Supplementary Materials
1 Integration of multiple STAT pathways (Fig 1A main text)
This section provides more details regarding the interactions between multiple STAT pathways,
schematically represented in Fig 1A in the main text. Several STATs, for example STAT3 and
STAT5 (1, 2), lead to the production of IFN-γ and IL-10. The molecular mechanism of IFN-γ
and IL-10 production via STAT3 and STAT5 is as follows.
Extracellular IL-2, IL-6 and IL-21 bind to their complementary receptors. The receptors remain
in complex with JAKs. Binding of the interleukins to their receptors induces the
autophosphorylation of the receptors and the bound JAKs (3). The phosphorylated
receptor-JAK complexes can be dephosphorylated by SHP-1 (4). Phosphorylation of STAT3 is
performed by JAK in IL-2R:JAK and IL-6R:JAK complexes, while STAT5 is phosphorylated
by IL-2R:JAK and IL-21R:JAK (5-7). STAT3 and STAT5 are dephosphorylated by SHP-1 and
SHP-2 phosphatases respectively (8, 9). The phosphorylated STAT3 and STAT5 then can form
either homo- or hetero- dimers (5, 6). The dimers translocate into the cell nucleus and promote
transcription of genes responsible for IFN-γ and IL-10 production. Both IFN-γ and IL-10 are
degraded by various factors of dissociation such as the cytokine-receptor binding, diffusion and
cleavage by metalloproteases (10), which, in our model, we denote as Mp1 and Mp2 for IFN-γ
and IL-10, respectively.
Other STAT pathways, in addition to STAT3 and STAT5, can also lead to the induction of
IFN-γ and IL-10 production. The production of IFN-γ is induced by the following interleukins:
IL-12, IL-21, IL-2 and IL-35 (11-14). Interleukins IL-12 and IL-35 activate STAT4 through the
JAK-STAT pathway (6) while IL-21 and IL-2 activate STAT5 (15, 16). The production of anti-
inflammatory IL-10 is up-regulated by STAT1, STAT3 and STAT6 (2, 17, 18). In particular,
STAT1 is activated by IL-6 and IL-35; STAT3 by IL-2 and IL-6; STAT6 by IL-3 and IL-4 (6,
7, 19, 20).
2 Mathematical model
In this section, mathematical details of the model for STAT-STAT interactions are provided.
We derived the equations employed in the STAT3-STAT5 (Fig 1B), STAT3-STAT4 (Fig 1C)
and the combined STAT3-STAT4-STAT5 (Fig 5A) circuits. Table S1 shows the short names
and abbreviations used in our model. Table S1. Abbreviations used in the STAT phosphorylation model.
Abbreviation Meaning
I2 IL-2
RJ2 IL-2 Receptor complex with JAK
I2RJ2 IL-2 Receptor:JAK complex with bound IL-2
RpJ2 Phosphorylated IL-2 Receptor:JAK complex
P2 SHP-1 phosphatase
RpJ2P2 Phosphorylated IL-2 Receptor:JAK complex with SHP-1
I6 IL-6
RJ6 IL-6 Receptor:JAK complex
I6RJ6 IL-6 Receptor:JAK complex with bound IL-6
RpJ6 Phosphorylated IL-6 Receptor:JAK complex
P6 SHP-2 phosphatase
RpJ6P6 Phosphorylated IL-6 Receptor:JAK complex with SHP-2
I12 IL-12
RJ12 IL-12 Receptor:JAK complex
I12RJ12 IL-12 Receptor:JAK complex with bound IL-12
RpJ12 Phosphorylated IL-12 Receptor:JAK complex
P12 JAK phosphatase
RpJ12P12 Phosphorylated IL-12 Receptor:JAK complex with P12
phosphatase
I35 IL-35
RJ35 IL-35 Receptor:JAK complex
I35RJ35 IL-35 Receptor:JAK complex with bound IL-35
RpJ35 Phosphorylated IL-35 Receptor:JAK complex
P35 JAK phosphatase
RpJ35P35 Phosphorylated IL-35 Receptor:JAK complex with P35
phosphatase
I21 IL-21
RJ21 IL-21 Receptor:JAK complex
I21RJ21 IL-21 Receptor:JAK complex with bound IL-21
RpJ21 Phosphorylated IL-21 Receptor:JAK complex
P21 JAK phosphatase
RpJ21P21 Phosphorylated IL-21 Receptor:JAK complex with P21
phosphatase
S3 STAT3
RpJ2S3 Phosphorylated IL-2 Receptor:JAK complex with STAT3
RpJ6S3 Phosphorylated IL-6 Receptor:JAK complex with STAT3
S3p Phosphorylated STAT3
P3 SHP-1 phosphatase
P3S3p Phosphorylated STAT3 complex with SHP-1
S4 STAT4
S4p Phosphorylated STAT4
RpJ12S4 Phosphorylated IL-12 Receptor:JAK complex with STAT4
RpJ35S4 Phosphorylated IL-35 Receptor:JAK complex with STAT4
P4 PTP phosphatase
P4S4p Phosphorylated STAT4 complex with PTP phosphatase
S5 STAT5
S5p Phosphorylated STAT5
RpJ2S5 Phosphorylated IL-2 Receptor:JAK complex with STAT5
RpJ21S5 Phosphorylated IL-21 Receptor:JAK complex with STAT5
P5 SHP-2 phosphatase
P5S5p Phosphorylated STAT5 complex with SHP-2
phosphatase
S33 STAT3:STAT3 homodimer
S34 STAT3:STAT4 heterodimer
S44 STAT4:STAT4 homodimer
S35 STAT3:STAT5 heterodimer
S55 STAT5:STAT5 homodimer
Gg Gene responsible for IFN-γ production
S44Gg IFN-γ gene complex with STAT4:STAT4 homodimer
S55Gg IFN-γ gene complex with STAT5:STAT5 homodimer
S44S55Gg IFN-γ gene complex with STAT4:STAT4 and
STAT5:STAT5
Ig IFN-γ
Mp1 Metalloprotease that cleaves IFN-γ
IgMp1 Metalloprotease complex with IFN-γ gene
Ign Non-active IFN-γ
G10 IL-10 gene
S33G10 IL-10 gene complex with STAT3:STAT3 homodimer
Sp1 SP1 transcription factor
Sp1a SP1 transcription factor in active form
S33Sp1aG10 IL-10 gene complex with STAT3:STAT3 and Sp1a
Sp1aG10 Complex of Sp1a with IL-10 gene
Mp2 Metalloprotease that cleaves IL-10
I10Mp2 Metalloprotease complex with IL-10 gene
I10n Non-active IL-10
2.1 Model for the STAT3-STAT5 circuit
The biochemical reactions involved in the STAT3-STAT5 circuit (Fig 1B) are given by:
2
3
4 5
6
7 8
9
10 11
12
13 14
15
16
18
1
2
17
1
3
I2 RJ2 p
p P2 p P2 P2
I6 RJ6 p
p P6 p P6 P6
I2RJ2 R J2
R J2 R J2 RJ2
I6RJ6 R J6
R J6 R J6 RJ6
I21RJ21I21 RJ21 p
p P21 p P21 P21
R J21
R J21 R J21 RJ21
R J2 S3 R J 3p p p2S R
f
f
f
f
f
f
f
f
f
f
f
f
a
a
f
f
f
f
f
a
f
→←
→←
→←
→←
→←
→←
→←
→
→
→
→
→
→
+ →
+
+ +
+
+ +
+
+ +
4 5
6
7 8
9
10 11
12
13 14
15
17
18
20
21
2
16
1
2
9
J2 S3p
R J6 S3 R J6S3 R J6 S3p
P3 S3p P3S3p P3 S3
R J2 S5 R J2S5 R J2 S5p
R J21 S5 R J21S5 R J21 S5p
P5 S5p P5S5 P5 S5
S3 S3p S33
S
p p p
p p p
p p
3p S5p S35
S5 S5
p
p
p
p
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
→←
→←
→←
→←
→←
→←
→←
+
+ → +
+ → +
+ → +
+ → +
+ → +
+
+
+23
24
1
2
p
I2 Sp1
55
1
S
I2Sp
a
a
l
l
→←
→←+
(S1)
3
4
3
4
1
2
2
5
3 4
1
5
6
7
8
9
8
9
6
7
I2Sp1 CD46 Sp1a
CD46 Sp1 CD46Sp1
CD46Sp1 I2 Sp1a
S55 Gg S55Gg
S55Gg Ig
Ig Mp1 IgMp1 Ign Mp1
S33 G10 S33G10
S33G10 Sp1a S33Sp1aG10
Sp1a G10 Sp1aG10
Sp1aG10 S33 S
l
l
l
l
l
l
k
k
k
k
k
k
k
k
k
k
k
k
l
k
→←
→←
→←
→←
→←
→←
→←
→←
→←
+
+
+
+
+ +
+
+
→
→
+
+
6
6
6
10 11
12
33Sp1aG10
S33G10 I10
Sp1aG10 I10
S33Sp1aG10 I10
I10 Mp2 I10Mp2 I10n Mp2k
k
l
l
l
k→←
→
→
+→+
→
The system of reactions (S1) can be divided into three major subsystems of interactions: i)
Cytokine-receptor interactions, ii) STAT phosphorylation and dimerization, iii) Cytokine
production.
2.1.1 Cytokine-receptor interactions In the most general case the reactions can be written as follows:
2
3
4 5
6
1
C RJ
Rp
CRJ RpJ,
J P RpJP RJ P,
q
q
q
q
q
q
→←
→←
+
+ +
→
→
(S2)
where C is cytokine, RJ is Receptor:JAK complex, P is phosphatase and small p denotes
phosphorylated state.
The ODEs for the system (S2):
[ ] [ ][ ] ( )[ ]
[ ] [ ] [ ][ ] [ ]
[ ] [ ][ ] ( )[ ]
2 3
2 4 6
4 5 6
1 ,
,
.
C q q
R
d CRJ q RJ CRJdtd RpJ q CRJ q P q RpJPdtd RpJP q RpJ P
pJ
q q RpJPdt
+
+
= −
= − +
= −
(S3)
Corresponding conservation equations:
[ ] [ ] [ ] [ ][ ] [ ]
,
,T
T
R RJ RCRJ RpJ
P P
pJP
RpJP
+ +
= +
= + (S4)
where TR and TP are the total amounts of receptors and phosphatase, respectively. Here we
neglected STAT-receptor interactions since STAT proteins do not have a significant impact on
receptor dephosphorylation.
Equations (S4) can be written as follows:
,,T
T
PR
pα β γ ω
ω
+ +
=
+
+
= (S5)
where [ ] [ ] [ ] [ ] [ ] [ ] , , , , , .RJ CRJ RpJ RpJP p P c Cα β γ ω= = = = = =
The ODEs (S3) can be rewritten in the following way:
( )
( )
2 3
2 4 6
6
1
4 5
,
,
.
d qdtd q q p qdtd
c q
q
q
q qpdt
β α β
γ β γ ω
ω γ ω
= −
= − +
+
+= −
(S6)
We need to find steady-state solutions of Equations (S6):
( )
( )
2 3
2 4 6
4 5 6
10 ,0 ,0 ,
,.
T
T
qq q p qc q q
q qq
P pR
p
α β
β γ ω
γ
α β γ ω
ω
ω
= −
= − +
= −
+ +
=
+
=
+
+
+
(S7)
We found the concentrations of the complexes:
( )
( )
5 6 2
4
2 3 2
5 5
2 2
2 3 1
1 3 2
,
,
,
TT
T T
T
PP
qq q
q q Q
Q Q Q
Q
P Pq qq q Q Pq c Qc
γ γω
γγ
γ γβ ω
γ γ
β γα
γ
=
= = =
= =
=+ ++
+ +
++
(S8)
where 2 31
1qq qQ +
= and 5 62
4qq qQ +
= are the Michaelis constants for phosphorylation and
dephosphorylation, respectively, and 23
5
Q qq
= .
We can write the following equation using the conservation Equation for the receptor (S5):
1
2 3 3
1 1 .TT
QRQ Q QP
cγ
γγ
+ +⎞⎛
= + ⎟⎜+ ⎝ ⎠ (S9)
As a result, we obtained a quadratic equation: 20 ,γ γ δχ+= − (S10)
where 12
3 3
1 1TTQQ RQc Q
Pχ⎞⎛
= +− + ⎜⎝
+ ⎟⎠, 2TR Qδ = .
The solution of Equation (S10) is:
2 42 2
χ δγ
χ+
+= − (S11)
Equation (S11) can be rewritten as follows:
21 4 ,2 c c
ρ ργ λ δ λ
⎞⎛ ⎞⎛ ⎟⎜ + + − −⎜ ⎟⎜ ⎟⎝ ⎠⎝ ⎠= (S12)
where 23
1 1T TQ R PQ
λ =⎞⎛
+ ⎟⎜⎝
+⎠
− and 1
3
TP QQ
ρ = .
