Content uploaded by Kumar Selvarajoo
Author content
All content in this area was uploaded by Kumar Selvarajoo
Content may be subject to copyright.
Inferring the mechanistic basis for the dynamic response of the
MyD88-dependent and –independent pathways
[Extended Abstract]
Kumar Selvarajoo
kumar@ttck.keio.ac.jp
Mohamed Helmy
Mohelmy78@yahoo.com
Masaru Tomita
tomita@ttck.keio.ac.jp
Masa Tsuchiya
tsuchiya@ttck.keio.ac.jp
Institute for Advanced Biosciences, Keio University, 14-1 Baba-cho, Tsuruoka, Yamagata 997-0035, Japan.
ABSTRACT
We constructed dynamic in silico Toll-like
receptor (TLR) 4 models constituting the
Myeloid Differentiation factor 88 (MyD88)-
dependent and –independent pathways and
investigated the experimental induction of
MyD88-dependent pathway genes Tnf, Il1
,
Cxcl1 and the MyD88-independent pathway
genes Ccl5, Cxcl10, Ifit1 in wildtype and several
knock-out (KO) conditions. By fitting our model
with wildtype experimental data and analysing
with MyD88 KO, TRIF KO and MyD88 and
TRIF (double) KO conditions, we infer that the
crosstalk between TRIF to TRAF6 and TRIF to
TAB/TAK complex via RIP1 is key for the
quantitative induction of MyD88-dependent
pathway genes Tnf, Il1
, Cxcl1 in MyD88 KO
conditions. Our systemic approach provides a
novel perspective to dynamic TLR4 pathways
analysis.
Keywords
MyD88, TLR4, Signaling, Gene Expression,
Computational Models.
INTRODUCTION
The Toll-like receptors (TLRs), play a critical
role in mammalian first line of defense against
invading pathogens by recognizing a variety of
pathogen-associated molecular patterns (PAMPs)
such as lipopolysaccarides (LPS). The activation
of TLRs in immune cells results in the induction
of proinflammatory cytokines which together
with antigen presenting capacity recruit naïve T-
cells and activate acquired immunity (Akira et al,
2001). The malfunction of these processes leads
to proinflammatory diseases such as autoimmune
diseases, arthrosclerosis, asthma etc (O’neill,
2006).
Among the 13 known mammalian TLRs, the
TLR4 is the most well studied. The activation of
TLR4, by LPS, triggers the MyD88-dependant
pathway and the MyD88-independant pathways
(Selvarajoo, 2006). The main role of the MyD88-
dependant pathway is to initiate the induction of
proinflammatory cytokines such as IL-6, IL-12
and TNF-α. The MyD88-independant pathway,
on the other hand, induces Type I Interferons
(IFNs) and chemokines such as CCL5 and
CXCL10 (O’neill, 2006).
Although the many individual components of the
TLR4 signaling pathway are well studied and
characterized, the dynamic behavior of these
pathways remains poorly understood. For
example, the MyD88-dependent pathway genes
Il1
, Cxcl1, Tnf were still induced in MyD88
knock-out (KO) conditions in contrast to popular
expectation (Hirotani et al, 2005). We
investigated the mechanistic reason for such
International Conference on Molecular Systems Biology 2008
110
observations by adopting a systemic approach;
dynamic computational modeling with
comparison to experimental data of the TLR4
system. Our model demonstrates that the
crosstalk between the MyD88-dependant
pathway and -independant pathways at i) TIR-
domain-containing adapter-inducing interferon-β
(TRIF) to TNF receptor-associated factor
(TRAF) 6 and ii) TRIF to TAK-1 binding
proteins/Transforming growth factor-ß–Activated
Kinase (TAB/TAK) complex are vital for the
induction of Il1
, Cxcl1, Tnf in MyD88 KO
conditions.
METHODS
Computational Model: Recently, we published
our macrophage TLR4 model developed using
ordinary differential equations with pulse
perturbation given to TLR4 (Selvarajoo,
2006:2007). We extended this model to simulate
the induction of MyD88-dependent genes Tnf,
Il1
, Cxcl1 in four genotypes, wildtype, MyD88
KO, TRIF KO, MyD88/TRIF double KO
conditions (Model I) (Fig. 1A). Briefly, the TLR4
stimulated with LPS, leads to the activation of
MyD88-dependant and -independant pathway.
