mammalian MAPK scaffold protein
is Kinase Suppressor of Ras (KSR). It binds to Raf, MEK, and
ERK to facilitate ERK activation at the plasma membrane (31,
34). Genetic and biochemical studies in nematodes, flies, and
mammals confirm that KSR1 is essential for proper MAPK
signal transmission in vivo (34). KSR1 binds to protein phos-
phatase 2A (17, 32) and casein kinase 2 (35), positive regula-
tors of Raf activity (17, 32, 35). KSR1 is also regulated by a
positive feed-forward loop from Ras through IMP (28). Thus,
KSR1 coordinates multiple MAPK positive regulatory loops,
placing KSR1 in a prime position to regulate MAPK system
sensitivity, output, or both.
We examined here activation of the MAPK cascade in single
T cells. We show that engagement of TCR with superantigen
(SAg) results in a digital ERK response, whereas stimulation
through a G-protein coupled receptor (GPCR), CXCR4, gen-
erates a graded ERK output, formally demonstrating that T
cells generate multiple system outputs from the MAPK mod-
ule. We then show that the MAPK scaffold KSR1 does not
rewire the MAPK pathway to generate digital or graded out-
puts. Instead, the primary function of KSR1 is to modulate
MAPK system sensitivity. Finally, we demonstrate that KSR1
protein levels are regulated during T-cell activation, revealing
KSR1 as a likely control point for T-cell responsiveness. These
findings have important implications for our understanding of
T-cell regulation by the MAPK pathway in vivo.
MATERIALS AND METHODS
Cell culture, plasmids, and antibodies. Jurkat T cells and Daudi B cells were
maintained in RPMI 1640 supplemented with 10% fetal bovine serum, glu-
tamine, penicillin, and streptomycin. Goat anti-mouse immunoglobulin G and
goat anti-human immunoglobulin G conjugated to phycoerythrin or biotin were
from Jackson Immunoresearch Laboratories and streptavidin-TC was from BD
Biosciences. The KSR1 knockdown cells and pMX-KSR1-green fluorescent pro-
tein (GFP) fusion were previously reported (13). In some cases, GFP fused to a
catalytically inactivated HDAC6 was used as a GFP fusion control for its simi-
larity in size to KSR1-GFP, but no observable difference was seen with GFP only.
MSCV-KSR1-FLAG-IRES-GFP and its variants were a generous gift from R.
Lewis (University of Nebraska Medical Center). Anti-phospho-ERK, anti-FLAG
(M2), and antitubulin antibodies were from Sigma. For technical reasons, the
data for Fig. 3B use pERK from Cell Signaling (Rb polyclonal). Anti-human
TCR (c305) was produced as ascitic fluid. Anti-mCD3 (2C11) and anti-mCD28
(37.51) antibodies are from BD Biosciences and Southern Biotech, respectively.
Anti-KSR1 and anti-cRaf are from Transduction Labs, and anti-MEK1 and
anti-ERK2 are from Santa Cruz Biotechnology, Inc. Staphylococcal enterotoxin
E (SEE) was from Toxin Technologies, and SDF-1 was kindly provided by D.
Fremont (Washington University).
Transfection and intracellular staining. Jurkat T cells were electroporated
(250 V, 960 ?F) with the indicated DNA. Prior to use, cells were rested in
serum-free medium 30 min prior to stimulation. Cells were then fixed in ice-cold
methanol for at least 20 min. In cells expressing GFP, cells were fixed in 3%
paraformaldehyde for 20 min prior to permeabilization in ice-cold methanol.
Cells were then resuspended in blocking medium (1% bovine serum albumin and
2 mM EDTA in phosphate-buffered saline) and incubated with anti-pERK
antibody (1:750 of ascites), followed by the appropriate secondary antibody. The
data were then acquired by a fluorescence-activated cell sorting scan flow cy-
tometer and analyzed by using FlowJo software. During analysis, antigen-pre-
senting cells (APCs) stained with anti-human immunoglobulin were gated out.
Primary T-cell isolation and chemotaxis assay. T cells were isolated according
to the protocol from the EasySep negative selection CD4?T-cell kit (StemCell
Technologies), and purity was measured by flow cytometry with cells stained for
CD4?and Thy1?. Chemotaxis assays were performed as previously described
(43). Briefly, cells were rested for 1 h at 37 degrees in migration medium (RPMI
1640 with 0.5% bovine serum albumin and 20 mM HEPES [pH 7.0]). A total of
5 ? 105cells were placed in the 6.5-mm Transwell insert with a pore size of 5.0
?m (Costar) and allowed to migrate for 4 h. The migration index was determined
by counting the number of cells that migrated through divided by the number of
cells that migrated through in the absence of chemokine.