Using Equation (S12) we can now write for [ ]2RpJ ,[ ]6RpJ and [ ]21RpJ in non-dimensional
form respectively:
[ ][ ]
[ ]
[ ][ ]
[ ]
[ ][ ]
1
1 1
2
1
1 1
3
2 2
2
3
2 2
5
2
2 2
4
4 4
36
3
1 12
22
1 1 42
,2
1 16
62
1 1 4
2 2
2 2 2
6 6
6 6 6
21
6,
2
1 12
21
211
2
t t
t t t
t t
t t t
t t
M r p
M r p r M
M r p
M r
Mn i n
w
Mn i n
Mn i n
w
Mn i n
Mn i
M
wM r p
n
p r
⎞⎛+ + ⎟⎜
⎝ ⎠= − +
⎞⎛ ⎞⎛+ + +⎟⎜ ⎟⎜⎜ ⎟⎝ ⎠⎝ ⎠+
⎞⎛+ + ⎟⎜
⎝ ⎠= − +
⎞⎛ ⎞⎛+ + +⎟⎜
− +
− +
− +
− + ⎟⎜⎜ ⎟⎝ ⎠⎝ ⎠+
⎞⎛+ + ⎟⎜
⎝=
−⎠
+
− +
[ ]5
6 6
2
3 3
1 1 421
,
21 21 2
2
1t t tM
n i nM r p r M
⎞⎛ ⎞⎛+ + +⎟⎜ ⎟⎜⎜ ⎟⎝
− +⎠⎝ ⎠+
(S13)
where
[ ] [ ] [ ] [ ] [ ] [ ]
[ ] [ ] [ ] [ ]
[ ] [ ]
2 3 5 6 22 1
1 4 5
8 9 11 12 83 4 2
7 10 11
12 62 2, , ,3 3 3 3 3
216 6 , ,3 3 3 3
213
2 2 2 , 2 , 2 , , , 6
3 36
6 , 6 , 6 , , , 213 3
2121 , 21
T T
T T T T T T T
T T
T T T
t t
t t
t
T T T
T
T
RpJi w r p M M n i
f S f S fRpJ
w r p M M n if S f
I IR
S f
P f f f f fS S S S S
IR P f f f f
RpJ
fS S S S
RS
w r
= = = = = =
= = = = = =
= =
+ += =
+ +=
14 15 17 18 145 6 3
13 16 17
, 21 ,21 .3 3
, ,3 3
T
T T T Ttp M M n
f SP f f f f f
S S f S f+
= = =+
=
2.1.2 STAT phosphorylation and dimerization The ODEs describing biochemical reactions in the STAT subsystem:
[ ] [ ][ ] ( )[ ]
[ ] [ ][ ] ( )[ ]
[ ] [ ] [ ] [ ][ ] [ ]
[ ] [ ] [ ][ ] [ ]
[ ] [ ][ ] ( )[ ]
[ ] [ ][ ] ( )[ ]
1 2 3
2
1
4 5 6
5 7 9
220 21 22
7 8 9
10 11 1
9
2
2 3 2 3 2 3 ,
6 3 6 3 6 3 ,
3 2 3 6 3 3 3 3 3
2 3 2 33 3 5 35 ,
3 3 3 3 3 3 ,
2 5 2 5 2 5 ,
p p
p p p
d RpJ S a RpJ S a a RpJ Sdtd RpJ S a RpJ S a a RpJ Sdtd S a RpJ S a RpJ S a P S p a P Sdt
a S a S a S S a Sd P S a P S a P Sdtd RpJ S a RpJ S a a RpJ Sdt
p p a p
= − +
= − +
= + − + −
− + − +
= − +
= − +
[ ] [ ][ ] ( )[ ]
[ ] [ ] [ ] [ ][ ] [ ]
[ ][ ] [ ] [ ] [ ]
[ ] [ ][ ] ( )[ ]
[ ] [ ] [ ]
[ ] [ ][ ] [ ]
13 14 15
11 14 18
21 22 23 24
17 18
20
21
16
2
1
22
6
219
21 5 21 5 21 5 ,
5 2 5 21 5 5 5 5 5
3 5 35 2 5 2 55 ,
5 5 5 5 5 5 ,
33 3 33 ,
35 5 35 ,
5
3
d RpJ S a RpJ S a a RpJ Sdtd S a RpJ S a RpJ S a P S a P Sdt
a S S a S a S p a Sd P S a P S a a P Sdtd S a S p a Sdtd S a S p a Sdtd Sdt
p p p
p p
p p p
S p
= − +
= + − + −
− + − +
= − +
= −
= −
[ ] [ ] [ ]223 245 5 55 .a S p a S= −
(S14)
Conservation equations are given by:
[ ] [ ] [ ] [ ] [ ] [ ][ ] [ ] [ ] [ ] [ ] [ ][ ][ ]
3 3 3 2 33 35 2 3 6 3 [ 3 3 ],
5 5 5 2 55 35 2 5 21 5 [ 5 5 ],
3 3 [ 3 3 ],
5 5 [ 5 5 ].
T
T
T
T
S S S p S S RpJ S RpJ S P S
S S S p S S RpJ S RpJ S P S
P P P S
P P P S
p
p
p
p
= + + + + + +
= + + + + + +
= +
= +
(S15)
Then we can normalize Equations (S15) to 3TS :
[ ] [ ] [ ] [ ] [ ] [ ] [ ][ ] [ ] [ ] [ ] [ ] [ ] [ ][ ] [ ][ ] [ ]
1 3 3 2 33 35 2 3 6 3 3 3
5 5 5 2 55 35 2 5 21 5 5 5
3 3
,
,
3 3
5 5 5 ,
,
5
t
t
t
s s s s w s w s p s
s s s s s w s w s p s
p p p
p p
p
s
p p p
p
s
p
p
= + + + + + +
= + + + + + +
= +
= +
(S16)
where
[ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ]3 3 2 3 6 3 3 3 53 , 3 , 2 3 , 6 33 3 3 3
, 3 3 ,3
,3
5 T
T T T T T Tt
p pp p
SS S RpJ S RpJ
S SS P S Ss s w s w s p s s
S S S= = = = = =
[ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ]5 5 2 5 21 5 5 5 5 , 5 , 2 5 , 21 5 ,
3 35
35 ,
3 3T T T T T
S S RpJ S Rpp J S P Ss s w s w
pp p
S S S S Ss p s= = = = =
[ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ]33 35 55 3 533 , 35 , 5
3 3 3 3, , 5 ,
35 3
T T T T TS S S SS S S P P
s s s p pS
= = = = = 33 5 .
33 , 5T T
Tt
Tt
P Pp pS S
= =
The ODEs (S14) can be written in non-dimensional form as follows:
[ ] [ ][ ] [ ]
[ ] [ ][ ] ( )[ ]
[ ] [ ] [ ] [ ][ ] [ ]
[ ] [ ] [ ][ ] [ ]
[ ] [ ][ ] ( )[ ]
[ ] [ ][ ] ( )[ ]
[ ] [ ][ ] ( )
1
3 4 5
2 4 6 8
18 19 20 21
6 7 8
9 10 11
12 1 1
2
3 4
2 3 2 3 2 3 ,
6 3 6 3 6 3 ,
3 2 3 6 3 3 3 3 3
2 3 2 33 3 5 35 ,
3 3 3 3 3 3 ,
2 5 2 5 2 5 ,
21 5 21 5 21
d w s m w s w sdd w s m w s m m w sdd s p m w s m w s m p s m p sd
m s p m s m s p s p m sd p s p m p s p m m p s pdd w s m w s m m w sdd w s m
p p
w s m m wd
τ
τ
τ
τ
τ
τ
= −
= − +
= + − + −
− + − +
= − +
= − +
= − + [ ]
[ ] [ ] [ ] [ ][ ] [ ]
[ ][ ] [ ] [ ] [ ]
[ ] [ ][ ] ( )[ ]
[ ] [ ] [ ]
[ ] [ ][ ] [ ]
[ ] [ ] [ ]
10 13 15 17
16 17
2
220 21 22 23
15
18
20 21
222 23
19
5 ,
5 2 5 21 5 5 5 5 5
3 5 35 2 2 55 ,
5 5 5 5 5 5 ,
33 3 33 ,
35 3 5 35 ,
55 5 5
5
5 ,
s
d s m w s m w s m p s m p sd
m s s m s m m sd p s p m p s m m p sdd s m s m sdd s m s p s
p p p
p p s p
p
p m sdd s m
p
m
p
sd
ps
τ
τ
τ
τ
τ
= + − + −
− − +
= − +
= −
= −
−
+
=
(S17)
where
( ) 12 3 1
2 3 2 3 2 3 2 3 2 3
2 3 2 3 2
2 4 5 62 3 4 5
7 8 9 10 11 126 7 8 9 10 11
13
3 2 3 2 3 2 3
2 3
14
2
1512 13 14
3 2
, 3 , , 3 , , ,
3 , , , 3 , , ,
3 , ,
T T
T T
T
t a a m S m m S m ma a a a a a a a a a
m S m m m S m ma a a a a a a a a a a a
m
a a a a a
a a a a a a
a S m ma a a
a aa a
τ = + = = = = =+ + + + +
= = = = = =+ + + + + +
= = =+ + +
16 17 1815 16 17
19 20 21 22 23 2418 19 20 21
3
22 23
2 3 2 3 2 3
2 3 2 3 2 3 2 3 2 3 2 3
, 3 , , ,
3 , , 3 , , 3 , .
T
T T T
m S m ma a a a a a a
m S m m S m m S ma a a a a a a a
a a a
a a a a a aa a a a
= = =+ + +
= = = = = =+ + + + + +
We need to find steady-state solutions of Equations (S17):
[ ][ ] [ ][ ][ ] ( )[ ]
[ ] [ ] [ ] [ ][ ] [ ]
[ ] [ ] [ ][ ] [ ][ ][ ] ( )[ ][ ][ ] ( )[ ][ ][ ] ( )[ ]
[ ] [ ] [ ] [ ]
1
3 4 5
2 4 6 8
18 19 20 21
6 7 8
9 10 11
12 13 14
10 1
2
13 5
0 2 3 2 3 ,
0 6 3 6 3 ,
3 2 3 6 3 3 3 3 3
2 3 2 33 3 5 35 ,
0 3 3 3 3 ,
0 2 5 2 5 ,
0 21 5 21 5 ,
5 2 5 21 5 5
m w s w s
m
p
w s m m w sd s p m w s m w s m p s m p sd
m s p m s m s p s p m s
m p s p m m p s p
m w s m m w s
m w s m m w sd s m w s m w s m p
p
dp
τ
τ
= −
= − +
= + − + −
− + − +
= − +
= − +
= − +
= + − [ ] [ ]
[ ][ ] [ ] [ ] [ ][ ][ ] ( )[ ][ ] [ ][ ][ ] [ ][ ] [ ]
220 21 22 23
15
18
20 21
1
2
7
16 17
219
22 23
5 5 5
3 5 35 2 2 55 ,
0 5 5 5 5 ,
0 3 33 ,
0 3 5 35 ,
0 5 55 .
5
s m p s
m s s m s m m s
m p s m m p s
m s m s
m s p s p m s
m s m
p p
p p s p
p
p
s
p
p
+ −
− − +
= − +
= −
= −
= −
+
(S18)
Equations (S18) can be simplified as follows:
[ ][ ] [ ][ ][ ] ( )[ ]
[ ] [ ] [ ] [ ]
[ ][ ] ( )[ ][ ][ ] ( )[ ][ ][ ] ( )[ ]
[ ] [ ] [ ] [ ]
[ ][ ] ( )[ ][ ]
1
3 4 5
2 4 7
6 7 8
9 10 11
12 13 14
16
16 17
2
10 13
15
8 191
0 2 3 2 3 ,
0 6 3 6 3 ,
3 2 3 6 3 3 3 ,
0 3 3 3 3 ,
0 2 5 2 5 ,
0 21 5 21 5 ,
5 2 5 21 5 5 5 ,
0 5 5 5 5 ,
0 3 3
m w s w s
m w s m m w sd s p m w s m w s m p sdm p s p m m p s p
m w s m m w s
m w s m m w sd s m w s m w s m p s
p
p p
p p
p
dm p s m m p s
m s m s
τ
τ
= −
= − +
= + −
= − +
= − +
= − +
= + −
= − +
= − [ ][ ][ ] [ ][ ] [ ]
20 21
222 23
3 ,
0 3 5 35 ,
0 5 55 .p
m s p s p m s
m s m s
= −
= −
(S19)
Then using conservation Equations (S16) and System (S19) we obtained:
[ ] [ ] [ ] [ ]
[ ] [ ] [ ] [ ]
2 4 7
10 1 63 1
3 2 3 6 3 3 3 ,
5 2 5 21 5 5 5 ,
pd s p m w s m w s m p sdd s m w s m w s sp pm pd
τ
τ
= + −
= + −
(S20)
where
[ ] [ ]
[ ]
[ ] [ ]
[ ]
[ ] [ ]
[ ] [ ][ ]
[ ] [ ]
[ ] [ ][ ]
[ ] [ ][ ]
[ ] [ ][ ]
[ ] [ ][ ]
[ ] [ ] [ ] [ ] [ ]
[ ]
7 8
6
16 17
15
18
20
21
22
23
3
4 5
9
10 11
12
13 14
2
9
3
1
2
1
1
33 3 3 ,
3
55 5 5 ,
5
33 3 ,
35 3 5 ,
55 5 ,
2 3 2 3 ,
6 3 6
2 5 2
21 5 21
3 2 33 35 3 33
1 2
3 ,
5 ,
5 ,
1
t
t
s pp s p m m s p
ms p
p s p m m s pm
ms s pmms s s pmms s pm
w s m wmw s w
m mmw s w
m mmw s w
m ms s s p s
s
p
p
p
s
s
s
s
p pmm w
=+
+
=+
+
=
=
=
=
=+
=+
=+
− − −=
+ +
−
[ ]
[ ] [ ] [ ] [ ] [ ]
[ ] [ ]
4 5
9 12
10 11 13 14
,6
5 5 2 55 35 5 55 .
1 2 21
t
wm m
s s s s p ss m mw w
m m m m
p p+
− − −
+
−=
+ ++
(S21)
We can rewrite Equations (S21):
[ ] [ ][ ]
[ ] [ ][ ]
[ ] [ ]
[ ] [ ][ ]
[ ] [ ]
[ ] [ ][ ]
[ ] [ ][ ]
[ ] [ ][ ]
[ ] [ ][ ]
[ ] [ ] [ ] [ ] [ ][ ] [ ]
[ ] [ ] [ ] [ ] [ ][ ] [ ]
9
12
2
13
142
15
7
8
10
11
7 8
10
33 3 3 ,
3
55 5 5 ,
5
333 ,
3 535 ,
555 ,
2 32 3 ,
66 3 ,
22 5 ,
2121 5 ,
3 2 33 35 3 33 ,
2 61
5 5 2 55 35 5 55
2 211
3
5
5
1
t
t
t
s pp s p
M s p
s pp s p
M s p
s ps
Ms s p
sM
s ps
Mw
w sM
ww s
Mw
w sMw
w sM
s s s p ss
w wM M
s s s
p
p
p
s
s
s
s
p p
p s p ss
w wM
p
=+
=+
=
=
=
=
=
=
=
− − −=
+ +
−
−
−
+ +
− −=
11
,
M
(S22)
where we denoted the Michaelis constants
2 3 5 6 8 9 11 12 14 15 17 187 8 9 10 11 12
1 4 7 10 13 16
20 22 2413 14 15
19 21 23
3 3 3 3 3, , , , , ,
, , .
