Suppl. Figure 1 A

CD31-

CD45-

CD34-

CD14-

CD29+

CD90+

CD73+

CD166 +

CD105 +

Suppl. Fig. 1 Characterization of BM-MSC surface markers. (A) The morphology of bone marrow-derived MSCs was photographed under microscope. (B) Bone marrow-derived MSCs were subjected to flow cytometry analysis to assess the expression of surface markers. The isolated BM-MSCs expressed the BM-MSC-specific markers (CD29, CD73, CD90, CD105, and CD166), but not the endothelial and hematopoietic cell markers (CD31, CD34, and CD45), as well as negative for macrophage-specific marker (CD14).

B

0

20

40

60

80

100

120

140

Control Low CM High CM

% o

f mig

rato

ry c

ells

BM-MSC-2

*

0

20

40

60

80

100

120

140

Control Low CM High CM

% o

f mig

rato

ry c

ells

BM-MSC-3

*

0

20

40

60

80

100

120

Control Low CM High CM

% o

f mig

rato

ry c

ell

EM-MSC-1

* *

0

20

40

60

80

100

120

Control Low CM High CM

% o

f mig

rato

ry c

ell

EM-MSC-2

* *

Suppl. Figure 2

Suppl. Fig. 2 MSC derived from different donors inhibited LAC cell migration. To dissect that the inhibitory effect of MSCs on LAC cell migration is not unique to a specific strain of MSCs, we collected 2 other BM-MSCs (BM-MSC-2 and 3; (1, 2)) from 2 donors and 2 endometrium MSCs (EM-MSC-1 and 2; (3, 4)) from another 2 donors for the TransWell migration assay. Conditioned medium (CM) were collected after 24-hour incubation with 5�105 (low) or 2�106 (high) of each MSC cells. CL1-5 cells were subjected to the migration assay in the presence of CM or control medium. The number of migratory cells were counted and presented as percentages relative to the controls. As shown in the figure, the migratory CL1-5 cells was inhibited by both BM-MSC- and endometrium MSC-derived CM in a dose-dependent manner, and the number of migratory cells was reduced up to 70% when treated with high concentration of CM.

OSM

18S

BM-MSC-1

0

5

10

15

20

25

30

35

18S OSMCt

qPCR of BM-MSC-1

ELISA

0

2

4

6

8

10

12

14

OSM

con

cent

ratio

n (p

g/m

l)

Suppl. Figure 3

A B

C

Suppl. Fig. 3 MSC expresses and secretes OSM (A) The medium supernatants of BM-MSCs derived from 3 individuals, along with the control medium, were subjected to an ELISA assay to assess the level of secreted OSM from MSC. (B) BM-MSC was analyzed by quantitative real-time PCR to assess the mRNA expression of OSM. The OSM mRNA was detected in the BM-MSC with a Ct value around 30. (C) The amplified OSM and robosomal 18S mRNA were visualized in agarose gel.

Suppl. Figure 4

0

5

10

15

20

25

30

35

0 2.5 5 10 20

Prol

ifera

tive

chan

ge (F

old)

OSM (ng/ml)

CL1-5 24 h

72 h

0

5

10

15

20

25

30

35

0 2.5 5 10 20

Prol

ifera

tive

chan

ge (F

old)

OSM (ng/ml)

A549 24 h

72 h

* *

*

*

Suppl. Fig. 4 OSM inhibited the proliferation of CL1-5 and A549 LAC cells. CL1-5 and A549 cells were seeded in 96-well culture plates with the density of 1000 cells/well and treated with different concentration of OSM as indicated for 24 and 72 hours. The cell density was assessed by WST-1 assay and the presented as fold change in comparison to non-treated control.

0

20

40

60

80

100

120

140

0 5 20

% o

f mig

rato

ry c

ell

OSM (ng/ml)

H2170

0

20

40

60

80

100

120

140

0 5 20

% o

f mig

rato

ry c

ell

OSM (ng/ml)

H520

* *

*

Suppl. Figure 5

Suppl. Fig. 5 OSM inhibited the migratory capability of lung squamous cell carcinoma. The lung squamous carcinoma cell lines, H2170 and H520 (acquired from ATCC), were subjected to a TransWell migration assay with or without the treatment of OSM (5 and 20 ng/ml).