The MAPK module generates both digital and graded out-
puts in T cells. To begin our studies, we wanted to develop a
robust, single-cell system for studying different MAPK system
outputs in T cells. We chose Jurkat T cells due to their ease of
transfection and well-characterized ERK activation profile in
response to TCR activation. Because the peptide specificity of
the Jurkat TCR is unknown, we stimulated the cells with the
SAg, SEE, bound to APCs (15). This physiologic stimulation
resulted in rapid and synchronous ERK activation, with most
cells remaining active for more than 60 min (Fig. 2). To char-
acterize whether ERK activation in response to SEE results in
a graded or digital output in Jurkat cells, we activated cells with
serial dilutions of SEE and measured the active ERK output at
a single cell level by flow cytometry using an antibody that
specifically recognizes the active, biphosphorylated form of
ERK (pERK). Importantly, SEE was diluted prior to incubat-
ing with APCs to ensure that all stimulations had the same
number of APCs. Strikingly, we observed only two states of
ERK activation in response to SEE stimulation at 3 min. As
the SEE levels increased, the number of cells that became
activated also increased in a dose-dependent fashion without
individual cells displaying intermediate levels of ERK activa-
tion (Fig. 3A). Since this type of bimodal output response is the
FIG. 1. Graded versus digital signaling. Hypothetical curves repre-
senting the relationship between input (stimuli) and output (pERK) at
the single-cell level for graded compared to digital signaling. On the
right of each curve are hypothetical flow-cytometric histograms depict-
ing increasing pERK levels in relation to increasing stimuli for the two
systems. In a graded system, the pathway transmits continuous infor-
mation that is proportional to the input stimulus. In contrast, the
all-or-nothing digital output can switch between two steady states but
cannot rest in intermediate states, thereby functioning as a digital
switch with the low and high steady states representing “off” and “on,”
respectively (11). These different signal outputs can be used to drive
discrete cell fate decisions within a single cell (37).
VOL. 29, 2009KSR1 DOES NOT ALTER FUNDAMENTAL MAPK PATHWAY OUTPUTS2083
hallmark of digital systems, we conclude that ERK activation
in Jurkat T cells by the SAg, SEE, is digital. To investigate
whether stimulation of the TCR alone is sufficient to generate
the digital response, we stimulated cells with an anti-TCR
antibody. This response was also digital; however, anti-TCR
stimulation resulted in a slightly broader “on” peak (Fig. 3B).
This difference could be due to other receptor-ligand interac-
tions between the T-cell and the APC that help to enforce the
“digitalness” of the system since T cells are only activated by
APCs in physiologic settings.
To determine whether all pathways leading to the activation
of ERK in Jurkat T cells results in a digital response, we
stimulated cells with serially diluted amounts of the chemo-
kine, SDF-1. SDF-1 induces ERK activation in cells by trig-
gering the GPCR, CXCR4 (12, 39). In contrast to SEE acti-
vation, as SDF-1 levels increased, individual cells showed
intermediate levels of pERK with the degree of ERK activa-
tion corresponding to SDF-1 input (Fig. 3C). These results
confirm that T cells display a graded ERK output in response
to SDF-1. Interestingly, the pharmacologic activator of the
MAPK pathway, phorbol myristate acetate (PMA), behaved in
a graded manner as well (Fig. 3D). In combination with the
SEE stimulations above, these data show that T cells can gen-
erate both digital and graded MAPK system outputs in vivo.
We conclude that the system output from the MAPK module
in T cells is highly plastic and suggest that the type of output
generated from the module is determined by signals emanating
from the initiating receptor.
It is important to note that while ERK activation in response
to SDF-1 and PMA, although distinct from the strong bimodal
character of TCR signaling, did not increase in a perfectly
linear fashion as the stimulus increased. We suspect that true
analog outputs are rare in biological circuits that are used to
generate digital as well as graded outputs, since some degree of
nonlinearity is necessary for digital signaling to occur (11, 18).
Rather, weakly switch-like outputs are used to functionally
mimic analog circuits, as is the case for SDF-1 and PMA
activation of the MAPK module in T cells. For this reason
we specifically use the word “graded” instead of “analog”
Reducing KSR1 levels lowers the sensitivity of the digital
MAPK system without changing its fundamental system out-
put. In mammals, there are two KSR isoforms, KSR1 and
KSR2. We focused our study on KSR1, since we have been
unable to detect KSR2 expression in mouse or human T cells.
It is therefore unlikely that KSR2 plays any role in T cells. In
addition, we previously demonstrated that KSR1-deficient T
cells show reduced levels of ERK activation when stimulated
with antibodies to the TCR or PMA (31), confirming that
KSR1 is essential for proper ERK activation in T cells. How-
ever, these experiments were done at the population level and
therefore cannot discriminate whether the reduced ERK out-
put was due to a change in system sensitivity or whether the
system output was fundamentally changed within individual
First, we sought to formally determine the role of KSR1 in
modulating MAPK system output. We utilized a previously
reported KSR1 knockdown Jurkat cell line to assess how re-
ducing the level of KSR1 affected the MAPK system output in
response to SEE activation at a single cell level (13). KSR1
knockdown cells still exhibited digital ERK activation after
TCR engagement. This was demonstrated by the bimodal dis-
tribution of pERK staining (Fig. 4A) and the absence of any
cells with intermediate staining. In addition, the intensity of
pERK staining for both the “off” and the “on” populations was
similar to the controls even though KSR1 levels were signifi-
cantly decreased. These results show that KSR1 is not respon-
sible for generating the digital MAPK system output.
