a. b.

Supplementary Fig. 1: Synergy of IFN-β and IFN-γ on STAT1

activation. a) Spleen cells C57BL/6 were stimulated with IFN-β or

IFN-γ or both for 0, 15, 30 and 60 mins and phosphorylation of

STAT1 in CD4 T-cells was assessed by flow cytometry. b) IFN-β

requires IFN-γ for optimal STAT1 signaling. C57BL/6 and IFNγR-/-

spleen cells were stimulated with IFN-β 0, 15, 30 and 60 mins and

phosphorylation of STAT1 in CD4 T-cells was assessed by flow

cytometry. *p<0.05.

% p

STAT

1

% p

STAT

1

Duration of Stimulation (mins) Duration of Stimulation (mins)

Nature Medicine: doi:10.1038/nm.2110

IL-17

CD

4

2.5

3.7

0.8

1.7

16.6

17.0

Non TH1 TH17

–

+

IFN

-β

IL-10

CD

4

Non TH1 TH17

–

+

IFN

-β

2.0

2.4

5.2

4.4

4.5

1.2

IFN-γ

CD

4

60.8

53.1

5.7

19.1

0.6

0.9

–

+

IFN

-β

Non TH1 TH17

a.

b.

c.

Supplementary Fig. 2: Direct effect of IFN-β on CD4 T-cells. Purified CD4

T-cells were stimulated with plate-bound anti-CD3 and anti-CD28 in non-

polarizing, TH1 and TH17 conditions in the presence or absence of IFN-β. a)

IL-17, b) IFN-γ and c) IL-10 production in CD4 cells were assessed by flow

cytometry.Nature Medicine: doi:10.1038/nm.2110

Foxp3

Events

26.9

24.9

TregTH1 TH17

1.5

1.4

1.0

1.1

–

IFN

-β

Supplementary Fig. 3: IFN-β does not affect the differentiation of

Foxp3+ Tregs. CD8 depleted spleen cells stimulated with or

without IFN-β in TH1 (IL-12), TH17 (TGFβ/IL-6) and Treg

(TGFβ) conditions and percentage of CD4+ FoxP3+ cells was

assessed by flow cytometry.

Nature Medicine: doi:10.1038/nm.2110

IL-10

CD

4Anti IFN-γ Anti IL-10No Ab

–

+

IFN

-β

14.7

41.7

14.4

34.5

19.9

45.6

IL-10

CD

4

Anti IFN-γ Anti IL-10No Ab

–

+

IFN

-β

5.5

12.9

6.2

13.7

6.2

4.9

a.

b.

Supplementary Fig. 4: Effect of inhibiting IFN-γ or IL-10

signaling during the induction of IL-10 by IFN-β in a) TH1

conditions and b) TH17 conditions with APCs. CD8 depleted

spleen cells were stimulated with or without IFN-β in TH1 or TH17

conditions in the presence or absence of anti-IFN-γ or anti-IL-10. Nature Medicine: doi:10.1038/nm.2110

IFNβ (Units/ml) IFNβ (Units/ml)

a. b.

c.

IFNβ (Units/ml)

Supplementary Fig. 5: Effect of IFN-β on IL-27. a) IFN-β induces

IL-27 in non-polarizing and TH1 conditions but not TH17 conditions.

CD8 depleted spleen cells were stimulated with or without IFN-β in

non-polarizing, TH1, and TH17 conditions and IL-27 was analyzed by

ELISA. (b and c) IFN-β requires IFN-γ to induce IL-27 in non-

polarizing conditions (b) and TH1 conditions (c). CD8 depleted spleen

cells were stimulated with or without IFN-β in non-polarizing, TH1

conditions in the presence or absence of anti-IFN-γ and IL-27 was

analyzed by ELISA. Results for these experiments are the mean ± SD

of triplicates.

IL-2

7 (

pg m

l-1)

IL-2

7 (

pg m

l-1)

IL-2

7 (

pg m

l-1)

Nature Medicine: doi:10.1038/nm.2110

a. b. c. d.

e. f. g. h.

i. j. k. l.

m. n. o. p.

q.

