images.nature.com100 nm) bkm 120 (500 nm) ehicle leq 506 0 bkm 120 (100 nm) leq+bkm (100 nm) (500...

1

(a) Histogram of pten expression from first database analyzed 7. Black bars

indicate samples considered to express low pten (22% of tumors); crosshatched

bars indicate high pten samples. (b) GBMs with decreased pten exhibit

increased activation of the Shh signaling pathway; robust z-scores for gli1 and

gli2 are higher in tumors expressing low pten compared with others

(Supplementary Figure 1a), (*) p = 0.005. (c) Copy number analysis of pten

status (Broad database, TCGA 8). Red line indicates cut-off used for loss of pten

Vehicle LDE 225 BKM 120 Combo

hBT1

12

0.4

0.5

0.6

0.7

0.8

0.9

1

1.1

1.2

1.3

1.4hBT 112

(PTEN-Deficient)

Vehicle LDE 225 BKM 120(100 nM)

Combo

*

1.4

0.4

0.2

0

0.6

0.8

1

1.21.4

0.4

0.2

0

0.6

0.8

1

1.2

1.3

0.60.50.4

0.70.8

0.9

1.11

1.2

*BK

M 1

20(1

00 n

M)

Cy

BKM

120

(5

00 n

M)

Cy+

BKM

(1

00 n

M)

Cy+

BKM

(5

00 n

M)

BKM

120

(1

00 n

M)

LEQ

506

BKM

120

(5

00 n

M)

LEQ

+BKM

(1

00 n

M)

LEQ

+BKM

(5

00 n

M)

Vehi

cle

LDE

225

GD

C-0

941

(500

nM

)

LDE+

GD

C

(500

nM

)

GD

C-0

941

(100

nM

)

LDE+

GD

C

(100

nM

)

Vehi

cle

Rel

ativ

e vi

abilit

y

hBT 70 Alternative PI3K inhibitor hBT 70 Alternative Shh pathway inhibitors

Rel

ativ

e vi

abilit

y

caSupplementary Figure 1

b

e

Vehicle LDE 225

PTENParental

pten gli1 gli2

BKM 120 Combo

DAY 7DAY 14

DAY 21 (7d post-treatment) DAY 28 (14d post-treatment)

hBT112 Over Time

0.6

0.8

0

1.2

1.4

1.6

1.0

0.5

0

1.5

2.0

2.5

1.8

* *

* *

**

n.s.

f

d

g h

i

Relative PTEN Copy Number

Freq

uenc

y

hBT# PTEN Status by WB

PTEN Status by aCGH

EGFR (aCGH) EGFRvIII (IHC) IDH1 IDH2

DNA change TP53 Protein

Change Variant Class 112 Not Detected Normal copy Amplified WT WT g.7579548G>A p.P47S Missense Mutation

g.7579801G>C None 5'UTR 145 Not Detected Loss normal copy WT WT g.7579619G>T None Intron 188 Present Loss Amplified No WT WT g.7579619G>T None Intron 239 Present Normal copy Gain WT WT g.7579801G>C None 5'UTR