KSR1 depletion did, however, significantly modulate the
sensitivity of the system to SEE input. In the KSR1 knockdown
line, significantly fewer cells were activated at any given input
relative to control cells (Fig. 4A). Thus, high KSR1 expression
increases the probability of a cell activating for a given input,
whereas the loss of KSR1 expression decreases probability of
activation. These findings are entirely consistent with early
work showing that optimal KSR expression increases the prob-
ability of the all-or-nothing cell fate decision of Xenopus oocyte
maturation (6). In combination with our data, the two studies
clearly demonstrate that KSR1 does not affect the digital na-
ture of the MAPK response. Instead, these results show that in
digital MAPK systems the scaffold protein KSR1 functions to
lower the threshold for ERK activation.
Reducing KSR1 levels lower the sensitivity of the graded
MAPK system without changing its fundamental system re-
sponse. We next explored the possibility that KSR1 might be
involved in facilitating graded ERK responses. ERK activation
occurs only after the phosphorylation of a threonine and ty-
rosine residue by its upstream activator, MEK. Experimental
FIG. 2. Activation of the MAPK pathway in Jurkat T cells is rapid and sustained. Jurkat T cells were stimulated with SEE-coated (1 ?g/ml)
APCs for the indicated time points. Cells were then fixed and stained for pERK and analyzed by flow cytometry. Samples were also stained with
anti-human immunoglobulin to exclude APCs during analysis.
2084 LIN ET AL.MOL. CELL. BIOL.
FIG. 3. Jurkat T cells can generate both digital and graded responses. (A) SEE was serially diluted twofold starting at 250 ng/ml prior to
incubation with APCs. Cells were then stimulated for 3 min, fixed, and stained with an anti-pERK monoclonal antibody and the appropriate
secondary antibody, followed by analysis by flow cytometry. (B) Jurkat T cells were stimulated with twofold serially diluted concentrations of
anti-TCR monoclonal antibody starting at a 1:500 dilution of ascitic fluid for 3 min and analyzed with an anti-pERK polyclonal antibody. (C) Jurkat
T cells were stimulated with twofold serially diluted concentrations of SDF-1 starting at 12.5 nM for 3 min and analyzed as described for panel
A. (D) Cells were stimulated with 2-fold serial dilutions starting at 25 ng of PMA/ml for 3 min and analyzed as described for panel A.
data demonstrate that when ERK and MEK are in solution,
each of the phosphorylation events occurs independently, a
mechanism termed “distributive phosphorylation” (5, 10). Ki-
netic modeling demonstrates that the requirement for two
independent events to activate ERK could contribute to a
digital response. (27). Since KSR1 couples MEK and ERK
together and therefore the two phosphorylation events (4),
KSR1 may be predicted to enhance the graded response. Re-
cent results in yeast studies provide strong experimental sup-
port for this hypothesis (42); however, the generality of this
model has not been tested.
We therefore tested whether suppression of KSR1 in Jurkat
cells affected the graded response of ERK to SDF-1. SDF-1
stimulation continued to produce a graded ERK output in
KSR1-depleted Jurkat T cells (Fig. 4B). KSR1 depletion did,
however, reduce the sensitivity of the system, since KSR1-
depleted cells generated less pERK in response to SDF-1 stim-
ulation relative to control cells at various doses of SDF-1. To
FIG. 4. KSR1 knockdown increases the threshold required to activate ERK at a per-cell basis. (A) Control T cells or KSR1 knockdown cells
were stimulated for 3 min with serially diluted SEE-coated APCs followed by staining with an anti-pERK monoclonal antibody and the appropriate
secondary antibody. Cells were then analyzed by flow cytometry. Shown are fourfold serial dilutions starting at 250 ng of SEE/ml. (B) Control T
cells or KSR1 knockdown cells were stimulated for 3 min with serially diluted SDF-1, followed by staining and analysis as described for panel A.
Shown are twofold serial dilutions starting at 5 nM SDF-1. Inset numbers represent the mean fluorescence intensity. The panels on the right show
the overlay of all of the stimulation points for the individual control and KSR1 knockdown cell lines. The lower panels show the overlay between
control and KSR1 knockdown cells for each individual stimulation point.
2086 LIN ET AL.MOL. CELL. BIOL.
ensure these results were due to the reduction in KSR1 ex-
pression and not an off-target effect of the KSR1 knock-down,
we repeated these experiments with mouse embryo fibroblasts
derived from wild-type or KSR1 knockout mice (31) using
epidermal growth factor as the input signal. Precisely the same
results were obtained in the MEF KSR1 knockout system as
for the Jurkat KSR1 knockdown system; the absence of KSR1
reduced the sensitivity of the MAPK response but did not
change the graded system output (data not shown). In combi-
nation, these results argue that, in contrast to the yeast Ste5
scaffold protein, mammalian KSR1 does not enhance graded
signaling from the MAPK pathway.