3
3 3 3
T T T T T T
T T T
a a a a aM M M M M Ma S a S a S a S a S a Sa a aM M Ma S a
a a a a a a a
S a S
= = = = = =
= = =
+ + + + + +
We look for steady-state solutions of System (S20):
[ ] [ ] [ ][ ] [ ] [ ]
2 4 7
10 13 16
0 2 3 6 3 3 3 ,
0 2 5 21 5 5 5 .
m w s m w s m p s
m
p
pw s m w s m p s
= + −
= + − (S23)
We can rewrite Equations (S23) as follows:
[ ] [ ] [ ][ ] [ ] [ ]4 5
6 7
0 2 3 6 3 3 3 ,
0 2 5 21 5 5 5 ,
n w s n w s p s
n w s n w s ps
p
p
= + −
= + − (S24)
where 2 5 11 144 5 6 7
8 8 17 17
, , ,a a a an n n na a a a
= = = = .
System (S24) can be rewritten after substituting solutions from Equation (S22):
[ ] [ ] [ ][ ]
[ ][ ]
[ ] [ ][ ] [ ]
[ ] [ ] [ ][ ]
[ ][ ]
[ ] [ ][ ] [ ]
2
13 14
7 8 8 7
9 8 7
2
15 14
10 11 11 10
12 11 10
4 5
6 7
3 3 50 3 2
3 2 63 1 1,
3 2 6
5 3 50 5 2
5 2 215 1 5 ,
5 2 21
t
t t
s p s s ps p
M M
s p M M M w M wpM s p n M w n M w
s p s s ps p
M M
s p M M M w M wp sM s p n M w n M w
p
p
= + + +
⎞⎛ + ++ + −⎟⎜
+ +⎝ ⎠
= + + +
⎞⎛ + ++ + −⎟⎜
+ +⎝ ⎠
(S25)
We solved System (S25) for and numerically.
2.1.3 Cytokine production In general case, transcription factor T can activate gene G by forming a complex with the gene
TG:
(S26)
The ODEs for Equation (S26):
(S27)
Conservation equation that follows from Equations (S27):
(S28)
where is the total concentration of the gene.
Equation (S28) can be written as follows:
(S29)
where .
[ ]3s p [ ]5s p
2
1
G TGTh
h
→←+
[ ] [ ][ ] [ ]
[ ] [ ][ ] [ ]
1
1
2
2
,
.
d G h G h TGdtd
T
TTG h G h TGdt
= − +
= −
[ ] [ ],TG GG T= +
TG
,TG λ θ= +
[ ] [ ] ,G TGλ θ= =
The ODEs (S27) can be rewritten in the following way:
(S30)
where .
We need to find steady-state solutions of Equation (S30):
(S31)
We can find from Equations (S31):
(S32)
where is the Michaelis constant.
In the most general case the reactions of the activation of a gene G by two transcription factors
T1 and T2 are:
(S33)
The ODEs for System (S33):
(S34)
1
21
2 ,
,
d h hdtd h hdt
T
T
λ λ θ
θ λ θ
= − +
= −
[ ]T T=
210 ,.T
TGh hλ θ
λ θ
−
=
=
+
θ
,TTG
Qh Tθ =
+
2
1
hQhh
=
2
3
4
1
3
4
1
2
T1 G
T1G T2 T1T2G
T
T1G
T22 G
T2G
G
T1 T1T2G
b
b
b
b
b
b
b
b
→←
→←
→←
→←
+
+
+
+
[ ] [ ][ ] [ ] [ ][ ] [ ]
[ ] [ ][ ] [ ] [ ][ ] [ ]
[ ] [ ][ ] [ ] [ ][ ] [ ]
[ ] [ ][ ] [ ] [ ][ ] [ ]
2 4
2 3 4
1 3
3 4 1 2
1
3 4 1 2
1 2 ,
1 1 2 1 2 ,
2 2 1 1 2 ,
1 2 2 1 2 1 1
1 2
1 1
2 2
1 2 2 .
d G b G b T G b G b T Gdtd T G b G b T G b T T T Gdtd T G
T T
T T G b
T T G b
T G
b G b T G b T T T Gdtd T T G b T T T G b T T T Gb T G bdt
= − + − +
= − − +
= − − +
= − + −
Conservation equation that follows from Equations (S34):
(S35)
where is the total amount of the gene.
Equation (S35) can be written as follows:
(S36)
where
The ODEs (S34) can be rewritten in the following way:
(S37)
where .
We found steady-state solutions of Equations (S37):
(S38)
Next, we found concentrations of the complexes:
[ ] [ ] [ ] [ ]1 2 ,1 2T T G TG TGG T G= + + +
TG
,TG ψ ξ ϕ ν+ + +=
[ ] [ ] [ ] [ ], 1 , 2 , 1 2 .G T G T G T T Gψ ξ ϕ ν= = = =
2 4
2 3 4
3 4 1 2
3 4
1 3
1 2
1
,
2 ,
1 2
,
1 ,
2 1
1
2
d b b b bdtd b b b Tdtd b b b Tdt
T T
T b
T b
b bd b T b Tdt
ψ ψ ξ ψ ϕ
ξ ψ ξ ν
ϕ ψ
ξ
ϕ
ξ
ϕ ν
ν ν νϕ
= − + − +
= − − +
= − − +
= − + −
[ ] [ ]1 1 , 2 2T T T T= =
2 3 4
3 4 1
1
2
3 4 1 2
0 2 ,0 1
12 ,
0 2 1 ,.T
b b b Tb b b Tb T b
Tb
bG
T
bT
b
ψ ξ ν
ψ ϕ ν
ν ν
ν
ξ
ϕ
ξ ϕ
ψ ξ ϕ
= − − +
= − − +
=
+
−
+=
− +
+
(S39)
where are the Michaelis constants.
If a protein is activated by the first and the second transcription factors at the same time, then
its concentration is proportional to the concentration only:
(S40)
which is a probability of the two transcription factors to be bound to the same gene.
If a protein is activated by the first or the second transcription factors, then it is proportional to
the sum of concentrations :
(S41)
which is a probability of either of the two transcription factors to be bound to the same gene.
( )( )
( )( )
( )( )
4
23
1 22 4
1 3
2
11
1 22 4
1 3
1 22 4
1 3
11
1 21 2
22
1 21 2
1 2 1 2 ,1 2
1
,
2
,
T T
T T
T T
b TQb TbG G
Qb T Qb Tb bT Tb b
b TQbTbG G
Qb T Qb Tb bT Tb b
T T T TG GQb T Qb Tb bT T
b b
ξ
ϕ
ν
= =⎞ + +⎞⎛⎛
+ + ⎟⎟⎜⎜⎝ ⎝⎠ ⎠
= =⎞ + +⎞⎛⎛
+ + ⎟⎟⎜⎜⎝ ⎝⎠ ⎠
= =⎞ + +⎞⎛⎛
+ + ⎟⎟⎜⎜⎝ ⎝⎠ ⎠
1 31
42
2 ,bQb Qb bb b
= =
ν
1 2
1 2 ,1 2T
T TGQb T Qb T
ν = ⋅+ +
ξ ϕ ν+ +
( )( )
( )( )( ) ( )
( )( )
( )( )
2 1
1 2
2 1
1 2
2 1
1 2
1 2 1 2
1 2
1 2 1 21 2
1 2 1 2 1 2 1 2 ,1 2
1 2 2 1 1 2,
1 2
1 2 1 2 ,1 2 1 2
1
,
21
T
T
T
T
T
Qb T QbT T TGQb T Qb TQb T QbT T T T T T TG
Qb T Qb T
T Qb T T Qb T T TG
Qb T Qb T
T T T TGQb T Qb T Qb T Qb T
T TGQb T Qb
ξ ϕ ν
ξ ϕ ν
ξ ϕ ν
ξ ϕ ν
ξ ϕ ν
+ ++ + =
+ +
+ + + −+ + =
+ +
+ + + −+ + =
+ +
⎞⎛+ + = + − ⎟⎜ ⎟+ + + +⎝ ⎠
+ + = ++ + 1 2
1 2 ,2 1 2
T TT Qb T Qb T
⎞⎛− ⎟⎜ + +⎝ ⎠
CD46 The full mechanism of how CD46 enhances IL-10 production is still not clear. We assumed
here that the mechanism of reactions is similar to the one described in Equation (S33). Thus it
can be written for the concentration of SP1 in non-dimensional form according to Equation
(S40):
(S42)
where , , , , and .
We assumed here that the gene interaction with the transcription factor and its subsequent
expression lead to the mRNA translation and certain cytokine secretion. The produced cytokine
then can be degraded by a metalloprotease. In this case the biochemical reactions can be written
as follows:
(S43)
where TG is the transcription factor complex with gene, C is the active cytokine, Mp is the
metalloprotease, CMp is cytokine-metalloprotease complex and Cn is a non-active cytokine.
The ODEs for the reactions in System (S43):
(S44)
Conservation equation that follows from Equations (S44):
(S45)
We found steady-state solutions of System (S44):
[ ] [ ][ ]
[ ][ ]16 17
2 461 1 ,
2 46ti cd
sp a spM i M cd
= ⋅+ +
[ ] [ ]113T
SP asp a
S=
113T
tT
SPspS
= [ ] [ ]46463T
CDcd
S= 2
161 3T
lMl S
= 417
3 3T
lMl S
=
1 2
3
TG C,
C Mp CMp Cn Mp,u
u
KP
u→← →+ +
→
[ ] [ ] [ ][ ] [ ]
[ ] [ ][ ] ( )[ ]
[ ] [ ][ ] ( )[ ]
1 3
1 2 3
1 2 3
,
,
.
d C KP TG u C Mp u CMpdtd CMp u C Mp u u CMpdtd Mp u C Mp u u CMpdt
= − +
= − +
= − + +
[ ] [ ]TMp Mp CMp= +
(S46)
Thus, we obtain a system of equations:
(S47)
where
We can find from this system of Equations (S47):
, (S48)
where is the Michaelis constant.
We next substituted from equation (S48) to the first equation in System (S47):
(S49)
Since C should be positive and the maximum value for is 1, , which implies that
the rate of the cytokine production should be less than its maximum rate of the degradation by
metalloprotease.
Equation (S49) can be written as follows:
(S50)
[ ] [ ][ ] [ ][ ][ ] ( )[ ]
1 3
1 2 3
0 ,
0 .
KP TG u C Mp u CMp
u C Mp u u CMp
= − +
= − +
( )2
1 2 3
0 ,0 ,
,T
KP uu C u u
Mp
ϑ σ
ζ σ
ζ σ
= −
= − +
= +
[ ] [ ] [ ] [ ],, , .C C CG MT Mp pζ σϑ = = ==
σ
TCMp
Qu Cσ =
+
2 3
1
u uQuu+
=
σ
2
2
2
0 ,
,
.1
T
T
T
CKP u MpQu C
KP QuCu Mp KPQuC u Mp
KP
ϑ
ϑϑ
ϑ
= −+
=−
=−
TMpϑ 2KP u<
,1T
QuC MpQpϑ
=−
where , .
When the cytokine production is up-regulated by one transcription factor only, from
Equation (S32) and thus it can be written:
(S51)
If the cytokine production is up-regulated by two transcription factors at the same time,
as shown in Equation (S40), it can be written:
(S52)
If the cytokine production is up-regulated by either of the two transcription factors,
as shown in Equation (S41), and thus it can be written:
(S53)
IFN-γ and IL-10 production Since IFN-γ is activated by STAT55 only (Fig 1B) we can write using Equation (S51):
(S54)
where .
IL-10 gene can be activated by either STAT33 or CD46. Thus, it can be written according to
Equation (S53):
2
KPQpu
= 1Qp <
ϑ θ=
.1T
TTG
Qh
QuC Mp
QpT+
=−
ϑ ν=
1 2
21
.11
2
T
T
QuC
T TGQb T Qb T
Mp
Qp
=
+ +
−
ϑ ξ ϕ ν= + +
1 2 1 2
1 2 1 21 2
.
2
1
1
T
TT T T TG
Qb T Qb T Qb
QuC Mp
QT b T
pQ
=
⎞⎛+ − ⎟⎜ + +⎝ ⎠
−
+ +
[ ]
[ ][ ]
18
819
555
1
5
,1
t
t
Mig mp
ns
gsM
g+
=−
[ ] [ ] 4 5 2 518 19 8 8
3 1 4
1, 1 , , , , , 13 3 3 3 3
T Tt t
T T T T T
Ig Mp Gg k k k lig mp gg M M n nS S S k S k S k
+= = = = = = <
(S55)
where .