Suppl. Figure 6

Migration assay

pSTAT1(y701) Short exposure

STAT1

pSTAT1(y701) Long exposure

�TUBULIN

shSC

shST

AT1-

1

shST

AT1-

2

shSC

shST

AT1-

1

shST

AT1-

2

Control OSM 20 ng/ml

0.5 hr

STAT3

NANOG

SNAIL

SLUG

�TUBULIN

shSC

shST

AT3-

1

shST

AT3-

2

shSC

shST

AT3-

1

shST

AT3-

2

Control OSM 20 ng/ml

24 hr

A B

C

0

50

100

150

200

250

300

% o

f mig

rato

ry c

ells

A549

CL1-5

*

*

*

*

Suppl. Fig. 6 STAT3 did not mediate the OSM-dependent inhibition of EMT markers. (A) CL1-5 cells with stable knockdown of STAT1 were treated with or without OSM (20 ng/ml) for 0.5 hours, followed by Western blotting analysis of the indicated proteins. (B) A549 and CL1-5 cells with or without stable knockdown of STAT1 were subjected to a TransWell migration assay. The graph presented the relative migratory cell number in comparison to scrambled shRNA-transfected cells, respectively. (C) CL1-5 cells with stable knockdown of STAT3 were treated with or without OSM (20 ng/ml) for 24 hours, followed by Western blotting analysis of the indicated proteins.

0

0.2

0.4

0.6

0.8

1.0

Rela

tive

Ratio

of

Slug

mRN

A

0

1

2

3

4

5

Rela

tive

Ratio

of

E-ca

dher

in m

RNA

* *

*

*

*

Suppl. Figure 7

Control

CM

OSM

CM+aOSM

Control

CM

Ki67 Hoechst33258 Merge A B

D

C Control CM OSM CM+aOSM

Vimentin

Hoechst33258

Merge 20 �m 20 �m

Suppl. Fig. 7 OSM mediated the MSC-CM-dependent suppression of tumor proliferation and EMT in vivo. (A) Tumors excised from mice subcutaneously transplanted with A549 cells and treated with control medium, CM, OSM (20 ng/ml), or OSM-neutralized CM (CM+aOSM). (B) Xenograft tumors treated with control or conditioned medium were subjected to immunofluorescence staining with Ki-67 and Hoechst33258. Conditioned medium suppressed the proliferation of xenograft tumor, compared with the control. (C) Tumor tissues from A549-transplanted mice treated with CM, OSM, or CM+aOSM were subjected to immunofluorescent staining for Vimentin and Hoechst33258. (D) Immunocompromised mice were transplanted through tail veins with A549 cells and then treated with control medium (100 �l), CM (100 �l), OSM (20 ng in 100 �l of control medium), or OSM-neutralized CM (CM+aOSM; 100 �l of CM preincubated with 2 �g/ml aOSM for 1 hour) through intraperitoneal injection. Mice were sacrificed 3 months after transplantation. Tumor sections from lungs were subjected to quantitative real-time PCR to assess the mRNA expression level of E-cadherin and Slug.

Targets of Antibodies Supplier

Oct4 Cell Signaling

Nanog Cell Signaling

Snail Cell Signaling

Slug Cell Signaling

E-cadherin BD

Vimentin AbD

Cytokeratin 18 Cell Signaling

Stat1 Cell Signaling

�-Tubulin Sigma

�-Actin NOVUS

OSM R&D Systems

Suppl. Table 1

Suppl. Table 1 List of antibodies used in the report.

Supplementary Methods

Isolation, expansion, and characterization of human MSCs

In collaboration with Dr. OK Lee, human bone marrow was obtained from 3 healthy

donors during fracture surgeries. All samples were collected with informed consent.