Increasing KSR1 expression levels do not reconfigure a dig-
ital or graded MAPK system. We next tested whether increas-
ing expression of KSR1 rewired the digital SEE system to
generate a graded output. Although it has been previously
published that ectopic expression of MAPK scaffolds inhibit
MAPK activation (23, 44), none of these studies have carefully
examined this effect on digital or graded systems at a single cell
level. Jurkat cells were transfected with a KSR1-GFP fusion to
allow us to examine the effect of increasing amounts of KSR1
on digital SAg-induced ERK activation. Strikingly, when the
levels of KSR1 reached a critical level, ERK activation was
completely inhibited (Fig. 5A). Importantly, the SEE system
retained a digital system output as KSR1 levels increased.
Moreover, KSR1 inhibition of the SEE system was digital,
whereas KSR1 inhibition of the SDF-1 system was graded,
mirroring the activation responses (Fig. 5A). This is clearly
shown when a narrow GFP gate is drawn at the inflection point
between on and off states so as to maintain roughly equal
numbers of cells that are on and off. Cells exist at an interme-
diate state when stimulated with SDF-1, whereas with SEE the
cells display a bimodal distribution. It is important to note that
while the width of the GFP gates are identical for the two
stimuli, the gate for cells stimulated with SDF-1 had to be
shifted slight to the right to maintain roughly equal numbers of
cells that are on and off, reflecting that slightly higher levels of
KSR1 are required to inhibit SDF-1-induced ERK activation
compared to SEE-induced ERK activation (Fig. 5A). In com-
bination, these data strongly argue that KSR1 does not recon-
figure the MAPK module from a digital to a graded output. It
is interesting that increases in KSR1 levels did not appreciably
increase the sensitivity of the digital SEE system or the graded
SDF-1 system in Jurkat cells (Fig. 5B and C). This result
indicates that KSR1 is expressed at close to the optimal levels
for signal transduction in this cell line. From these experi-
ments, we conclude that KSR1 does not engineer the MAPK
module to generate different system outputs but functions
solely to determine the sensitivity of the systems in which it is
KSR1 scaffold function is critically dependent on MEK
binding. High-level expression of scaffold proteins inhibits sig-
naling cascades by blocking the productive interactions of
pathway constituents (6, 19, 24, 36, 41). Early studies charac-
terizing the role of KSR1 demonstrated that MEK-KSR1 bind-
ing is crucial for KSR1 function (30, 44). However, more re-
cent results suggest the role of KSR1 in regulating the MAPK
pathway is more complicated than originally thought. For ex-
ample, KSR1 participates in Raf activation (35), which may
explain why the loss of MEK binding did not inhibit the ability
of KSR1 (C809Y) to alter the biological actions of oncogenic
Ras (20). To confirm that KSR1 functions as a scaffold in T
cells, we examined the inhibitory ability of ectopically ex-
pressed mutant KSR1 proteins with compromised scaffold
function. As seen previously, ectopic expression of wild-type
KSR1 completely inhibited ERK activation once the inhibitory
level of KSR1 expression was reached (Fig. 5A and 6B). To
confirm that KSR1 functions as a scaffold in T cells, we tested
the mutant KSR1 protein that cannot bind MEK (C809Y) (30,
44) for its ability to inhibit MAPK signaling. Western blotting
of immunoprecipitated KSR1 wild type and KSR1 mutant
(C809Y) confirmed that the mutant protein does not bind
MEK even under TCR-stimulated conditions (Fig. 6A). The
KSR1 (C809Y) mutant failed to inhibit ERK activation at any
level of expression, confirming that the scaffolding the MAPK
module is a crucial function of KSR1 in T cells (Fig. 6B).
If the role of scaffolds, such as KSR1, is to simply link MEK
to ERK, the inhibitory effect due to overexpression of KSR1
should in theory be a dose dependent, graded inhibition due to
a “dilution” effect as previously modeled (24). To determine
how the dose-dependent inhibition of ERK activation due to a
competitive inhibitor at a single cell level would appear, we
ectopically expressed a kinase-dead version of MEK that func-
tions as a dominant negative. Its inhibition of ERK is much
more graded at high expression levels, which is consistent with
the idea of a competitive inhibitor (Fig. 6C). Also, ectopic
expression of another scaffold that links MEK and ERK, MP-1
(38), did not induce the same inhibitory effect as KSR1 over-
expression (data not shown). Clearly, the mechanism of inhi-
bition induced by ectopic expression of KSR1 is not as simple
as originally thought and will require further investigation into
possible mechanisms involving cooperative binding and feed-
KSR1 is important for proper activation of the MAPK path-
way and biological output in primary T cells. To confirm our
findings in a more physiologic system, we directly examined the
effect of the loss of KSR1 expression on ERK activation in
primary T cells isolated from KSR1-deficient mice. Primary
CD4?T cells isolated from wild-type and KSR1 knockout mice
were subjected to stimulation with either anti-TCR antibodies
or SDF-1. As previously reported, the absence of KSR1 se-
verely impaired TCR-induced ERK activation (Fig. 7A) (31).