2.2 Model for the STAT3-STAT4 circuit
The biochemical reactions involved in the STAT3-STAT4 circuit (Fig 1C) are as follows:
[ ]
[ ][ ]
[ ][ ]
[ ][ ]
[ ][ ]21
20
22 21 229
1
33 1 33 110
33 1
,2
3
0
1
1
3
t
ts sp a s sp a
M s M sp a M s M
Mi mp
n gsp a
⎞⎛+ − ⎟⎜
+ + + + ⎠
−
⎝
=
11 12 7 9 620 21 22 9 9
10 6 8 11
2 102 , 10 , , , , , 13 3 3 3 3
t tt t
T T T T T
Mp G k k k k lmp g M M M n nS S k S k S k S k
+= = = = = = <
2
3
4 5
6
7 8
9
10 11
12
13 14
15
16 1
18
1
7
19
21
I2 RJ2 p
p P2 p P2 P2
I6 RJ6 p
p P6 p P6 P6
I2RJ2 R J2
R J2 R J2 RJ2
I6RJ6 R J6
R J6 R J6 RJ6
I12RJ12 R J12
R J1
I12 RJ12 p
p P12 R J12 RJ12
I
2 p P1
35RJ
2 P12
I35 RJ35 35
f
f
f
f
f
f
f
f
f
f
f
f
f
f
f
f
f
f
f
f
→←
→←
→←
→←
→←
→←
→←
+
+ +
+
+
→
→
→
→
→
+
+
+ +
+
→
→20
R 5pJ3f
(S56)
22 23
24
4 5
6
7 8
9
10 11
12
13 14
1
1
3
5
2
R J35 R J35 RJ35
R J2 S3 R J2S3 R J2 S3p
R J6 S3 R J6S3 R J6 S3p
P3 S3p P3S3p P3 S3
R J12 S4 R J12S4 R J12 S4p
R J35 S4
p P35 p P35 P35
p
R J35S4 R J35 S4p
P4 S4p
p p
p p p
p p p
p p p
f
a
a
a
a
a
a
f
a
a
a
a
a
a
a
f
a
a
→←
→←
→←
→←
→←
→←
+ +→
+ → +
+ → +
+ → +
+ → +
+ → +
+17
18
20
21
22
23
24
1
2
3
4
3
4
1
2
2
16
19
1
5
3
5
p
p
p p
I2 Sp1 I2Sp1
I2Sp1 CD46 Sp1a
CD46 Sp1 CD46Sp1
CD46Sp1 I2 Sp1a
S44 Gg S44Gg
S
P4
44Gg I
S4 P4 S4
S3 S3p S33
S3p S4p S34
S4 S4 S44
g
Ig Mp1 IgMp1
a
a
l
l
l
l
l
l
l
l
k
k
k
k
a
a
a
a
a
a
k
a
l
→←
→←
→←
→←
→←
→←
→←
→←
→←
→←
→
+
+
+
+
+
+
+
+
+
+
→
→4
6
7
Ign Mp1
S33 G10 S33G10k
k
→←
+
+
2.2.1 Cytokine-receptor interactions According to Equations (1.12) we can write for , , and in
non-dimensional form respectively:
8
9
8
9
6
7
6
6
6
10 11
12
S33G10 Sp1a S33Sp1aG10
Sp1a G10 Sp1aG10
Sp1aG10 S33 S33Sp1aG10
S33G10 I10
Sp1aG10 I10
S33Sp1aG10 I10
I10 Mp2 I10Mp2 I10n Mp2
k
k
k
k
k
k
k
k
l
l
l
k
→←
→←
→←
→←
→
→
→
+
+
+
+ +→
[ ]2RpJ [ ]6RpJ [ ]21RpJ [ ]35RpJ
[ ][ ]
[ ]
[ ][ ]
[ ]
1
1 1
2
1
1 1
3
2
2
2 2
2
42
3
24 4
2
1 12
22
1 1 42
,21 1
66
2
1 1 46
2 2
2 2 2
6 6
6 6 6
,2
t t
t t t
t t
t t t
Mn i n
w
Mn i n
Mn i n
w
M r p
M r p r M
M r p
M ri
r MMpn n
⎞⎛+ + ⎟⎜
⎝ ⎠= − +
⎞⎛ ⎞⎛+ + +⎟⎜ ⎟⎜⎜ ⎟⎝ ⎠⎝ ⎠+
⎞⎛+ + ⎟⎜
⎝ ⎠= − +
⎞⎛ ⎞⎛+ + +⎟⎜ ⎟⎜⎜ ⎟⎝ ⎠
− +
− +
−
⎠
+
+⎝+
−
(S57)
where
2.2.2 STAT phosphorylation and subsequent dimerization ODEs for the STAT phosphorylation and dimerization module are given by:
[ ][ ]
[ ]
[ ][ ]
[ ]
5
3 3
2
5
3 3
7
4 4
2
7
6
84
6 6
8
48
12 12
12 12 1
1 112
122
1 1 412
,2
1 135
352
1 1 43
2
35 35
35 5 35
5
,2
3
t t
t t t
t t
t t t
Mn i n
w
Mn i n
Mn i n
w
M r p
M r p r M
M r p
M r p r MMn i n
− +
− +
− +
− +
⎞⎛+ + ⎟⎜
⎝ ⎠= − +
⎞⎛ ⎞⎛+ + +⎟⎜ ⎟⎜⎜ ⎟⎝ ⎠⎝ ⎠+
⎞⎛+ + ⎟⎜
⎝ ⎠= − +
⎞⎛ ⎞⎛+ + +⎟⎜ ⎟⎜⎜ ⎟⎝ ⎠⎝ ⎠+
[ ] [ ] [ ] [ ] [ ] [ ]
[ ] [ ] [ ] [ ]
[ ] [ ]
2 3 5 6 22 1
1 4 5
8 9 11 12 83 4 2
7 10 11
12 62 2, , ,3 3 3 3 3
126 6 , ,3 3 3 3
123
2 2 2 , 2 , 2 , , , 6
3 36
6 , 6 , 6 , , , 123 3
1212 , 12
T T
t T T T T T T
T T
T T T
t t
t t
t
T T T
T
T
RpJi w r p M M n i
f S f S fRpJ
w r p M M n if S f
I IR
S f
P f f f f fS S S S S
IR P f f f f
RpJ
fS S S S
RS
w r
= = = = = =
= = = = = =
= =
+ += =
+ +=
[ ] [ ] [ ] [ ]
14 15 17 18 145 6 3
13 16 17
20 21 23 24 207 8 4
19 22 23
12 ,3 3
35
, 12 , , ,3 3
3535 35 , 35 , 35 , , 35 35, .
3 3 33,
33
T
T T T T
T T
T T T T T
t
t tT
p M M nf S f S f
Rp
P f f f f fS S
I R P f f fJi w r p M M n
f S f S ff f
S S S S
= = =+ +
+
=
= == = = = =+
(S58)
Conservation equations:
(S59)
Conservation Equations (S59) in non-dimensional form:
(S60)
where
[ ] [ ][ ] ( )[ ]
[ ] [ ][ ] ( )[ ]
[ ] [ ] [ ] [ ][ ] [ ]
[ ] [ ] [ ][ ] [ ]
[ ] [ ][ ] ( )[ ]
[ ] [ ][ ] ( )
1 2 3
2
1
4 5 6
5 7 9
220 21 22
7 8 9
10 11 1
9
2
2 3 2 3 2 3 ,
6 3 6 3 6 3 ,
3 2 3 6 3 3 3 3 3
2 3 2 33 3 5 35 ,
3 3 3 3 3 3 ,
12 4 12 4 12
p p
p p p
d RpJ S a RpJ S a a RpJ Sdtd RpJ S a RpJ S a a RpJ Sdtd S a RpJ S a RpJ S a P S p a P Sdt
a S a S a S S a Sd P S a P S a P Sdtd Rp
p p a
J S a RpJ S a a Rp
p
Jdt
= − +
= − +
= + − + −
− + − +
= − +
= − + [ ]
[ ] [ ][ ] ( )[ ]
[ ] [ ] [ ] [ ][ ] [ ]
[ ][ ] [ ] [ ] [ ]
[ ] [ ][ ] ( )[ ]
[ ] [ ] [ ]
[ ] [ ][ ] [ ]
13 14 15
11 14 18
21 22 23 24
17 18
20
21 22
16
2
16
219
4 ,
35 4 35 4 35 4 ,
4 12 4 35 4 4 4 4 4
3 4 34 2 4 2 44 ,
4 4 4 4 4 4 ,
33 3 33 ,
34 34 ,3 4
S
d RpJ S a RpJ S a a RpJ Sdtd S a RpJ S a RpJ S a P S a P Sdt
a S S a S a S p a Sd P S a P S a a P Sdtd S a S p a
p p p
p
Sdtd S a S p a Sdt
p
p
d
p p
S p
= − +
= + − + −
− + − +
= − +
= −
= −
[ ] [ ] [ ]2 242
344 4 44 .S a S p a Sdt
= −
[ ] [ ] [ ] [ ] [ ] [ ][ ] [ ] [ ] [ ] [ ] [ ][ ][ ]
3 3 3 2 33 34 2 3 6 3 [ 3 3 ],
4 4 4 2 44 34 12 4 35 4 [ 4 4 ],
3 3 [ 3 3 ],
4 4 [ 4 4 ].
T
T
T
T
S S S p S S RpJ S RpJ S P S
S S S p S S RpJ S RpJ S P S
P P P S
P
p
P
p
p
pP S
= + + + + + +
= + + + + + +
= +
= +
[ ] [ ] [ ] [ ] [ ] [ ] [ ][ ] [ ] [ ] [ ] [ ] [ ] [ ][ ] [ ][ ] [ ]
1 3 3 2 33 34 2 3 6 3 3 3
4 4 4 2 44 34 12 4 35 4 4 4
3 3 3 3
4 4
,
,
,
,4 4
t
t
t
s s s s w s w s p s
s s s s s w s w s p s
p p p s
p p p s
p p
p p
p
p
= + + + + + +
= + + + + + +
= +
= +
ODEs (S58) in non-dimensional form can be written as follows:
(S61)
where
[ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ]3 3 2 3 6 3 3 3 43 , 3 , 2 3 , 6 33 3 3 3
, 3 3 ,3
,3
4 T
T T T T T Tt
p pp p
SS S RpJ S RpJ
S SS P S Ss s w s w s p s s
S S S= = = = = =
[ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ]3 34 4 12 4 35 4 4 4
4 , 4 , 12 4 , 35 4 , 3
43
3
4 ,T T T T T
pS S RpJ S RpJ S P Ss s w s w s p
pp p
S S S S Ss= = = = =
[ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ]33 34 44 3 4 3 433 , 34 , 44 ,3 3 3
3 ,3 3 3 3
4 , 3 , 4 .T T
T T T T T Tt t
T
S S S PS S S
P P Ps s s p p p pS S S S
= = = = = = =
[ ] [ ][ ] [ ]
[ ] [ ][ ] ( )[ ]
[ ] [ ] [ ] [ ][ ] [ ]
[ ] [ ] [ ][ ] [ ]
[ ] [ ][ ] ( )[ ]
[ ] [ ][ ] ( )[ ]
[ ] [ ][ ] ( )[ ]
1
3 4 5
2 4 6 8
18 19 20 21
6 7 8
9 10 11
12 13 14
2
2 3 2 3 2 3
6 3 6 3 6 3
3 2 3 6 3 3 3 3 3
2 3 2 33 3 4 34
3 3 3 3 3 3
12 4 12 4 12 4
35 4 35 4 35 4
d w s m w s w sdd w s m w s m m w sdd s p m w s m w s m p s m p sd
m s p m s m s p s p m sd p s p m p s p m m p s pdd w s m w s m m w sdd w s m w s m m w s
p
d
p
τ
τ
τ
τ
τ
τ
= −
= − +
= + − + −
− + − +
= − +
= − +
= − +
[ ] [ ] [ ] [ ][ ] [ ]
[ ][ ] [ ] [ ] [ ]
[ ] [ ][ ] ( )[ ]
[ ] [ ] [ ]
[ ] [ ][ ] [ ]
[ ] [ ] [ ]
10 13 15
220 21 22 23
15
18
20 21
17
16 17
219
222 23
4 12 4 35 4 4 4 4 4
3 4 34 2 2 44
4 4 4 4 4 4
33 3 33
34 3 4 34
44 4 4
4
4
d s m w s m w s m p s m p sd
m s s m s m m sd p s p m
p p p
p p s p
p s m m p sdd s m s m sdd s m s p s p
p p
p
p
m sdd s m s m sd
τ
τ
τ
τ
τ
= + − + −
− − +
= −
=
= −
−
+
+
−
=
( ) 12 3 1
2 3 2 3 2 3 2 3 2 3
2 3 2 3 2
2 4 5 62 3 4 5
7 8 9 10 11 126 7 8 9 10 11
13
3 2 3 2 3 2 3
2 3
14
2
1512 13 14
3 2
, 3 , , 3 , , ,
3 , , , 3 , , ,
3 , ,
T T
T T
T
t a a m S m m S m ma a a a a a a a a a
m S m m m S m ma a a a a a a a a a a a
m
a a a a a
a a a a a a
a S m ma a a
a aa a
τ = + = = = = =+ + + + +
= = = = = =+ + + + + +
= = =+ + +
16 17 1815 16 17
19 20 21 22 23 2418 19 20 21
3
22 23
2 3 2 3 2 3
2 3 2 3 2 3 2 3 2 3 2 3
, 3 , , ,
3 , , 3 , , 3 , .