Supplementary Fig. 6: Effect of IFN-β on chemokine/cytokine

profiles in antigen specific TH1 and TH17 differentiation. Lymph

nodes from MOGp35-55 immunized mice were re-stimulated in MOGp35-

55 for 3 days with IL-12 or IL-23 in the presence or absence of IFN-β.

Chemokines and cytokines were assessed by Luminex multiplex

analysis or ELISA.

Nature Medicine: doi:10.1038/nm.2110

Supplementary Fig. 7: Effects of IFN-β treatment in different EAE

models. (a and b) Clinical scores from SJL mice with passive EAE induced

by adoptive transfer of (a) TH1 and (b) TH17 cells that were treated with

rmIFN-β or PBS every second day from day 0 to 10 post transfer (n=6 mice

per group). (c and d) Clinical scores from (c) C57BL/6 and (d) IFNγR-/-

mice treated daily with rmIFN-β of PBS from day 10 to day 17 post EAE

induction (n=4 to 5 mice per group). (e and f) Clinical scores from (e)

C57BL/6 and (f) IFNγR-/- mice treated daily for 10 days with rmIFN-β or

PBS beginning at disease score of 2 or 3 (n=6 to 9 mice per group).

Treatment doses indicated with arrows. *p<0.05.

Clin

ical

sco

re

Time after induction (d)

Clin

ical

sco

re

Time after induction (d)

e. f.

ControlrmIFNβ

ControlrmIFNβ**

*****

Clin

ical

sco

reC

linic

al S

core

Clin

ical

sco

reC

linic

al S

core

Time after induction (d) Time after induction (d)

Time after initiation of therapy (d) Time after initiation of therapy (d)

ControlrmIFNβ

ControlrmIFNβ* ** **

a. b.

c. d.

Nature Medicine: doi:10.1038/nm.2110

C5

7B

L/6

IFNγR

-/-

IL-17

Control rmIFN-β

IFN

-γ

17.0 1.5

4.7

5.0 0.3

0.0

7.3 0.8

5.5

10.2 4.5

4.3

IL-17

IFN

-γ

Control rmIFN-β

8.0 0.0

3.4

5.1 0.0

0.0

4.0 0.0

2.9

7.3 2.5

8.7

C5

7B

L/6

IFNγR

-/-

a. b.

Supplementary Fig. 8: Frequencies of the IFN-γ and IL-17

producing CD4 T-cells in the spinal cords (a) brainstem/cerebellum

(b) 12 days post induction of EAE in C57BL/6 or IFNγR-/- mice

treated with rmIFN-β or PBS.

Nature Medicine: doi:10.1038/nm.2110

a.

b.

c.

600

400

200

0600 8004002000

600

400

200

01,500 2,0001,0005000

600

400

200

01,500 2,0001,0005000

4,000

3,000

2,000

03,000 4,0002,0001,0000

1,000

Responders Non-Responders Healthy

4,000

3,000

2,000

01,500 2,0001,0005000

1,000

3,000

2,000

01,500 2,0001,0005000

1,000

400

300

200

0600 8004002000

100

400

300

200

0600 8004002000

100

800

600

400

0600 8004002000

200

IL-1

7F

IL-1

7F

IL-1

7F

IL-1

7F

IL-1

7F

IL-1

7F

IFN-β IFN-β IFN-β

MIP1-β MIP1-β MIP-β

MIP1-β MIP1-β MIP-β

IFN

-β

IFN

-β

IFN

-β

Supplementary Fig. 9: Correlation of a) IL-17F vs IFN-β levels,

b) IL-17F vs MIP1β levels and c) IFN-β vs MIP1β in serum from

responders, non-responders and healthy controls. R2 values close

to 1 demonstrate that the cytokines are positively correlated.

Nature Medicine: doi:10.1038/nm.2110

Supplementary Table 1. Demographic and clinical characteristics of patients

with relapsing remitting multiple sclerosis and their clinical response to IFN-β

therapy.