The MyD88-dependant pathways leads to the
activation of the transcription factors Activator
Protein -1 (AP-1) and Nuclear Factor–kappa B
(NF-κB) which results in the transcription of
proinflammatory genes such as Tnf, Il1
, Cxcl1.
The MyD88-independant pathway activates
Interferon Regulatory Factor 3 (IRF-3) and NF-
κB, inducing chemokine genes such as Ccl5,
Cxcl10, Ifit [Fig. 1]. The details of model
development and parameter selection are found in
(Selvarajoo, 2006:2007).
Experiments: We utilized experimental data
obtained from murine macrophages, with LPS
stimulation, for wildtype, MyD88-/- mice, TRIF-
/- mice and MyD88-/-TRIF-/- mice. Affymetrix
mouse expression array A430 microarray chips
were used for gene expression detection (Hirotani
et al., 2005).
IкB
IKK Complex
Extracellular
Cytoplasm
TLR4
Nucleus
ex. Ccl5
,
Cxcl10
,
Ifit1
ex. Tnf
,
Il1
,
Cxcl1
LPS
NF-кB
p38
JNK
ELK1
NF-кB
MKK4/7MKK/
MKK/
p38ERK
TAB/TAK
Complex
IRAK1 IRAK4
TRAF6
MyD88
NF-кBAP-1
JNK
I1 I2 I3
TBK1
IRF-3
TRAM
NF-кBIRF-3
TRIF
IкB
IKK Complex
Extracellular
Cytoplasm
TLR4
Nucleus
ex. Ccl5
,
Cxcl10
,
Ifit1
ex. Tnf
,
Il1
,
Cxcl1
LPS
NF-кB
p38
JNK
ELK1
NF-кB
MKK4/7MKK/
MKK/
p38ERK
TAB/TAK
Complex
IRAK1 IRAK4
TRAF6
MyD88
NF-кBAP-1
JNK
I1 I2 I3
TBK1
IRF-3
TRAM
NF-кBIRF-3
RIP1
TRIF
Figure 1 Schematic representation of TLR4 signaling
pathways. A) Model I which contains one crosstalk
mechanism between the MyD88-dependant and -
independant pathways (TBK1→NF-B). B) Model II
which contains a total of three crosstalk mechanisms
between the MyD88-dependant and -independant pathways
(TBK1→NF-B, TRIF → TRAF6 and TRIF→ RIP1→
TAB/TAK complex).
A
B
International Conference on Molecular Systems Biology 2008
111
RESULTS AND DISCUSSION
We focused our simulation on LPS primary
signaling mechanism, for up to 1 hour only when
secondary effects such as negative feedback
mechanisms are unlikely to affect our original
TLR4 signaling. We used semi-quantitative
wildtype experimental profiles of signaling
molecules Extracellular Regulated MAP Kinase
(ERK), p38 and transcription factors AP-1 and
NF-κB to fit our model (Kawai et al, 1999).
Using the same model we simulated the profiles
of gene expressions of Il1
, Cxcl1, Tnf in
wildtype conditions (Fig. 2A-C). To test whether
our model can predict other experimental
conditions, we performed in silico MyD88 KO
conditions (shutting down the reaction TLR4 to
MyD88 in Fig.1). Fig. 2A-C, shows our
simulation does not recapitulate the experimental
outcome (Hirotani et al 2005) suggesting that our
model may exclude key signaling properties of
the TLR4 signaling.
Recently, there are reports suggesting crosstalk
between the MyD88-dependent and –independent
pathways although these have not yet been
demonstrated to occur in macrophages (Sato et
al., 2003; Selvarajoo, 2007). To test whether
these crosstalk also operate in macrophages, we
included them into our model (TRIF to TRAF-6
reaction and TRIF to TAB/TAK complex via
Receptor-Interacting Protein (RIP) 1 activation,
Model II) (Fig. 1B) and selected the parameters
values of the new reactions, without much
affecting other parts of the model, to fit the
wildtype profile of the three genes (Fig. 3A-C).
This time when we simulated the MyD88 KO
conditions, our simulation results matched the
experimental observation (Fig. 3A-C). To further
test the simulation capability of our revised
Model II, we performed the simulation of two
other experimental conditions, TRIF KO and
MyD88/TRIF double KO conditions and found
our Model II simulation consistent with the
experimental results. These results indicate that
the two crosstalk mechanisms are vital for TLR4
signaling in macrophages.