T
T T T
m S m ma a a a a a a
m S m m S m m S ma a a a a a a a
a a a
a a a a a aa a a a
= = =+ + +
= = = = = =+ + + + + +
We need to find steady-state solution of System (S61):
(S62)
We can simplify System (S62):
(S63)
Then using conservation Equations (S60) and System (S63) we obtained:
(S64)
[ ][ ] [ ][ ][ ] ( )[ ]
[ ] [ ] [ ] [ ][ ] [ ]
[ ] [ ] [ ][ ] [ ][ ][ ] ( )[ ][ ][ ] ( )[ ][ ][ ] ( )[ ]
[ ] [ ] [ ] [ ][ ]
1
3 4 5
2 4 6 8
18 19 20 21
6 7 8
9 10 11
12 13 14
10 13 15
2
0 2 3 2 3
0 6 3 6 3
3 2 3 6 3 3 3 3 3
2 3 2 33 3 4 34
0 3 3 3 3
0 12 4 12 4
0 35 4 35 4
4 12 4 35 4 4 4
m w s w s
m w s m m w sd s p m w s m w s m p s m p sd
m s p m s m s p s p m s
m p s p m m p s p
m w s m m w s
m w s m m w sd s m w s
p
m w s mp ppd
p
s
τ
τ
= −
= − +
= + − + −
− + − +
= − +
= − +
= − +
= + − [ ]
[ ][ ] [ ] [ ] [ ][ ][ ] ( )[ ][ ] [ ][ ][ ] [ ][ ] [ ]
220 21 22 23
15
18
20 21
2
17
16 17
2
22 23
19
4 4
3 4 34 2 2 44
0 4 4 4 4
0 3 33
0 3 4 34
0 4 44
4
m p s
m s s m s m m s
m p s m m p s
m s m s
m s p s p m s
m s m s
p
p p s p
p p
p
p
+ −
− − +
= − +
= −
= −
= −
+
[ ][ ] [ ][ ][ ] ( )[ ]
[ ] [ ] [ ] [ ]
[ ][ ] ( )[ ][ ][ ] ( )[ ][ ][ ] ( )[ ]
[ ] [ ] [ ] [ ]
[ ][ ] ( )[ ][ ] [ ]
1
3 4 5
2 4 7
6 7 8
9 10 11
12 13 1
16
16 1
4
10 13
15
1
7
298 1
2
0 2 3 2 3
0 6 3 6 3
3 2 3 6 3 3 3
0 3 3 3 3
0 12 4 12 4
0 35 4 35 4
4 12 4 35 4 4 4
0 4 4 4 4
0 3 33
0
m w s w s
m w s m m w sd s p m w s m w s m p sdm p s p m m p s p
m w s m m w s
m w s m m w sd s m w s m w s m p sdm p s m m p s
m s
p
s
p p
m
m
p p
p
τ
τ
= −
= − +
= + −
= − +
= − +
= − +
= + −
= − +
= −
= [ ][ ] [ ][ ] [ ]0 21
222 23
3 4 34
0 4 44p
s p s p m s
m s m s
−
= −
[ ] [ ] [ ] [ ]
[ ] [ ] [ ] [ ]
2 4 7
10 13 16
3 2 3 6 3 3 3 ,
4 12 4 35 4 4 4 ,
d s p m w s m w s m p sdd s m
p
p w s m w s m p sd
p
τ
τ
= + −
= + −
where
Or we can rewrite it as follows:
[ ] [ ]
[ ]
[ ] [ ]
[ ]
[ ] [ ]
[ ] [ ][ ]
[ ] [ ]
[ ] [ ][ ]
[ ] [ ][ ]
[ ] [ ][ ]
[ ] [ ][ ]
[ ] [ ] [ ] [ ] [ ]
[ ] [ ]
7 8
6
16 17
15
18
20
21
22
23
3
4 5
9
10 11
12
13 14
3
4 5
2
19
2
1
1
33 3 3
3
44 4 4
4
33 3
34 3 4
44 4
2 3 2 3
6 3 6
12 4 12
35 4 35
3 2 3
3
4
4
3 34 3 33
1 2 6
1
t
t
s pp s p m m s p
ms p
p s p m m s pm
ms s pmms s s pmms s pm
w s m w
p
p
p
s
s
s
s
p
mw s wm m
mw s wm mmw s w
m ms s s p s
s mm w wm m
p
=+
+
=+
+
=
=
=
=
−
=+
=+
=+
− − −=
+ ++
[ ] [ ] [ ] [ ] [ ]
[ ] [ ]9 12
10 11 13 14
4 4 2 44 34 4 44
1 12 35
t p ps s s s p ss m mw w
m m m m
− − −=
+ ++ +
−
(S65)
where we denote the Michaelis constants
When considering steady-state solutions of System (S64) we can write:
(S66)
Or we can rewrite Equations (S66) as follows:
(S67)
[ ] [ ][ ]
[ ] [ ][ ]
[ ] [ ]
[ ] [ ][ ]
[ ] [ ]
[ ] [ ][ ]
[ ] [ ][ ]
[ ] [ ][ ]
[ ] [ ][ ]
[ ] [ ] [ ] [ ] [ ][ ] [ ]
[ ] [ ] [ ] [ ] [ ][ ] [ ]
11
14
2
15
162
17
9
10
12
13
9 10
12 13
33 3 3
3
44 4 4
4
333
3 434
444
2 32 3
66 3
1212 4
3535 4
3 2 33 34 3 33
2 61
4 4 2 44 34 4 4
3
412
4
1
351
4
t
t
t
s pp s p
M s p
s pp s p
M s p
s ps
Ms s p
sM
s ps
Mw
w sM
ww s
Mw
w s
p
p
p
s
s
s
s
p p
p
Mw
w sM
s s s p ss
w wM M
s s s s p ss
w wM
p
M
=+
=+
=
=
=
=
=
=
=
− − −=
+ +
−
+ +
−
− − −=
2 3 5 6 8 9 11 12 14 159 10 11 12 13
1 4 7 10 13
17 18 20 22 2414 15 16 17
16 19 21 23
, , , , ,
,
3 3 3 3 3
3 3 3, , .
3
T T T T T
T T T T
a a a a a aa a a aM M M M Ma S a S a S a S a Sa a a aM M M Ma S a S a S a S
a
= = = = =
= =
+ +
+=
+
=
+ +
[ ] [ ] [ ][ ] [ ] [ ]
2 4 7
10 13 16
0 2 3 6 3 3 3 ,
0 12 4 35 4 4 4 .
m w s m w s m p s
m
p
pw s m w s m p s
= + −
= + −
[ ] [ ] [ ][ ] [ ] [ ]5 6
7 8
0 2 3 6 3 3 3 ,
0 12 4 35 4 4 4 ,
n w s n w s p s
n w s
p
n w s p s p
= + −
= + −
where .
We can rewrite System (S67) substituting Equations (S65):
(S68)
We find and in System (S68) numerically.
CD46 It can be written for SP1 in non-dimensional form according to Equation (S40):
(S69)
where , , , , and .
2.2.3 IFN-γ and IL-10 production Since IFN-γ is activated in this module by STAT44 only (Fig 1C) we can write using Equation
(S51):
(S70)
where .
According to Equation (S53) we can write:
2 5 11 145 6 7 8
8 8 17 17
, , ,a a a an n n na a a a
= = = =
[ ] [ ] [ ][ ]
[ ][ ]
[ ] [ ][ ] [ ]
[ ] [ ] [ ][ ]
[ ][ ]
[ ] [ ][ ] [ ]
2
15 16
9 10 10 9
11 10 9
2
17 16
12 13 13 12
14 13
6
8 1
5
27
3 3 40 3 2
3 2 63 1 1,
3 2 6
4 3 40 4 2
4 12 354 1 4 ,
4 12 35
t
t t
s p s s ps p
M M
s p M M M w M wpM s p n M w n M w
s p s s ps p
M M
s p M M M w M wp s
M s p n M w n M w
p
p
= + + +
⎞⎛ + ++ + −⎟⎜
+ +⎝ ⎠
= + + +
⎞⎛ + ++ + −⎟⎜
+ +⎝ ⎠
[ ]3s p [ ]4s p
[ ] [ ][ ]
[ ][ ]18 19
2 461 1 ,
2 46ti cd
sp a spM i M cd
=+ +
[ ] [ ]113T
SP asp a
S=
113T
tT
SPspS
= [ ] [ ]46463T
CDcd
S= 2
181 3T
lMl S
= 419
3 3t
lMl S
=
[ ]
[ ][ ]
20
921
444
1
4
,1
t
t
Mig mp
ns
gsM
g+
=−
[ ] [ ] 4 5 2 520 21 9 9
3 1 4
1, 1 , , , , , 13 3 3 3 3
T Tt t
T T T T T
Ig Mp Gg k k k lig mp gg M M n nS S S k S k S k
+= = = = = = <
(S71)
where
2.3 Combined STAT3-STAT4-STAT5 model
The reactions involved in STAT3-STAT4-STAT5 circuit (Fig 5A):
[ ]
[ ][ ]
[ ][ ]
[ ][ ]
[ ][ ]
22
23 24 23 2410
33 1 33 110
33
10 2
1 3
1
3
,
1t
t
s sp a s sp aM s M
Mi mp
nsp a M s M sp a
g⎞⎛
+ − ⎟⎜+ + +⎝ ⎠
−
+
=
11 12 7 9 622 23 24 10 10
10 6 8 11
2 102 , 10 , , , , , 1.3 3 3 3 3T T
t tT T T T T
Mp G k k k k lmp g M M M n nS S k S k S k S k
+= = = = = = <
2
3
4 5
6
7 8
9
10 11
12
13 14
15
1
I2RJ2 R J2
R J2 R J2 RJ2
I6RJ6 R J
I2 RJ2 p
p P2 p P2 P2
I6 RJ6 p
p
6
R J6 R J6 RJ6
I12
P6 p
RJ12 R
P6 P6
I12 RJ1 J1p 22
f
f
f
f
f
f
f
f
f
f
f
f
f
f
f
→←
→←
→←
→←
→←
→
→
→
+
+ +
+
+ →
→
+
+
(S72)
16 17
18
19 20
21
22 23
24
25 26
27
28
3
29
2
30
1
R J12 R J12 RJ12
I35RJ35 R J35
R J35 R J35 RJ35
I21RJ21 R J21
R J
p P12 p P12 P12
I35 RJ35 p
p P35 p P35 P35
I21 RJ21 p
p P21 p P21 P221 R J21 RJ21
R J2 S3 R J2S3
1
p p RpJ
f
f
f
f
f
f
f
f
f
f
a
a
f
f
f
a
f
f
→←
→←
→←
→←
→←
→←
→
→
+ +
+
+ +
+
+ →
+
→
→
+→
4 5
6
7 8
9
10 11
12
13 14
15
17
1
1
0
2
6
8
19 2
1
2 S3p
R J6 S3 R J6S3 R J6 S3p
P3 S3p P3S3p P3 S3
R J12 S4 R J12S4 R J12 S4p
R J35 S4 R J35S4 R J35 S4p
P4 S4p P4S4 P4 S4
R J2 S5 R J2S5 R J2 S5p
p p p
p p p
p p p
p
p
R J21 S
p p
p
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
→←
→←
→←
→←
→←
→←
+
+ → +
+ → +
+ → +
+ → +
+ → +
+ → +
+22 23
24
25 26
27
28
29
30
31
5 R J21S5 R J21 S5p
P5 S5p P5S5 P5 S5
S3 S3
p p
p S33
S3p S4p S
p
4
p
3
a
a
a
a
a
a
a
a
a
a
→←
→←
→←
→←
→ +
+ → +
+
+
32
33
34
35
36
37
1
2
3
4
3
4
1
2
1
1
1
5 6
7
p p
p p
S44 Gg S44Gg
S44Gg S55 S44S55Gg
S55 Gg S55Gg
S55Gg S44 S44S55Gg
S44Gg Ig
S55Gg Ig
S44S55Gg Ig
Ig Mp1
S4 S4 S4
IgMp1 Ign Mp1
I2 Sp1
4
S3p S5p S35
S5 S5 S55
a
a
a
k
k
k
k
k
k
k
a
a
k
k
k
a
l
l
l
k
→←
→←
→←
→←
→←
→←
→←
→←
+
+
+
+
+ +
→
→
→
→
+
+
+
+
2
3
4
5
4
5
2
3
8
9
10
11
10
11
8
9
6
6
I2Sp1
I2Sp1 CD46 Sp1a
CD46 Sp1 CD46Sp1
CD46Sp1 I2 Sp1a
S33 G10 S33G10
S33G10 Sp1a S33Sp1aG10
Sp1a G10 Sp1aG10
Sp1aG10 S33 S33Sp1aG10
S33G10 I10
Sp1aG10 I10
S33Sp
l
l
l
l
l
l
l
l
k
k
k
k
k
k
k
k
l
l
→←
→←
→←
→←
→←
→←
→←
→←
+
+
+
+
+
+
+
→
→6
12 13
14
1aG10 I10
I10 Mp2 I10Mp2 I10n Mp2k
k
l
k→← →+ +
→
2.3.1 Cytokine-receptor interactions According to Equation (S12) we can write for , , , and
in non-dimensional form respectively:
(S73)
[ ]2RpJ [ ]6RpJ [ ]21RpJ [ ]35RpJ
[ ]21RpJ
[ ][ ]
[ ]
[ ][ ]
[ ]
[ ][ ]
1
1 1
2
1
1 1
3
2 2
2
3
2 2
5
2
2 2
4
4 4
36
3
1 12
22
1 1 42
,21 1
66
2
1 1 4
2 2
2 2 2
6 6
6 6 6
12
6,
21 1
11
212
2
2
t t
t t t
t t
t t t
t t
M r p
M r p r M
M r p
M r
Mn i n
w
Mn i n
Mn i n
w
Mn i n
Mn i
M
wM r p
n
p r
⎞⎛+ + ⎟⎜
⎝ ⎠= − +
⎞⎛ ⎞⎛+ + +⎟⎜ ⎟⎜⎜ ⎟⎝ ⎠⎝ ⎠+
⎞⎛+ + ⎟⎜
⎝ ⎠= − +
⎞⎛ ⎞⎛+ + +⎟⎜
− +
− +
− +
− + ⎟⎜⎜ ⎟⎝ ⎠⎝ ⎠+
⎞⎛+ + ⎟⎜
⎝=
−⎠
+
− +
[ ]
[ ][ ]
[ ]
[ ][ ]
[ ]
2
5
3 3
7
4 4
2
7
4 4
6 6
8
8 8
10
10
9
5 5
9
5
12 12 12
35 35
35 35 35
21
1 1 412
,2
1 135
352
1 1 435
,2
1 121
212
2
21
21 211
t t t
t t
t t t
t t
t t
Mn i n
Mn i n
w
Mn i n
Mn i n
w
Mn i
M r p r M
M r p
M r p r M
M r p
M r p
⎞⎛ ⎞⎛+ + +⎟⎜ ⎟⎜⎜ ⎟⎝ ⎠⎝ ⎠+
⎞⎛+ + ⎟⎜
⎝ ⎠= − +
⎞⎛ ⎞⎛+ + +⎟⎜ ⎟⎜⎜ ⎟⎝ ⎠
− +
− +
− +
− +
⎝ ⎠+
⎞⎛+ + ⎟⎜
⎝ ⎠= +
+
−
−
+ 10
2
5
1 14
,2
21 tr Mn
⎞⎛ ⎞⎛+ + +⎟⎜ ⎟⎜⎜ ⎟⎝ ⎠⎝ ⎠
where
2.3.2 STAT phosphorylation and dimerization ODEs for the STAT module are given by:
[ ] [ ] [ ] [ ] [ ] [ ]
[ ] [ ] [ ] [ ]
[ ] [ ]
2 3 5 6 22 1
1 4 5
8 9 11 12 83 4 2
7 10 11
12 62 2, , ,3 3 3 3 3
126 6 , ,3 3 3 3
123
2 2 2 , 2 , 2 , , , 6
3 36
6 , 6 , 6 , , , 123 3
1212 , 12
T T
T T T T T T T
T T
T T T
t t
t t
t
T T T
T
T
RpJi w r p M M n i
f S f S fRpJ
w r p M M n if S f
I IR
S f
P f f f f fS S S S S
IR P f f f f
RpJ
fS S S S
RS
w r
= = = = = =
= = = = = =
= =
+ += =
+ +=
[ ] [ ]
[ ] [ ]
[ ] [ ] [ ] [ ]
14 15 17 18 145 6 3
13 16 17
20 21 23 24 207 8 4
19 22 23
, 12 , , , 353 3
3535 , 35 , 35 , , ,
3 321
3512 , ,3 3 3
35 35 ,3 3 3
21 221 21 , 21 , 1,3 3 3
T
T T T T T
T T
T
t
t t
t
T T T T
T
T T T
p M M n if S f S f
RpJw r p M M n
f S
IP f f f f fS S S
R P f f f f fS S S f S f
RpJI RS S S
i w r p
= = = =
= = = = =
+ +=
+ +=
= == 26 27 29 30 269 10 5
25 28 29
21 , ,21 .3
,3 3
T
T Tt
T
M M nf S f
P f f fS ff f
S= = = =
+ +
[ ] [ ][ ] ( )[ ]
[ ] [ ][ ] ( )[ ]
[ ] [ ] [ ] [ ][ ] [ ]
[ ] [ ] [ ][ ] [ ][ ][ ] [ ]
[ ] [ ][ ] ( )[ ]
1 2 3
4 5 6
5 7 9
228 29 30 31
34 35
7 8 9
2
2 3 2 3 2 3 ,
6 3 6 3 6 3 ,
3 2 3 6 3 3 3 3 3
2 3 2 33 3 4 34
3 5 35 ,
3 3 3 3 3 3 ,
d RpJ S a RpJ S a a RpJ Sdtd RpJ S a RpJ S a a RpJ Sdtd S a RpJ S a RpJ S a P S p a P Sdt
a S a S a S S a S
a S S a S
p p
p p p
p p
p pd P S a P S a P Sdt
a p
= − +
= − +
= + − + −
− + − + −
− +
= − +
(S74)
Conservation equations:
(S75)
Conservation Equations (S75) in non-dimensional form:
[ ] [ ][ ] ( )[ ]
[ ] [ ][ ] ( )[ ]
[ ] [ ] [ ] [ ][ ] [ ]
[ ][ ] [ ] [ ] [ ]
[ ] [ ][ ] ( )[ ]
[ ]
10 11 12
13 14 15
11 14 18
30 31 32
16
2
16
1
1
9
33
17 8
12 4 12 4 12 4 ,
35 4 35 4 35 4 ,
4 12 4 35 4 4 4 4 4
3 4 34 2 4 2 44 ,
4 4 4 4 4 4 ,
2 5
d RpJ S a RpJ S a a RpJ Sdtd RpJ S a RpJ S a a RpJ Sdtd S a RpJ S a RpJ S a P S a P Sdt
a S S a S a S p a Sd P S a P
p
S a a P Sdt
p p
p p
p
d
p p
RpJ S adt
= − +
= − +
= + − + −
− + − +
= − +
= [ ][ ] ( )[ ]
[ ] [ ][ ] ( )[ ]
[ ] [ ] [ ] [ ][ ] [ ]
[ ][ ] [ ] [ ] [ ]
[ ] [ ][ ] ( )[ ]
[ ] [ ] [ ]
[ ]
20
20
21
22 23 24
23 25 27
234 35 36 37
25 26 27
28 292
2 5 2 5 ,
21 5 21 5 21 5 ,
5 2 5 21 5 5 5 5 5
3 5 35 2 5 2 55 ,
5 5 5 5 5 5 ,
33 3 33 ,
34
RpJ S a a RpJ S
d RpJ S a RpJ S a a RpJ Sdtd S a RpJ S a RpJ S a P S p a P Sp p
pdt
a S S a S a S a Sd P S a P S a P Sdtd S a S p a Sdtd
p p
p p
S adt
a p
− +
= − +
= + − + −
− + − +
= − +
= −
= [ ][ ] [ ]
[ ] [ ] [ ]
[ ] [ ][ ] [ ]
[ ] [ ] [ ]
30 31
32 33
34 35
26
2
3 37
4 34 ,
44 4 44 ,
3
3
35 5 35 ,
55 5 55 .