Strikingly, SDF-1 stimulation also demonstrated a severe ERK
activation defect in KSR1-deficient primary T cells, confirming
our results obtained with Jurkat T cells (Fig. 7A). Although the
crucial role of KSR1 during TCR mediated T-cell activation is
well established (31), it is unknown how the loss of KSR1
affects SDF-1-mediated T-cell function. To directly address
this question, we performed chemotaxis assays on primary T
cells isolated from wild-type and KSR1 knockout mice. In line
with their impaired ERK activation, KSR1-deficient T cells
exhibited a substantial defect in their ability to migrate toward
SDF-1 (Fig. 7B). From these results, we conclude that KSR1
does play a physiologically important role in regulating the
sensitivity of ERK activation to SDF-1.
KSR1 expression is carefully regulated following T-cell ac-
tivation. To investigate our hypothesis that KSR1 plays a cen-
tral role in setting ERK activation thresholds, we examined
whether KSR1 levels were actively regulated during T-cell re-
sponses. Naive versus primed T cells are thought to have dif-
VOL. 29, 2009 KSR1 DOES NOT ALTER FUNDAMENTAL MAPK PATHWAY OUTPUTS2087
ferent activation requirement for proliferation and cytokine
release (reviewed in reference 1). To determine whether KSR1
levels were playing a role in this differential activation require-
ment, we isolated CD4?T cells from mice and compared
KSR1 levels from freshly isolated cells to cells that had been
stimulated with anti-CD3 and anti-CD28 for 5 days. We also
compared KSR1 expression levels relative to the MAPK com-
ponents Raf, MEK, and ERK in these cells. Strikingly, KSR1
expression dramatically decreased as T cells differentiated rel-
ative to the components of the MAPK module (Fig. 8). These
FIG. 5. OverexpressionofKSR1inhibitsERKactivation.(A)JurkatTcellsweretransfectedwitheitheracontrolGFPorKSR1-GFPfusionand,18hlater,
the cells were stimulated for 3 min with either SEE-coated (1 ?g/ml) APCs or 1 ?M SDF-1. Cells were then stained with anti-pERK and analyzed by flow
cytometry. A narrow gate was drawn to show the cells at the inflection point of low to high pERK staining. Note that the gate is slightly shifted to the right in
the SDF-1 stimulation to maintain approximately equal numbers of low and high cells in the histogram. (B) Control GFP or KSR1-GFP transfected cells were
for 3 min with twofold serially diluted SDF-1 starting at 16 nM. Cells were then stained for pERK and analyzed by flow cytometry.
2088 LIN ET AL.MOL. CELL. BIOL.
data reveal that KSR1 expression level is indeed regulated
during T-cell activation relative to the components of the
The three-tiered architecture of MAPK modules has been
highly conserved throughout eukaryotic evolution, confirming
the utility of this circuit configuration in biology (22). It has
become increasingly clear that MAPK modules are highly plas-
tic, generating multiple types of signal outputs that are used by
cells to regulate divergent biological functions (14). Here we
provide direct experimental support that this paradigm holds
true for T cells. Activation of the MAPK module by the TCR
generates a digital response, whereas activation of the module
by a GPCR is graded. These distinct MAPK outputs are func-
tionally suited for their respective T-cell biological functions.
Since T-cell activation by the TCR is a discrete cell fate deci-
sion, a digital output from the MAPK module is appropriate to
drive this decision-making process (2). In the case of a chemo-
kine response, cells are interpreting a chemokine gradient to
determine in which direction to migrate. In this instance, the
appropriate MAPK output is a graded one, with the amount of
output increasing as the cell moves into areas with increasingly
higher concentrations of chemokines.
Our finding that output from the MAPK is highly plastic in
T cells suggests that signals emanating from the initiating re-
ceptor are used to set the type of MAPK response. Since
scaffolds can potentially integrate positive- and negative-feed-
back loops to shape the output of kinase cascades (3), we
tested whether differential usage of KSR1, the best-character-
ized mammalian ERK scaffold, was responsible for the differ-
ences in TCR versus GPCR MAPK signal output. Our findings
do not support a specific role of KSR1 in determining the
system output of the MAPK module. Instead, KSR1 enhanced
the efficiency of ERK activation in both graded and digital
systems, confirming recent predictions based on computational
models of scaffold function (25).
The distributive, two-step phosphorylation of ERK is an
important contributor to digital signal output (10, 18). Teth-
ering MEK and ERK in a scaffold complex is predicted to
suppress distributive phosphorylation and thereby inhibit dig-
ital output to enhance graded signaling (5), a prediction con-
FIG. 6. KSR1 inhibition requires MEK binding. (A) Jurkat T cells were transfected with vector control, wild-type (wt) KSR1-FLAG, or KSR
(C809Y)-FLAG and then left unstimulated or stimulated for 3 min with anti-TCR. Cells were then lysed and immunoprecipitated with an
anti-FLAG monoclonal antibody. Proteins were separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis and visualized by
immunoblotting with anti-FLAG and anti-MEK1 antibodies. (B) Jurkat T cells were transfected with control GFP, KSR1-IRES-GFP, or KSR
(C809Y)-IRES-GFP. After 18 h, the cells were stimulated for 3 min, as before, with SEE-coated APC (1 ?g/ml). Cells were fixed and stained as
described above. (C) Jurkat T cells were transfected with control GFP or a dominant-negative MEK (DN-MEK). Cells were then stimulated with
SEE-coated APC (1 ?g/ml), fixed, and stained as described above.