CONCLUSION
Recent high throughput microarray data analyses
have revealed that LPS stimulates the expression
of large number of genes, predominantly those
which encode proinflammatory cytokines. The
mechanisms by which these genes are temporally
regulated, however, is largely unknown. We
constructed dynamic in silico TLR4 models
constituting the MyD88-dependent and –
independent pathways and investigated the
experimental induction of MyD88-dependent
pathway genes Tnf, Il1
, Cxcl1 in wildtype and
several knock-out (KO) conditions. By fitting our
model with wildtype experimental data and
analysing with MyD88 KO, TRIF KO and
MyD88 and TRIF (double) KO conditions, we
infer that the crosstalk between TRIF to TRAF6
and TRIF to TAB/TAK complex via RIP1 is key
for the quantitative induction of MyD88-
dependent pathway genes Tnf, Il1
, Cxcl1 in
MyD88 KO conditions. Our integration of
simulation and experimental data provides a
novel perspective to dynamic TLR4 pathways
analysis.
WT
MyD88 KO
WT
MyD88 KO
Cxcl1
Il1
Tnf
Figure 2 Semi-quantitative simulation results of so-called MyD88-dependant genes A) Cxcl1, B) Il1β and C) Tnf, generated using Model I.
Blue line indicates wildtype (WT), pink line indicates MyD88 KO condition. The x-axis represents the simulation time in minutes and the y-
axis represents respective relative mRNA expression.
B
P
h
y
si
c
al
B
lo
c
ki
n
A
C
International Conference on Molecular Systems Biology 2008
112
REFERENCES
[1] Akira S, Takeda K, Kaisho T, Toll-like receptors: critical
proteins linking innate and acquired immunity. Nat
Immunol, 2, 675-80, 2001.
[2] Hirotani T et al, Regulation of lipopolysaccharide-inducible
genes by MyD88 and Toll/IL-1 domain containing adaptor
inducing IFN-beta. Biochem Biophys Res Commun,
11:328 , 383-392, 2005.
[3] Kawai T et al, Lipopolysaccharide stimulates the MyD88-
independent pathway and results in activation of IFN-
regulatory factor 3 and the expression of a subset of
lipopolysaccharide-inducible genes. J Immunol 167, 5887-
5894, 2001.
[4] Kawai T et al, Unresponsiveness of MYD88-Deficient Mice
to Endotoxin. Immunity. 11, 115-122, 1999.
[5] O'Neill LA, Targeting signal transduction as a strategy to
treat inflammatory diseases. Nat Rev Drug Discov. 5, 549-
563, 2006.
[6] Sato S, Sugiyama M, Yamamoto M, Toll/IL-1 receptor
domain-containing adap-tor inducing IFN-beta (TRIF)
associates with TNF receptor-associated factor 6 and
TANK-binding kinase 1, and activates two distinct
transcription factors, NF-қB and IFN-regulatory factor-3, in
the Toll-like receptor signaling. J Immunol. 171,4304-
4310, 2003.
[7] Selvarajoo K, Discovering differential activation machinery
of the Toll-like receptor 4 signaling pathways in MyD88
knockouts. FEBS Letters 580, 1457–1464, 2006.
[8] Selvarajoo K. ‘Decoding the signalling mechanism of Toll-
like receptor 4 signaling pathways in wildtypes and
knockouts’. In ‘E-Cell System: Basic Concepts and
Applications’ (edited by Ghosh S, Tomita M, Arjunan
SNV, Dhar PK. Kluwer Academic Publishers - Landes
Bioscience) (http://eurekah.com/chapter/3584), 2007.
Cxcl1
Il1
Time (min)
Time (min)
Tnf
Figure 3 Semi-quantitative simulation results of so-called MyD88-
dependant genes A) Cxcl1, B) Il1β and C) Tnf, generated using Model II.
Blue line indicates wildtype (WT), pink line indicates MyD88 KO
condition, green line indicates TRIF KO condition and cyan line indicates
MyD88/TRIF double KO condition. The x-axis represents the simulation
time in minutes and the y-axis represents respective relative mRNA
expression.
Time (min)
C
WT
MyD88 KO
TRIF KO
MyD88/TRIF DKO
WT
MyD88 KO
WT
MyD88 KO
TRIF KO
MyD88/TRIF DKO
B
P
h
y
si
c
al
B
lo
c
ki
n
g
T
h
e
o
r
y:
A
International Conference on Molecular Systems Biology 2008
113