S p a S
d S a S p a Sdtd S a S p a Sdtd S a S p a Sdt
S p
S p
−
= −
= −
= −
[ ] [ ] [ ] [ ] [ ] [ ] [ ][ ] [ ] [ ] [ ] [ ] [ ][ ] [ ] [ ] [ ] [ ] [ ][ ][ ][ ]
3 3 3 2 33 34 35 2 3 6 3 [ 3 3 ],
4 4 4 2 44 34 12 4 35 4 [ 4 4 ],
5 5 5 2 55 35 2 5 21 5 [ 5 5 ],
3 3 [ 3 3 ],
4 4 [ 4 4 ],
5 5 [ 5 5 ].
T
T
T
T
T
T
S S S p S S S RpJ S RpJ S P S
S S S p S S RpJ S RpJ S P S
S S S p S S RpJ
p
p
p
p
S RpJ S P S
P P P S
P P P S
P P P S
p
p
= + + + + + + +
= + + + + + +
= + + + + + +
= +
= +
= +
(S76)
where
ODEs (S74) in non-dimensional form are given by:
[ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ][ ] [ ] [ ] [ ] [ ] [ ] [ ][ ] [ ] [ ] [ ] [ ] [ ] [ ][ ] [ ][ ] [ ][ ] [ ]
1 3 3 2 33 34 35 2 3 6 3 3 3
4 4 4 2 44 34 12 4 35 4 4 4
5 5 5 2 55 35 2 5 21 5 5 5
3 3 3 3
4 4 4 4
5 5 5 5
,
,
,
,
,
,
t
t
t
t
t
s s s s s w s w s p s
s s s s s w s w s p s
s s s s s w s w s p s
p p p s
p p
p p p s
p
p p
p p
s pp
p
p
p
= + + + + + + +
= + + + + + +
= + + + + + +
= +
= +
= +
[ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ]3 3 2 3 6 3 3 3 43 , 3 , 2 3 , 6 33 3 3 3
, 3 3 ,3
,3
4 t
T T T T T Tt
p pp p
SS S RpJ S RpJ
S SS P S Ss s w s w s p s s
S S S= = = = = =
[ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ]3 34 4 12 4 35 4 4 4
4 , 4 , 12 4 , 35 4 , 3
43
3
4 ,T T T T T
pS S RpJ S RpJ S P Ss s w s w s p
pp p
S S S S Ss= = = = =
[ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ]5 5 2 5 21 5 5 555 , 5 , 5 , 2 5 , 21 5 ,3 3
5 5 ,3 3 3 3
T
T T T T T Tt
S S RpJp pp p
S S S S SS RpJ S P SSs s s w s w s p s
S= = = = = =
[ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ]33 34 44 35 55 3 433 , 34 , 44 , 35 , 55 , 3 , 4 ,
3 3 3 3 3 3 3T T T T T T TSS S S S S P P
s sS S S S S
s pS
s s p= = = = = = =
[ ] [ ]5 3 4 55 , 3 , 43 3 3
.3, 5T T T
T Tt t t
T T
PS S S
P P Pp p pS
p = = = =
(S77)
where
[ ] [ ][ ] [ ]
[ ] [ ][ ] ( )[ ]
[ ] [ ] [ ] [ ][ ] [ ]
[ ] [ ] [ ][ ] [ ][ ][ ] [ ]
[ ] [ ][ ] ( )[ ]
[ ] [ ][ ] ( )[ ]
[ ]
1
3 4 5
2 4 6 8
27 28 292
30
33 34
6 7 8
9 10 11
12
2 3 2 3 2 3
6 3 6 3 6 3
3 2 3 6 3 3 3 3 3
2 3 2 33 3 4 34
3 5 35
3 3 3 3 3 3
12 4 12 4 12 4
35 4
d w s m w s w sdd w s m w s m m w sdd s p m w s m w s m p s m p sd
m s p m s m s p s p m s
m s p s p m sd p s p m p s p m m p s pdd w s m w s m m w sdd w sd
p
m
p
τ
τ
τ
τ
τ
τ
= −
= − +
= + − + −
− + − + −
− +
= − +
= − +
= [ ][ ] ( )[ ]
[ ] [ ] [ ] [ ][ ]
[ ] [ ][ ][ ] [ ] [ ]
[ ] [ ][ ] ( )[ ]
[ ] [ ][ ] ( )[ ]
[ ] [ ][ ] ( )[ ]
[ ] [ ]
17
16 1
13 14
10 13 15
29
230 31 32
15 7
1918 20
21 22 23
19 2
35 4 35 4
4 12 4 35 4 4 4
4 4 3 4
34 2 2 44
4 4 4 4 4 4
2 5 2 5 2 5
21 5 21 5 21 5
5 2 5
4
w s m m w s
d s m w s m w s m p sd
m p s m s s
m s m m sd p s p m p s m m p sdd w s m w s m m w sdd w s m w s m m
p p
p p p
s p
w sdd s p m w s md
p p
τ
τ
τ
τ
τ
− +
= + − +
+ −
+ − +
= − +
=
− +
+
− +
=
= + [ ] [ ][ ] [ ]
[ ][ ] [ ] [ ] [ ]
[ ] [ ][ ] ( )[ ]
[ ] [ ] [ ]
[ ] [ ][ ] [ ]
[ ] [ ] [ ]
[ ] [ ][ ] [ ]
[ ] [ ] [ ]
2 24 26
33 34 35 36
24 25 26
27 28
29 30
231 32
33 34
235 36
2
2
21 5 5 5 5 5
3 5 35 2 5 2 55
5 5 5 5 5 5
33 3 33
34 3 4 34
44 4 44
35 3 5 35
55 5 55
w s m p s m p s
m s p s p m s m s p m sd p s p m p s p m m p s pdd s m s m sdd s m s p s p m sdd s m s m sdd s m s p s p m sdd s m s m
p
p
sd
p
p
p
τ
τ
τ
τ
τ
τ
− + −
− + − +
= − +
= −
= −
= −
= −
= −
We need to find a steady-state solution of System (S77):
(S78)
We can simplify System (S78):
( ) 12 3 1
2 3 2 3 2
2 4 5 6 72 3 4 5 6
8 9 10 11
3 2 3 2 3 2 3
2 3 2 3 2 3 2 3 2 3 2 3 2 3
2
12 13 147 8 9 10 11 12 13
1514
, 3 , , 3 , , , 3 ,
, , 3 , , , 3 , ,
t t t
t t
t a a m S m m S m m m Sa a a a a a a a a a a a
m m m S m m m S ma a a a a a a a a a a a a a
m
a a a a a a
a a a a a a
aa a
a
τ = + = = = = = =+ + + + + +
= = = = = = =+ + + + + + +
=+
16 17 18 19 20 2115 16 17 18 1
3 2 3 2 3 2 3 2 3 2 39 20
22 23 24 25 26 2
2 3
2 3 2 3 2 3 2 3
721 22 23 24 25 26
2827 2
2 3 2 3
2 3
, 3 , , , 3 , , ,
3 , , , 3 , , ,
3 ,
t t
t t
t
m S m m m S m ma a a a a a a a a a a a
m S m m m S m ma
a a a a a a
a a a a a aa a a a a a a a a a a
m S ma aa
= = = = = =+ + + + + +
= = = = = =+ + + + + +
=+ 2 3 2 3 2 3 2 3 2 3
2 3 2
29 30 31 32 338 29 30 31 32
34 35
3 2 3
36 3733 34 35 3
2 36
, 3 , , 3 , ,
3 , , 3 , .