VOL. 29, 2009 KSR1 DOES NOT ALTER FUNDAMENTAL MAPK PATHWAY OUTPUTS2089
firmed recently in the yeast MAPK scaffold Ste5; however, it is
important to note that the readouts in this system were tran-
scriptionally regulated (42). Given these results, it was surpris-
ing that KSR1 failed to enhance graded MAPK signaling. Our
data reveal that different scaffold proteins have qualitatively
different effects on MAPK signaling; thus, it may not be pos-
sible to make generalizations about how scaffolds modulate
MAPK signaling. Rather, the regulatory functions of individual
scaffold proteins need to be determined empirically. A chal-
lenge for the future will be to determine the biochemical basis
of digital and graded responses. It seems likely that a variety of
different factors, including scaffold proteins, localization within
the cell, positive and negative feedback, and the complexity of
signaling pathways, will each play roles in determining the final
character of a signaling response.
A central question in T-cell biology is how TCR responses
are tuned during T-cell development, activation, and differen-
tiation in the periphery (9). The threshold for discrimination
between pMHC ligands changes during T-cell maturation. As
T cells mature, they maintain the same sensitivity for foreign
agonist ligands while losing the capacity to respond to weak
self-pMHC complexes (8, 26). Recent experimental results
suggest that T cells may also modify their pMHC discrimina-
tion threshold in the periphery by using endogenous pMHC
ligands to boost T-cell response to agonist pMHC ligands (21,
40). These combined observations show that TCR responsive-
ness is not set for each TCR but is fine-tuned depending on the
differentiation state of the T cell. The molecular mechanisms
underlying the tuneability of the TCR are not well understood.
Our results show that KSR1 occupies a prime position from
which to regulate the plasticity of T-cell responses, with
changes in KSR1 expression level being able to modulate both
the threshold of TCR activation and the rate of chemotaxis.
Consistent with this role, we show that KSR1 expression levels
are tightly regulated during T-cell activation, as naive T cells
express significantly higher levels of KSR1 protein relative
to their primed counterparts. These data support the previous
observation that KSR1 is required for proper naive T-cell differ-
entiation into TH1 and TH2 cells but not for cytokine production
after restimulation of fully differentiated TH1 and TH2 cells (31).
Also, in support of this model, other groups have reported that
KSR1 function can be modulated by caspase-dependent cleavage
of KSR1 or changes in KSR1 protein half-life (29, 33). Future
work will explore changes in KSR1 expression during thymocyte
development and T-cell differentiation. Also, examining KSR1
levels in anergic or tolerized T cells could provide insight into the
biology of these cells.
In conclusion, we have demonstrated the MAPK module is
highly plastic in T cells, displaying fundamentally different sys-
tem outputs and sensitivities. We go on to show that the scaf-
fold protein, KSR1, modulates MAPK system sensitivity but
not system output. We also reveal that KSR1 expression level
is carefully regulated during T-cell maturation. These findings
identify KSR1 as a likely control point used to fine-tune
T-cell responses during T-cell activation and differentiation,
suggesting KSR1 as a key regulator of T-cell function and
plasticity in vivo.
FIG. 7. KSR1-deficient primary T cells fail to activate ERK in
response to TCR or SDF-1. (A) Primary CD4?T cells were isolated
from wild-type (wt) or KSR1 knockout mice to ?85% purity and
stimulated with either an anti-TCR antibody or 100 nM SDF-1 for 3
min. Cells were then lysed, and ERK activation was measured by
pERK blotting. (B) Chemotaxis of purified primary CD4?T cells from
either wild-type (wt) or KSR1 knockout mice was measured by a
Transwell migration assay. The migration index is a measure of the
number of cells that migrated into the bottom chamber, in the pres-
ence of the indicated SDF-1 concentration, divided by the number of
cells that migrated with no chemokine. The bars represent the aver-
ages, and the error bars indicate the standard deviations of samples
evaluated in triplicate.
FIG. 8. KSR1 levels are regulated in primary T cells after stim-
ulation. Primary CD4?T cells were isolated from wt mice and
stimulated with plate-bound anti-CD3 and anti-CD28 antibody for
5 days. Cells were then lysed, and proteins were resolved by sodium
dodecyl sulfate-polyacrylamide gel electrophoresis. KSR1, c-Raf,
MEK1, and ERK2 were visualized by immunoblotting with their
2090 LIN ET AL.MOL. CELL. BIOL.
ACKNOWLEDGMENTS Download full-text
We thank Erin Filbert, Shuba Srivatsan, and Kathleen Cato for
critical reading of the manuscript. We also thank David Fremont and
Rob Lewis for critical reagents.
J.L. is supported by a Cancer Research Institute Postdoctoral Fel-
lowship. A.H. is supported by an NHMRC C. J. Martin Research
1. Akbar, A. N., M. Salmon, and G. Janossy. 1991. The synergy between naive
and memory T cells during activation. Immunol. Today 12:184–188.