The MSCs were isolated following a previously reported protocol (1, 2, 5, 6) which

had been approved by the Institutional Review Board at the Taipei Veteran General

Hospital. Briefly, after negative immune-selection and limiting dilution from the

mononuclear fraction of the bone marrow aspirates, colonies obtained from signle

cells were cultured and expanded in T75 flasks with MesenPRO RS medium

(Invitrogen, Carlsbad, CA, USA). Cells were sub-cultured at the dilution of 1:3 when

reaching 60 - 70% confluence. The surface immune phenotype and multi-potency of

MSCs were confirmed as previously described (1, 2, 5, 6) prior to the commencement

of this study. The isolated BM-MSCs were maintained and applied to the experiments

for up to 12 passages of subculture. The endometrium MSCs were obtained from Dr.

SH Chiou with collaboration (3). In brief, the endometrial cells were obtained using

enzymatic digestion and by mechanical means. Single cells were separated from the

debris by a 40-�m sieve (Becton Dickinson, Franklin Lakes, NJ). MSCs were then

obtained by negative immune-selection and characterized as previous reported (3).

Immunoblotting assay

Total cell lysate extraction and immunoblotting analysis were performed as described

(7). Aliquot protein sample was boiled at 95°C for 5 min and separated on 10%

SDS-PAGE. The proteins were transferred to PVDF membrane (Pall.). Primary and

secondary antibodies were added as indicated. Reactive protein bands were detected

by the ECL detection system (Millipore). Used antibodies are listed in Supplemental

Table 1.

Flow cytometry analysis

To assess cell surface marker expression, MSCs were detached, washed, resuspended

in 100 �l PBS, and stained with fluorescein-conjugated anti-CD29, CD37, CD90,

CD105, CD166, CD14, CD31, CD34, CD45 antibodies (Biolegend) individually.

Cells were analyzed with a BD FACS Calibur apparatus (BD Biosciences).

shRNA

RNAi reagents were obtained from the National RNAi Core Facility located at the

Institute of Molecular Biology / Genomic Research Center, Academia Sinica,

supported by the National Core Facility Program for Biotechnology Grants of NSC

(NSC 100-2319-B-001-002). The target sequence of shSTAT1-1 (ID:

TRCN0000004267) and shSTAT1-2 (ID: TRCN0000280021) are

CTGGAAGATTTACAAGATGAA and CTGGAAGATTTACAAGATGAA,

respectively.

References

1. Lee KD, Kuo TK, Whang-Peng J, Chung YF, Lin CT, Chou SH, et al. In vitro

hepatic differentiation of human mesenchymal stem cells. Hepatology.

2004;40:1275-84.

2. Ho JH, Tseng TC, Ma WH, Ong WK, Chen YF, Chen MH, et al. Multiple

intravenous transplantations of mesenchymal stem cells effectively restore

long-term blood glucose homeostasis by hepatic engraftment and beta cell

differentiation in streptozosin-induced diabetic mice. Cell transplantation. 2011.

3. Chen YJ, Li HY, Chang YL, Yuan CC, Tai LK, Lu KH, et al. Suppression of

migratory/invasive ability and induction of apoptosis in adenomyosis-derived

mesenchymal stem cells by cyclooxygenase-2 inhibitors. Fertility and sterility.

2010;94:1972-9, 9 e1-4.

4. Huang TS, Hsieh JY, Wu YH, Jen CH, Tsuang YH, Chiou SH, et al. Functional

network reconstruction reveals somatic stemness genetic maps and

dedifferentiation-like transcriptome reprogramming induced by GATA2. Stem

Cells. 2008;26:1186-201.

5. Lee OK, Ko YC, Kuo TK, Chou SH, Li HJ, Chen WM, et al. Fluvastatin and

lovastatin but not pravastatin induce neuroglial differentiation in human

mesenchymal stem cells. Journal of cellular biochemistry. 2004;93:917-28.

6. Shih YR, Tseng KF, Lai HY, Lin CH, Lee OK. Matrix stiffness regulation of

integrin-mediated mechanotransduction during osteogenic differentiation of human

mesenchymal stem cells. Journal of bone and mineral research : the official journal

of the American Society for Bone and Mineral Research. 2011;26:730-8.

7. Kao CL, Huang PI, Tsai PH, Tsai ML, Lo JF, Lee YY, et al. Resveratrol-induced

apoptosis and increased radiosensitivity in CD133-positive cells derived from

atypical teratoid/rhabdoid tumor. Int J Radiat Oncol Biol Phys. 2009;74:219-28.