t t
t t
m S m m S ma a a a a a a a a a
m S m m S ma a a a a
a a
a a
a a a
a a a aa
= = = = =+ + + + +
= = = =+ + + +
[ ][ ] [ ][ ][ ] ( )[ ]
[ ] [ ] [ ] [ ][ ] [ ] [ ]
[ ] [ ][ ] [ ][ ][ ] [ ]
[ ][ ] ( )[ ][ ][ ] ( )[ ][ ][ ] ( )[ ]
[ ] [ ]
1
3 4 5
2 4 6 8 27
28 29 30
33 34
6 7 8
9 10 11
12 13 14
10
2
0 2 3 2 3
0 6 3 6 3
3 2 3 6 3 3 3 3 3 2 3
2 33 3 4 34
3 5 35
0 3 3 3 3
0 12 4 12 4
0 35 4 35 4
4 12 4
m w s w s
m w s m m w sd s p m w s m w s m p s m p s m s pd
m s m s p s p m s
m s p s p m s
m p s p m m p s p
m w s m m w s
m w s m m w sd s m w s
p p
pd
τ
τ
= −
= − +
= + − + − +
+ − + −
− +
= − +
= − +
= − +
= [ ] [ ][ ] [ ] [ ][ ]
[ ] [ ] [ ][ ][ ] ( )[ ][ ][ ] ( )[ ][ ][ ] ( )[ ]
[ ] [ ] [ ] [ ][ ] [ ]
[ ][ ] [ ] [ ]
13 15 29
230 31 32
15
18 20
21
1
22 23
19 22 24 26
33 34 3
7
16 17
1
5
9
2
35 4 4 4 4 4 3 4
34 2 2 44
0 4 4 4 4
0 2 5 2 5
0 21 5 21 5
5 2 5 21 5 5 5 5 5
3 5 35 2 5
4
2
m w s m p s m p s m s s
m s m m s
m p s m m p s
m w s m m w s
m w s m m w sd s p m w s m w s m p s m p sd
m s p
p p p
s p m s m s p m
p
s p
p p
p pτ
+ − + −
+ − +
= − +
= − +
= − +
= + − + −
− + − +
+
[ ][ ][ ] ( )[ ][ ] [ ][ ][ ] [ ][ ] [ ][ ][ ] [ ][ ] [ ]
36
24 25 26
27 28
29 30
231 32
33 34
235
2
36
55
0 5 5 5 5
0 3 33
0 3 4 34
0 4 44
0 3 5 35
0 5 55
s
m p s p m m p s p
m s m s
m s p s p m s
m s m s
m s p s p m s
m s m s
p
p
p
= − +
= −
= −
= −
= −
= −
(S79)
Then using conservation Equations (S76) and System (S79) we obtained:
(S80)
where
[ ][ ] [ ][ ][ ] ( )[ ]
[ ] [ ] [ ] [ ]
[ ][ ] ( )[ ][ ][ ] ( )[ ][ ][ ] ( )[ ]
[ ] [ ] [ ] [ ]
[ ][ ] ( )[ ][ ][ ] ( )
1
3 4 5
2 4 7
6 7 8
9 10 11
12 13 1
16
16 1
4
10 13
15
1
7
218 9 0
0 2 3 2 3
0 6 3 6 3
3 2 3 6 3 3 3
0 3 3 3 3
0 12 4 12 4
0 35 4 35 4
4 12 4 35 4 4 4
0 4 4 4 4
0 2 5 2
m w s w s
m w s m m w sd s p m w s m w s m p sdm p s p m m p s p
m w s m m w s
m w s m m w sd s m w s m w s m p sdm p s m m
p
p p
p pp s
m w s m m w s
τ
τ
= −
= − +
= + −
= − +
= − +
= − +
= + −
= − +
= − + [ ][ ][ ] ( )[ ]
[ ] [ ] [ ] [ ]
[ ][ ] ( )[ ][ ] [ ][ ][ ] [ ][ ] [ ][ ][ ] [ ][ ] [ ]
21 22 23
19 22 25
24 25 26
27 28
29 30
231 32
33 34
235 36
2
5
0 21 5 21 5
5 2 5 21 5 5 5
0 5 5 5 5
0 3 33
0 3 4 34
0 4 44
0 3 5 35
0 5 55
m w s m m w sd s p m w s m w s m p sdm p s p m m p s p
m
p
ps m s
m s p s p m s
m s m s
m s p s p m s
m s m s
p
p
τ
= − +
= + −
= − +
= −
= −
= −
= −
= −
[ ] [ ] [ ] [ ]
[ ] [ ] [ ] [ ]
[ ] [ ] [ ] [ ]
2 4 7
10 13
19 22 25
16
3 2 3 6 3 3 3
4 12 4 35 4 4 4
5 2 5 21 5 5 5
d s p m w s m w s m p sdd s m w s m w s m p sdd s p m w s m w s m p sd
p
p p
p
τ
τ
τ
= + −
= + −
= + −
Or we can rewrite it in the following way:
[ ] [ ]
[ ]
[ ] [ ]
[ ]
[ ] [ ]
[ ]
[ ] [ ]
[ ] [ ][ ]
[ ] [ ]
[ ] [ ][ ]
[ ] [ ]
[ ] [ ][ ]
[ ] [ ][ ]
[ ] [ ][ ]
7 8
6
16 17
15
25 26
24
27
28
29
30
31
32
33
34
35
36
3
4 5
2
2
9
10 11
2
1
33 3 3
3
44 4 4
4
55 5 5
5
33 3
34 3 4
44 4
35 3 5
55 5
2 3 2 3
6 3 6
12
3
4 412
35
t
t
t
s pp s p m m s p
ms p
p s p m m s pms p
p s p m m s pm
ms s pmms s s pmms s pmms s s p
p
p
p
p
p
s
s
mms s pm
w s m wmw s w
m mmw ss w
m m
w s
=+
+
=+
+
=+
+
=
=
=
=
=
=
=+
=+
[ ] [ ][ ]
[ ] [ ][ ]
[ ] [ ][ ]
[ ] [ ] [ ] [ ] [ ] [ ]
[ ] [ ]
[ ] [ ] [ ] [ ] [ ]
[ ] [ ]
[ ] [ ] [ ] [ ] [ ]
[ ]
12
13 14
18
19 20
21
22 23
3
4 5
9 12
10 11 13 14
18 21
19 20 22
1
4 35
2 5 2
21 5 21
3 2 33 34 35 3 33
1 2 6
4 4 2 44 34 4 44
1 12 35
5 5 2 5
4
5
5 35 5 55
1
1 2
5
t
t
m wm mmw s w
m mmw s w
m ms s s s p s
s mm w wm m
s s s s p ss m mw w
m m m ms s s s p s
s m mw
s
s
s
p
m m
p
p
m
p p
p
=+
=+
=+
− − − −=
+ ++
− − −=
+ ++ +
− − −
+
−
++
−
−=
+[ ]
23
21wm
(S81)
where we denoted the Michaelis constants
[ ] [ ][ ]
[ ] [ ][ ]
[ ] [ ][ ]
[ ] [ ]
[ ] [ ][ ]
[ ] [ ]
[ ] [ ][ ]
[ ] [ ]
[ ] [ ][ ]
[ ] [ ][ ]
[ ] [ ][ ]
[ ] [ ][ ]
[ ] [ ][ ]
[ ] [ ][ ]
[ ]
13
16
19
2
20
212
22
232
24
11
12
14
15
17
18
33 3 3
3
44 4 4
4
55 5 5
5
333
3 434
444
3 535
555
2 32 3
66
3
4
4
3
1212 4
3535 4
22 5
2121 5
3
5
5
1
t
t
t
s pp s p
M s p
s pp s p
M s p
s pp s p
M s p
p
p
p
p
p
s
s
s
s
s ps
Ms s p
sM
s ps
Ms s p
sM
s ps
Mw
w sMw
w sMw
w sM
ww s
Mw
w sM
s
s
s
ww s
Ms
=+
=+
=+
=
=
=
=
=
=
=
=
=
=
=−
=
[ ] [ ] [ ] [ ] [ ][ ] [ ]
[ ] [ ] [ ] [ ] [ ][ ] [ ]
[ ] [ ] [ ] [ ] [ ][ ] [ ]
11 12
14 15
17 18
3 2 33 34 35 3 32 6
1
4 4 2 44 34 4 44
12 351
5 5 2 55 35 5 55
2 211
t
t
s s s p sw wM M
s s s s p s
p p
p ps
w wM M
s s s s p ss
w wM
p p
M
−
− − − −
+ +
− − −=
+ +
−
+
− − −=
+
We looked for steady-state solutions of System (S80):
(S82)
We can rewrite Equations (S82) as follows:
(S83)
where .
We can rewrite System (S83) substituting Equations (S81):
(S84)
We found , and in System (S84) numerically.
2 3 5 6 8 9 11 12 14 15 17 1811 12 13 14 15 16
1 4 7 10 13 16
20 21 23 24 26 27 29 31 3317 18 19 20 21 22
19 22 25 28 30 32
23
, , , , , ,
, , ,
3 3 3 3 3 3
3,
3 3,
3 3 3,
T T T T T T
T T T T T T
a a a a a a a
a a
a a a a aM M M M M Ma S a S a S a S a S a Sa a a a a aM M M M M Ma S
aa S a S a S a S a S
aM
= = = = =+ +
=
= =
+ + + +
+= =
+=
+=
= 35 3724
34 363, .
3T T
aMa S a S
=
[ ] [ ] [ ][ ] [ ] [ ][ ] [ ] [ ]
2 4 7
10 13
1
1
22
6
9 25
0 2 3 6 3 3 3 ,
0 12 4 35 4 4 4 ,
0 2 5 21 5 5 5 .
m w s m w s m p s
m w s m w s m p s
m w s m w s
p
p
pm p s
= + −
= + −
= + −
[ ] [ ] [ ][ ] [ ] [ ][ ] [ ] [ ]
6 7
8 9
10 11
0 2 3 6 3 3 3 ,
0 12 4 35 4 4 4 ,
0 2 5 21 5 5 5 .
n w s n w s p s
n w s n w s p s
n w s n
p
s pw s
p
p
= + −
= + −
= + −
2 5 11 14 20 236 7 8 9 10 11
8 8 17 17 26 26
, , , , ,a a a a a an n n n n na a a a a a
= = = = = =
[ ] [ ] [ ][ ] [ ][ ]
[ ][ ]
[ ] [ ][ ] [ ]
[ ] [ ] [ ][ ]
[ ][ ]
[ ] [ ][ ] [ ]
[ ] [ ] [ ]
2
20 21 23
11 12 12 11
13 12 11
2
6 7
8 9
22 21
14 15 15 14
16 15 14
2
24
3 3 4 3 50 3 2
3 2 63 1 1,
3 2 6
4 3 40 4 2
4 12 354 1 4 ,
4 12 35
5 30 5 2
t
t t
s p s s p s s ps p
M M M
s p M M M w M wpM s p n M w n M w
s p s s ps p
M M
s p M M M w M wp s
M s p n M w n M
p p
p
w
p sp
Mps
s
= + + + +
⎞⎛ + ++ + −⎟⎜
+ +⎝ ⎠
= + + +
⎞⎛ + ++ + −⎟⎜
+ +⎝ ⎠
= + +[ ]
[ ][ ]
[ ] [ ][ ] [ ]
23
17 18 18 17
19 110 18 71 1
5
5 2 215 1 5 ,
5 2 21t t
s pM
s p M M M w M wp sM s p n M w n M w
⎧⎪⎪⎪⎪⎪
+
⎞⎛ + ++ + −⎟⎜
⎪⎪⎪⎪⎨⎪⎪⎪⎪⎪⎪⎪⎪⎪⎩ + +⎝ ⎠
[ ]3s p [ ]4s p [ ]5s p
CD46 It can be written for SP1 in non-dimensional form according to Equation (S40):
(S85)
where , , , , and .
2.3.3 IFN-γ and IL-10 production Since IFN-γ gene is activated by either STAT44 or STAT55, it can be written according to
Equation (S53):
(S86)
where
and
.
The production of IL-10 can be activated by either STAT33 or SP1 through CD46. Thus,
according to Equation (S53), it can be written:
(S87)
where
where .
[ ] [ ][ ]
[ ][ ]25 26
2 461 1 ,
2 46ti cd
sp a spM i M cd
=+ +
[ ] [ ]113T
SP asp a
S=
113T
tT
SPspS
= [ ] [ ]46463T
CDcd
S= 3
252 3T
lMl S
= 526
4 3T
lMl S
=
[ ]
[ ][ ]
[ ][ ]
[ ][ ]
[ ][ ]
27
1228 29 28 29
55 5
,1 144 444
555 544 54t
t
s s s sgg
s s
Mig mp
nM M s sM M
+ −+ +
=−
⎞⎛⎟⎜
⎝ ⎠+ +
[ ] [ ] 6 7 2 4 127 28 29 12
5 1 3 6
1, 1 , , , , , ,3 3 3 3 3 3
T Tt t
T T T T T T
Ig Mp Gg k k k k lig mp gg M M KM nS S S k S k S k S k
+= = = = = = =
12 1n <
[ ]
[ ][ ]
[ ][ ]
[ ][ ]
[ ][ ]
30
31 32 31 3213
33 1 33 110
33
10 2
1 3
1
3
,
1t
t
s sp a s sp aM s M
Mi mp
nsp a M s M sp a
g⎞⎛
+ − ⎟⎜+ + +⎝ ⎠
−
+
=
13 14 9 11 630 31 32 13
12 8 10 13
2 102 , 10 , , , , ,3 3 3 3 3T T
t tT T T T T
Mp G k k k k lmp g M M M nS S k S k S k S k
+= = = = = =
13 1n <
3 Model parameters
3.1 Parameter fitting
For the parameter fitting in the STAT3-STAT5 model, we used the Genetic Algorithm (GA)
tool integrated to MATLAB. As the criterion for fitting we chose the squared error, which can
be described as ( )21
N
i ii
SM E M=
= −∑ , where iE is experimental data for cytokine concentration
(IFN-γ or IL-10) corresponding to the i-th value of IL-2 concentration, M is the model
predictions for cytokine concentration corresponding to the same IL-2 concentration, i is the
number of experimental data point, 4N = is the total number of experimental data points. The
GA tool allows minimizing the squared error using the integrated algorithms for the optima
search.
We selected a "nominal" set of parameters "by hand", which qualitatively demonstrates the
switching between IFN-γ and IL-10. This set of parameters is represented as "Nom" in Table
S2. The sets of the optimized parameters and corresponding squared errors are also shown in
Table S2. We performed 15 optimization tests setting the allowable ranges for the parameters
ten-fold either side of the nominal values of parameters and chose the best fitting (set "O3" in
Table S2) with the smallest squared error 77.34 10SM −= ⋅ . Fig S1 shows the distribution of
five parameter sets with the closest minimum squared errors SM , namely sets "O2", "O3",
"O9", "O10" and "O12".
Table S2. Nominal, optimized parameters and squared error SM.
Fig S1. The distribution of the optimized parameters. The parameter sets with the closest minimum squared errors , namely "O2", "O3", "O9", "O10" and "O12". SM
Fig S2. Model predictions for the swapped parameters. A. STAT monomers. B. STAT homodimers. C. Produced cytokines.
Table S3. The effects of the parametric changes on the concentration of produced IFN-γ and IL-10 shown in Fig 3.
Fig Parameters
Perturbation
thin line
(optimized)
3A 9n 4163 10−⋅ 4191 10−⋅ 4220 10−⋅
20M 4119 10−⋅ 4149 10−⋅ 4179 10−⋅
3B
5ts 4198 10−⋅ 4247 10−⋅ 4297 10−⋅
18M 7.93 9.913 11.896
8n 481 10−⋅ 4101 10−⋅ 4121 10−⋅
14M 0.068 0.1 0.145
9M 69.185 47.714 32.445
3C 2tr
42.7 10−⋅ 427 10−⋅ 4268 10−⋅
1n 11.842 118.42 1184.2
3D 7M 23.63 10−⋅ 43.63 10−⋅ 63.63 10−⋅
3E 10M 470 447 10−⋅ 947 10−⋅
3F, G 6Q 41 10−⋅ 414 10−⋅ 4209 10−⋅
Table S4. Parameters in the STAT3-STAT4 subsystem and their correspondence to the
parameters in the STAT3-STAT5 subsystem.
3.2 Parameter sensitivity analysis
In the main text, we suggested that IFN-γ to IL-10 switching (Fig 2) is due to the competition
between STAT3 and STAT5 proteins. These model predictions are not obvious from the model
structure due to the following. The structure of our model presented in Fig 1 is symmetrical in
relation to the varied IL-2, i.e. IL-2 activates both STAT3 and STAT5. The model predictions
for the phosphorylated states of STAT3, STAT5 and their homodimers could be swapped in
Fig 2 if the parameters for STAT3 and STAT5 are swapped (Fig S2). However, in this case
there would be no cytokine switching (Fig S2C). Thus the model predictions depend on the
assumed structure of our model as well as on the chosen set of parameters. In this section, we
performed parameter sensitivity analysis (SA) (21, 22) to identify the parameter conditions for
the conclusions to hold.
First, we tested if the IFN-γ to IL-10 and STAT5 to STAT3 switchings are still present when
we vary the parameters compared to their optimal values (set "O3" in Table S3). Quantitatively,
we defined the switching of either IFN-γ to IL-10 or STAT5 to STAT3 as follows. We assumed
that the switching occurs when the 2-fold changes take place, which is typically considered as
significant change in Biology (23, 24). Thus, the changes should include at least 2-fold increase
in IL-10 and STAT3 as well as decrease of the peak of IFN-γ and STAT5 concentrations within
the range of the tested IL-2 concentrations.