2. Altan-Bonnet, G., and R. N. Germain. 2005. Modeling T-cell antigen dis-
crimination based on feedback control of digital ERK responses. PLoS Biol.
3. Bashor, C. J., N. C. Helman, S. Yan, and W. A. Lim. 2008. Using engineered
scaffold interactions to reshape MAP kinase pathway signaling dynamics.
4. Burack, W. R., and A. S. Shaw. 2000. Signal transduction: hanging on a
scaffold. Curr. Opin. Cell Biol. 12:211–216.
5. Burack, W. R., and T. W. Sturgill. 1997. The activating dual phosphorylation
of MAPK by MEK is nonprocessive. Biochemistry 36:5929–5933.
6. Cacace, A. M., N. R. Michaud, M. Therrien, K. Mathes, T. Copeland, G. M.
Rubin, and D. K. Morrison. 1999. 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. 19:229–240.
7. Daniels, M. A., E. Teixeiro, J. Gill, B. Hausmann, D. Roubaty, K. Holmberg,
G. Werlen, G. A. Hollander, N. R. Gascoigne, and E. Palmer. 2006. Thymic
selection threshold defined by compartmentalization of Ras/MAPK signal-
ling. Nature 444:724–729.
8. Davey, G. M., S. L. Schober, B. T. Endrizzi, A. K. Dutcher, S. C. Jameson,
and K. A. Hogquist. 1998. Preselection thymocytes are more sensitive to
T-cell receptor stimulation than mature T cells. J. Exp. Med. 188:1867–1874.
9. Feinerman, O., J. Veiga, J. R. Dorfman, R. N. Germain, and G. Altan-
Bonnet. 2008. Variability and robustness in T-cell activation from regulated
heterogeneity in protein levels. Science 321:1081–1084.
10. Ferrell, J. E., Jr., and R. R. Bhatt. 1997. Mechanistic studies of the dual
phosphorylation of mitogen-activated protein kinase. J. Biol. Chem. 272:
11. Ferrell, J. E., and W. Xiong. 2001. Bistability in cell signaling: how to make
continuous processes discontinuous, and reversible processes irreversible.
12. Ganju, R. K., S. A. Brubaker, J. Meyer, P. Dutt, Y. Yang, S. Qin, W. Newman,
and J. E. Groopman. 1998. The alpha-chemokine, stromal cell-derived factor-
1alpha, binds to the transmembrane G-protein-coupled CXCR-4 receptor and
13. Giurisato, E., J. Lin, A. Harding, E. Cerutti, M. Cella, R. E. Lewis, M.
Colonna, and A. S. Shaw. The MAP kinase scaffold KSR1 is required for
recruitment of ERK to the immunological synapse. Mol. Cell. Biol., in press.
14. Harding, A., and J. F. Hancock. 2008. Ras nanoclusters: combining digital
and analog signaling. Cell Cycle 7:127–134.
15. Hudson, K. R., H. Robinson, and J. D. Fraser. 1993. Two adjacent residues
in staphylococcal enterotoxins A and E determine T-cell receptor V beta
specificity. J. Exp. Med. 177:175–184.
16. Inder, K., A. Harding, S. J. Plowman, M. R. Philips, R. G. Parton, and J. F.
Hancock. 2008. Activation of the MAPK module from different spatial lo-
cations generates distinct system outputs. Mol. Biol. Cell 19:4776–4784.
17. Jaumot, M., and J. F. Hancock. 2001. Protein phosphatases 1 and 2A pro-
mote Raf-1 activation by regulating 14-3-3 interactions. Oncogene 20:3949–
18. Kholodenko, B. N. 2006. Cell-signalling dynamics in time and space. Nat.
Rev. Mol. Cell. Biol. 7:165–176.
19. Kornfeld, K., D. B. Hom, and H. R. Horvitz. 1995. The ksr-1 gene encodes
a novel protein kinase involved in Ras-mediated signaling in Caenorhabditis
elegans. Cell 83:903–913.
20. Kortum, R. L., H. J. Johnson, D. L. Costanzo, D. J. Volle, G. L. Razidlo,
A. M. Fusello, A. S. Shaw, and R. E. Lewis. 2006. The molecular scaffold
kinase suppressor of Ras 1 is a modifier of RasV12-induced and replicative
senescence. Mol. Cell. Biol. 26:2202–2214.
21. Krogsgaard, M., Q. J. Li, C. Sumen, J. B. Huppa, M. Huse, and M. M. Davis.
2005. Agonist/endogenous peptide-MHC heterodimers drive T-cell activa-
tion and sensitivity. Nature 434:238–243.
22. Kyriakis, J. M., and J. Avruch. 2001. Mammalian mitogen-activated protein
kinase signal transduction pathways activated by stress and inflammation.
Physiol. Rev. 81:807–869.
23. Laurent, M. N., D. M. Ramirez, and J. Alberola-Ila. 2004. Kinase suppressor
of Ras couples Ras to the ERK cascade during T-cell development. J. Im-
24. Levchenko, A., J. Bruck, and P. W. Sternberg. 2000. Scaffold proteins may
biphasically affect the levels of mitogen-activated protein kinase signaling
and reduce its threshold properties. Proc. Natl. Acad. Sci. USA 97:5818–
25. Locasale, J. W., A. S. Shaw, and A. K. Chakraborty. 2007. Scaffold proteins
confer diverse regulatory properties to protein kinase cascades. Proc. Natl.