Responder Non-responder

Number 12 14

Female/Male (n) 10/2 11/3

Median age at onset (yr) 27.6 [24.5; 35.8] 26.7 [19.3; 36.0]

Median age at start IFN-β (yr) 33.5 [30.3; 39.5] 33.0 [23.0; 37.8]

Median EDSS score around start IFN-β 2.5 [2.0; 3.5] 2.5 [1.8; 4.3]

Relapse rate in 2 yrs before start IFN-β 2 [2-3] 2 [1-3]

Relapse rate in 2 yrs after start IFN-β 0 [0; 0] 2 [1.5; 2.0]

Steroid interventions before start IFN-β (n) 0 [0; 2] 1 [0; 3]

Steroid interventions after start IFN-β (n) 0 [0; 0.5] 2 [1; 3]

Duration of IFN-β treatment (months) 80 [46; 141] 56 [38; 104]

Avonex 4 5

Rebif 2 8

Betaferon 6 1

Median values are shown with 25 and 75 percentiles.

IFN-β = Interferon-beta.

EDSS = Expanded Disability Status Scale

Nature Medicine: doi:10.1038/nm.2110

Supplementary Methods.

Mice. SJL, and Ifngr1tm/Agt/J (Ifngr1–/–) mice were purchased from Jackson Laboratory and

C57BL/6 mice were purchased from Jackson Lab or NCI-Fredrick bred at Stanford and/or UAB.

B6 Stat1–/– mice were provided by R. Lorenz (UAB). All animals were housed and treated in

accordance to with institutional guidelines and approved by the IACUC.

EAE induction. Age and sex matched C57BL/6 and Ifngr1–/– mice were induced with EAE by an

immunization 150 � g of MOG p35–55 (Biosynthesis) emulsified in CFA followed by an

intraperitoneal injection of with 500 ng of Bordetella pertussis toxin (Difco Laboratories) in PBS

at the time of, and two days following immunization.

The typical clinical manifestation of EAE in C57BL/6 mice is a progressive ascending

paralysis which starts in the tail and leads to forelimb paralysis. In mice with decreased IFN-�

signaling, EAE symptoms are atypical and characterized by defects in proprioception with axial

rotatory movement and ataxia with little hind limb paralysis1,2. Typical EAE symptoms monitored

daily using a standard clinical score ranging: 1) Loss of tail tone, 2) incomplete hind limb paralysis,

3) complete hind limb paralysis, 4) forelimb paralysis, 5) moribund/dead. Atypical EAE

symptoms were scored as follows: 1) hunched appearance, slight head tilt, 2) severe head tilt, 3)

slight axial rotation/staggered walking, 4) severe axial rotation/spinning, 5) moribund/dead. In our

experiments, we observed that 60-80% of the IFN-� R–/– mice exhibited atypical EAE (scoring

described in the methods) and this was not affected by IFN-� treatment.

Naïve Human CD4 T-cell Isolation. We obtained peripheral blood mononuclear cells from

healthy donors (Stanford Blood Center) by centrifugation through Ficoll (Histopaque 1077;

Sigma). CD4+ T cells were isolated by magnetic bead depletion of CD19+, CD14+, CD56+, CD16+,

CD36+, CD123+, CD8+, T cell receptor-� and T cell receptor-δ positive and glycophorin A–positive

Nature Medicine: doi:10.1038/nm.2110

cells on an AutoMACS instrument (Miltenyi Biotec). Naive CD45RA+ T cells were obtained by

depletion with anti-CD45RO and anti-CD25 magnetic beads (Miltenyi Biotec).

References

1. Wensky, A.K. et al. IFN-gamma determines distinct clinical outcomes in autoimmune

encephalomyelitis. J Immunol 174, 1416-23 (2005).

2. Stromnes, I.M., Cerretti, L.M., Liggitt, D., Harris, R.A. & Goverman, J.M. Differential

regulation of central nervous system autoimmunity by T(H)1 and T(H)17 cells. Nat Med

14, 337-42 (2008).

Nature Medicine: doi:10.1038/nm.2110