We varied the parameters up to 10-fold either side of their values in the optimized set. We used
the Latin Hypercube Sampling (LHS), which is considered as one of the most effective
strategies for sampling the parameters (22, 25). We performed the LHS sampling and checked
the assumed condition for switching for 1000 samples of the optimized parameters. Fig S3
illustrates the probability of the cases, in percent, where the switching of both cytokine and
STAT (C+S+), cytokine but not STAT (C+S-), not cytokine but STAT (C-S+), neither cytokine
nor STAT (C-S-) occurs for 1-10 fold change of the optimized parameters. The obvious result
that follows from Fig S3 is that when the parameters are not perturbed, which corresponds to
the 1-fold change, the probability of the presence of both cytokine and STAT switching is
100%. However, with an increase of the fold change up to ten, the probability for switching of
both cytokine and STAT decreases down to 2.1% (data shown in Table S5). The probability of
either of the cases (cytokine or STAT switching) is almost equal with an increase of the fold
change as it can be seen from Fig S3. Thus, we can conclude that the model with the optimized
parameters demonstrates both IFN-γ to IL-10 and STAT5 to STAT3 switching with higher
probability (more than 50%) when the parameters are perturbed within 2-fold, with modest
probability (between 10% and 50%) when the parameters are perturbed within 3-6-fold and
with low probability (less than 10%) when the parameters are perturbed within 7-fold and
higher.
Next, we identified the most sensitive parameters that have the greatest impact on the steady-
state concentrations of IFN-γ, IL-10, STAT5 and STAT3 for the three concentrations of IL-2:
10-10, 10-6 and 10-1, which correspond to the three assumed T cell phenotypes shown in Fig 1A:
Th1, Th1/Tr1 and Tr1 respectively. We performed the SA using the eFast method (26), because
it was reported as one of the most reliable methods of parameter sensitivity analysis (25). As a
tool for the eFast sensitivity analysis, we used the SBToolbox software (27). We performed the
SA over one order of magnitude of perturbation for 10000 simulations.
Fig S4-Fig S6 illustrate the results of the sensitivity analysis for IFN-γ, IL-10, STAT5 and
STAT3 by the SBToolbox for the non-dimensional IL-2 concentrations [ ] 102 10i −= (Fig S4),
[ ] 62 10i −= (Fig S5) and [ ] 12 10i −= (Fig S6). The bars indicate the sensitivity indices for each
of the parameters of our model. Here we classified the parameters as sensitive if the
corresponding sensitivity index is more than 0.5 .
It can be seen from Fig S4A, Fig S5A and Fig S6A that IFN-γ production is the most sensitive
to the following parameters: 18M , tgg , 8n , 19M , 1tmp and 5ts , which demonstrated high
sensitivity indices (more than 0.5 ) for all the three IL-2 concentrations. Parameters 1n , 2 ,tr 6n
and 2M are sensitive for [ ] 102 10i −= (Fig S4A) and [ ] 62 10i −= (Fig S5A). Parameter 14M is
sensitive only for [ ] 62 10i −= (Fig S5A) and [ ] 12 10i −= (Fig S6A). Another group of IFN-γ-
sensitive parameters that includes 5ts and 6n is involved in the STAT5 pathway activation,
which leads to the production of IFN-γ as shown in Fig 1B and Equations (6). There is also the
third group of IFN-γ-sensitive parameters consisting of 1n , 2tr and 2M that are involved in
the upstream activation of IL-2 receptor described by Equation (1). Finally, IFN-γ is sensitive
to the Michaelis constant of heterodimerization 14M .
The parameter sensitivity analysis revealed that the concentration of IL-10 is the most sensitive
to 10tg and 9n for [ ] 102 10i −= (Fig S4B), [ ] 62 10i −= (Fig S5B) and [ ] 12 10i −= (Fig S6B).
Parameter 9M is the most sensitive for the phosphorylated STAT3 for all the three IL-2
concentrations as it is shown in Fig S4C, Fig S5C and Fig S6C. Two parameters, namely, 5n
and 6Q , showed high sensitivity only for [ ] 102 10i −= (Fig S4C) and [ ] 62 10i −= (Fig S5C).
The concentration of phosphorylated STAT5 is the most sensitive to the total amount of
STAT5, modeled by parameter 5ts , for all the three tested IL-2 concentrations as it is shown in
Fig S4D, Fig S5D and Fig S6D. Parameter 6n demonstrated high sensitivity indices for
STAT5p for [ ] 102 10i −= (Fig S4D) and [ ] 62 10i −= (Fig S5D). This parameter is involved in
the activation of STAT5 pathway as shown in Fig 1B and Equations (6). The concentration of
STAT5p is sensitive to the Michaelis constant of heterodimerization 14M for [ ] 62 10i −= (Fig
S5D) and [ ] 12 10i −= (Fig S6D). Three parameters, namely 1n , 2tr and 2M , are sensitive for
[ ] 102 10i −= (Fig S4D) and [ ] 62 10i −= (Fig S5D) involved in the activation of the IL-2 receptor
and described by Equation (1).
Fig S3. The probability of cases when the switching occurs. The percentage of cases when the switching of both cytokine and STAT (C+S+), cytokine but not STAT (C+S-), not cytokine but STAT (C-S+), neither cytokine nor STAT (C-S-) occurs for 1-10 fold change of the optimized parameters.
Table S5. The percentage of cases when the switching of both cytokine and STAT (C+S+), cytokine but not STAT (C+S-), not cytokine but STAT (C-S+), neither cytokine nor STAT (C-S-) occurs for 1-10 fold change of the optimized parameters.
Fig S4. Parameter sensitivity analysis performed by eFAST for low concentrations of IL-2. Sensitivity indicators for the developed model for IFN-γ (A), IL-10 (B), STAT3 (C) and STAT5 (D) for non-dimensional IL-2 concentration [ ] 102 10i −= .
Fig S5. Parameter sensitivity analysis performed by eFAST for medium concentrations of IL-2. Sensitivity indicators for the developed model for IFN-γ (A), IL-10 (B), STAT3 (C) and STAT5 (D) for non-dimensional IL-2 concentration [ ] 62 10i −= .
Fig S6. Parameter sensitivity analysis performed by eFAST for high concentrations of IL-2. Sensitivity indicators for the developed model for IFN-γ (A), IL-10 (B), STAT3 (C) and STAT5 (D) for non-dimensional IL-2 concentration [ ] 12 10i −= .
4 FACS analysis
Fig S1. FACS analysis for STAT3p- and STAT5p-positive cells. Purified human CD4+ T cells were activated with a-CD3/CD46 antibodies in the presence of 50U/ml
rhIL-2 for 36 hours and percentage of STAT3p- and STAT5p-positive cells assessed by FACS analysis.
Cells were analyzed either as bulk population (bulk CD4+) or after subsorting populations into the three
IFN-γ +/- IL-10-secreting populations generated by CD3/CD46-activation. STAT3p and STAT5p values
were normalized against respective non-phosphorylated STAT protein levels. NA, non-activated, n=3;
shown are mean values ± SD of three independently performed experiments using a different donor
each time.
5 References
1. Herr F, Lemoine R, Gouilleux F, Meley D, Kazma I, Heraud A, et al. IL-2 phosphorylates STAT5 to drive IFN-gamma production and activation of human dendritic cells. Journal of immunology. 2014 Jun 15;192(12):5660-70. PubMed PMID: 24829413.
2. Stumhofer JS, Silver JS, Laurence A, Porrett PM, Harris TH, Turka LA, et al. Interleukins 27 and 6 induce STAT3-mediated T cell production of interleukin 10. Nature immunology. 2007 Dec;8(12):1363-71. PubMed PMID: 17994025. 3. Aaronson DS, Horvath CM. A road map for those who don't know JAK-STAT. Science. 2002 May 31;296(5573):1653-5. PubMed PMID: 12040185. 4. Jiao H, Berrada K, Yang W, Tabrizi M, Platanias LC, Yi T. Direct association with and dephosphorylation of Jak2 kinase by the SH2-domain-containing protein tyrosine phosphatase SHP-1. Molecular and cellular biology. 1996 Dec;16(12):6985-92. PubMed PMID: 8943354. Pubmed Central PMCID: 231702. 5. Collison LW, Delgoffe GM, Guy CS, Vignali KM, Chaturvedi V, Fairweather D, et al. The composition and signaling of the IL-35 receptor are unconventional. Nature immunology. 2012 Mar;13(3):290-9. PubMed PMID: 22306691. Pubmed Central PMCID: 3529151. 6. Delgoffe GM, Vignali DA. STAT heterodimers in immunity: A mixed message or a unique signal? Jak-Stat. 2013 Jan 1;2(1):e23060. PubMed PMID: 24058793. Pubmed Central PMCID: 3670269. 7. Fung MM, Rohwer F, McGuire KL. IL-2 activation of a PI3K-dependent STAT3 serine phosphorylation pathway in primary human T cells. Cellular signalling. 2003 Jun;15(6):625-36. PubMed PMID: 12681450. 8. Liu CY, Tseng LM, Su JC, Chang KC, Chu PY, Tai WT, et al. Novel sorafenib analogues induce apoptosis through SHP-1 dependent STAT3 inactivation in human breast cancer cells. Breast cancer research : BCR. 2013;15(4):R63. PubMed PMID: 23938089. Pubmed Central PMCID: 3978748. 9. Chen Y, Wen R, Yang S, Schuman J, Zhang EE, Yi T, et al. Identification of Shp-2 as a Stat5A phosphatase. The Journal of biological chemistry. 2003 May 9;278(19):16520-7. PubMed PMID: 12615921. 10. Valeyev NV, Hundhausen C, Umezawa Y, Kotov NV, Williams G, Clop A, et al. A systems model for immune cell interactions unravels the mechanism of inflammation in human skin. PLoS computational biology. 2010;6(12):e1001024. PubMed PMID: 21152006. Pubmed Central PMCID: 2996319. 11. Guo L, Junttila IS, Paul WE. Cytokine-induced cytokine production by conventional and innate lymphoid cells. Trends in immunology. 2012 Dec;33(12):598-606. PubMed PMID: 22959641. 12. Strengell M, Matikainen S, Siren J, Lehtonen A, Foster D, Julkunen I, et al. IL-21 in synergy with IL-15 or IL-18 enhances IFN-gamma production in human NK and T cells. Journal of immunology. 2003 Jun 1;170(11):5464-9. PubMed PMID: 12759422. 13. Kasahara T, Hooks JJ, Dougherty SF, Oppenheim JJ. Interleukin 2-mediated immune interferon (IFN-gamma) production by human T cells and T cell subsets. Journal of immunology. 1983 Apr;130(4):1784-9. PubMed PMID: 6403613. 14. Niedbala W, Wei XQ, Cai B, Hueber AJ, Leung BP, McInnes IB, et al. IL-35 is a novel cytokine with therapeutic effects against collagen-induced arthritis through the expansion of regulatory T cells and suppression of Th17 cells. European journal of immunology. 2007 Nov;37(11):3021-9. PubMed PMID: 17874423. 15. Scheeren FA, Diehl SA, Smit LA, Beaumont T, Naspetti M, Bende RJ, et al. IL-21 is expressed in Hodgkin lymphoma and activates STAT5: evidence that activated STAT5 is required for Hodgkin lymphomagenesis. Blood. 2008 May 1;111(9):4706-15. PubMed PMID: 18296629. Pubmed Central PMCID: 2343600. 16. Gilmour KC, Pine R, Reich NC. Interleukin 2 activates STAT5 transcription factor (mammary gland factor) and specific gene expression in T lymphocytes. Proceedings of the National Academy of Sciences of the United States of America. 1995 Nov 7;92(23):10772-6. PubMed PMID: 7479881. Pubmed Central PMCID: 40694. 17. Stewart CA, Metheny H, Iida N, Smith L, Hanson M, Steinhagen F, et al. Interferon-dependent IL-10 production by Tregs limits tumor Th17 inflammation. The Journal of clinical investigation. 2013 Nov 1;123(11):4859-74. PubMed PMID: 24216477. Pubmed Central PMCID: 3809773. 18. Saraiva M, O'Garra A. The regulation of IL-10 production by immune cells. Nature reviews Immunology. 2010 Mar;10(3):170-81. PubMed PMID: 20154735. 19. Sanz E, Hofer MJ, Unzeta M, Campbell IL. Minimal role for STAT1 in interleukin-6 signaling and actions in the murine brain. Glia. 2008 Jan 15;56(2):190-9. PubMed PMID: 18023015. 20. Wurster AL, Tanaka T, Grusby MJ. The biology of Stat4 and Stat6. Oncogene. 2000 May 15;19(21):2577-84. PubMed PMID: 10851056. 21. Chu Y, Jayaraman A, Hahn J. Parameter sensitivity analysis of IL-6 signalling pathways. IET systems biology. 2007 Nov;1(6):342-52. PubMed PMID: 18203580. 22. Saltelli A, Ratto M, Tarantola S, Campolongo F. Sensitivity analysis for chemical models. Chemical reviews. 2005 Jul;105(7):2811-28. PubMed PMID: 16011325.
23. McCarthy DJ, Smyth GK. Testing significance relative to a fold-change threshold is a TREAT. Bioinformatics. 2009 Mar 15;25(6):765-71. PubMed PMID: 19176553. Pubmed Central PMCID: 2654802. 24. Dalman MR, Deeter A, Nimishakavi G, Duan ZH. Fold change and p-value cutoffs significantly alter microarray interpretations. BMC bioinformatics. 2012;13 Suppl 2:S11. PubMed PMID: 22536862. Pubmed Central PMCID: 3305783. 25. Marino S, Hogue IB, Ray CJ, Kirschner DE. A methodology for performing global uncertainty and sensitivity analysis in systems biology. Journal of theoretical biology. 2008 Sep 7;254(1):178-96. PubMed PMID: 18572196. Pubmed Central PMCID: 2570191. 26. Saltelli AT, S.; Chan, K.P.S. A quantitative model-independent method for global sensitivity analysis of model output. Technometrics. 1999;41:39–56. 27. Schmidt H, Jirstrand M. Systems Biology Toolbox for MATLAB: a computational platform for research in systems biology. Bioinformatics. 2006 Feb 15;22(4):514-5. PubMed PMID: 16317076.