Acad. Sci. USA 104:13307–13312.
26. Lucas, B., I. Stefanova, K. Yasutomo, N. Dautigny, and R. N. Germain. 1999.
Divergent changes in the sensitivity of maturing T cells to structurally related
ligands underlies formation of a useful T-cell repertoire. Immunity 10:367–
27. Markevich, N. I., J. B. Hoek, and B. N. Kholodenko. 2004. Signaling switches
and bistability arising from multisite phosphorylation in protein kinase cas-
cades. J. Cell Biol. 164:353–359.
28. Matheny, S. A., and M. A. White. 2005. Ras-sensitive IMP modulation of the
Raf/MEK/ERK cascade through KSR1. Methods Enzymol. 407:237–247.
29. McKay, M. M., and D. K. Morrison. 2007. Caspase-dependent cleavage
disrupts the ERK cascade scaffolding function of KSR1. J. Biol. Chem.
30. Muller, J., A. M. Cacace, W. E. Lyons, C. B. McGill, and D. K. Morrison.
2000. Identification of B-KSR1, a novel brain-specific isoform of KSR1 that
functions in neuronal signaling. Mol. Cell. Biol. 20:5529–5539.
31. Nguyen, A., W. R. Burack, J. L. Stock, R. Kortum, O. V. Chaika, M. Afkar-
ian, W. J. Muller, K. M. Murphy, D. K. Morrison, R. E. Lewis, J. McNeish,
and A. S. Shaw. 2002. Kinase suppressor of Ras (KSR) is a scaffold which
facilitates mitogen-activated protein kinase activation in vivo. Mol. Cell.
32. Ory, S., M. Zhou, T. P. Conrads, T. D. Veenstra, and D. K. Morrison. 2003.
Protein phosphatase 2A positively regulates Ras signaling by dephosphory-
lating KSR1 and Raf-1 on critical 14-3-3 binding sites. Curr. Biol. 13:1356–
33. Razidlo, G. L., R. L. Kortum, J. L. Haferbier, and R. E. Lewis. 2004.
Phosphorylation regulates KSR1 stability, ERK activation, and cell prolifer-
ation. J. Biol. Chem. 279:47808–47814.
34. Ritt, D. A., I. O. Daar, and D. K. Morrison. 2005. KSR Regulation of the
Raf-MEK-ERK Cascade. Methods Enzymol. 407:224–237.
35. Ritt, D. A., M. Zhou, T. P. Conrads, T. D. Veenstra, T. D. Copeland, and
D. K. Morrison. 2007. CK2 Is a component of the KSR1 scaffold complex
that contributes to Raf kinase activation. Curr. Biol. 17:179–184.
36. Roy, F., G. Laberge, M. Douziech, D. Ferland-McCollough, and M. Therrien.
2002. KSR is a scaffold required for activation of the ERK/MAPK module.
Genes Dev. 16:427–438.
37. Santos, S. D., P. J. Verveer, and P. I. Bastiaens. 2007. Growth factor-induced
MAPK network topology shapes Erk response determining PC-12 cell fate.
Nat. Cell Biol. 9:324–330.
38. Schaeffer, H. J., A. D. Catling, S. T. Eblen, L. S. Collier, A. Krauss, and M. J.
Weber. 1998. MP1: a MEK binding partner that enhances enzymatic activa-
tion of the MAP kinase cascade. Science 281:1668–1671.
39. Sotsios, Y., G. C. Whittaker, J. Westwick, and S. G. Ward. 1999. The CXC
chemokine stromal cell-derived factor activates a Gi-coupled phosphoinosi-
tide 3-kinase in T lymphocytes. J. Immunol. 163:5954–5963.
40. Stefanova, I., B. Hemmer, M. Vergelli, R. Martin, W. E. Biddison, and
R. N. Germain. 2003. TCR ligand discrimination is enforced by compet-
ing ERK positive and SHP-1 negative feedback pathways. Nat. Immunol.
41. Sundaram, M., and M. Han. 1995. The C. elegans ksr-1 gene encodes a novel
Raf-related kinase involved in Ras-mediated signal transduction. Cell 83:
42. Takahashi, S., and P. M. Pryciak. 2008. Membrane localization of scaffold
proteins promotes graded signaling in the yeast MAP kinase cascade. Curr.
43. Takesono, A., R. Horai, M. Mandai, D. Dombroski, and P. L. Schwartzberg.
2004. Requirement for Tec kinases in chemokine-induced migration and
activation of Cdc42 and Rac. Curr. Biol. 14:917–922.
44. Yu, W., W. J. Fantl, G. Harrowe, and L. T. Williams. 1998. Regulation of the
MAP kinase pathway by mammalian Ksr through direct interaction with
MEK and ERK. Curr. Biol. 8:56–64.
VOL. 29, 2009 KSR1 DOES NOT ALTER FUNDAMENTAL MAPK PATHWAY OUTPUTS2091