TP53 TP53

50 100

150

200

250

300

350

400

450

500

550

600

650

700

750

800

850

900

950

1000

1050

1100

1150

1200

0

5

10

15

Num

ber o

f GB

Ms

pten Expressionpten loss pten neutral

Vehi

cle

* * *

Fold

cha

nge

rela

tive

to v

ehic

le

0.2

-0.2

0.4

0.6

0.8

1.2

1.4

1

0Low pten High pten

**

Robu

st Z

-sco

re

gli 1gli 2

Nature Medicine: doi:10.1038/nm.3328

2

(–0.3 threshold represents minimum in copy number distribution). (d) hBT112

neurospheres were imaged and viability was quantified by Trypan Blue exclusion

(n = 3). Error bars: +/– SEM. (*) F = 12, DFn = 1, DFd = 20, p = 0.0025 by two-

way ANOVA factorial interaction. Scale bar = 1 mm. (e) Combination treatment

of alternative PI3K inhibitor GDC-0941 with LDE225 reduces cell viability in

hBT70 monolayer (n = 3). Error bars: +/– SEM. F = 2.72, DFn = 2, DFd = 12, p =

0.1065 by two-way ANOVA factorial interaction. (f) Combination treatment of

BKM120 with Smo-inhibitors, Cyclopamine, or LEQ506 reduces cell viability in

hBT70 neurospheres (n = 3). Error bars: +/– SEM. (*) F = 5.60, DFn = 2, DFd =

2, p = 0.0192 for Cyclopamine and (*) F = 4.99, DFn = 2, DFd = 12, p = 0.0265

for LEQ506 for interaction with BKM120, by two-way ANOVA factorial interaction.

(g) hBT112 neurospheres were assayed for viability (n = 3). Error bars: +/– SEM,

(*) p < 0.05 and (**) p < 0.01 by one-way ANOVA. (h) Table of mutation status of

hBT112, 145, 188 and 239 GBMs. Absence of PTEN protein is the only known

factor common to lines responsive to combination therapy. (i) Acute

overexpression of PTEN in hBT112 cells decreases expression of gli1 and gli2 (*,

p = .005 for gli1 and p = .00006 for gli2 by Student t-test, mRNA normalized to

gapdh), (n = 3). Results shown are normalized to no virus.


3

(a) hBT145 cells were treated with vehicle, LDE225 (1 µM), BKM (100 nM) or

combination therapy. Apoptotic index measured by activated caspase-3 is

increased with combination therapy (n = 3). Error bars: +/– SEM, (*) p < 0.01 by

chi-square analysis. (b) Combination treatment leads to a significant reduction of

cell size as measured by M-phase cells, normalized to cells treated with vehicle.

Data represent means from > 50 cells per condition (n = 3). Error bars: +/– SEM,

(*) p = 0.002 by t-test. (c) Hematoxylin/Eosin staining of hBT112 tumor

xenografts at end treatment shows decreased cell size and increased pyknotic

nuclei with combination therapy. Scale bar = 50 µM. (d) Combination therapy

leads to increased deaths shortly post-mitosis; a cell divides and both daughter

cells rapidly die (75% die within one hour after division and 90% die within two

hours after division). All mitoses were analyzed; the proportion wherein both

daughters rapidly die is shown. Data represent means from > 52 cells per

0.88

0.9

0.92

0.98

0.96

0.94

1

1.06

1.04

1.02

*

5

10

25

20

15

*

0

Mitotic Catastrophe

Per

cent

of m

itose

s

Rel

ativ

e ce

ll si

ze

Supplementary Figure 2

b

d

a

0

0.2

0.4

0.6

0.8

1

hBT1

45

Act

ivat

ed C

aspa

se3+

cel

ls (%

) *

c

H&

E

Vehicle BKM 120 ComboLDE 225





4

condition, from three independent real time imaging experiments, (*) p < 0.05 by

chi-square analysis.

(a) Immunoblots of phosphorylated and total AKT in hBT70 cells treated for 24

hours with LDE225 (1 µM), BKM120 (500 nM) or combination, Actin: loading

control (n = 3). (b) Real-Time RT PCR for gli1 mRNA normalized to gapdh (n =

2). (c) Immunoblots of phosphorylated and total S6K and S6 proteins from

hBT70. Actin: loading control. (d) Phosphorylated S6 decreases in hBT112 cells

after treatment with PF-4708671, as shown by integrated intensity of staining

over number of cells counted. The effect is similar to that seen with combination

therapy. Data represent means of > 2000 cells (n = 3). Error bars: +/– SEM, (*) p

= 0.017, (**) p = 0.0008 by z-test, compared to one. (e) Immunoblot of cyclinD1

gli1 hBT70

pAKT (Ser)

pAKT (Thr)

AKT

Actin

hBT 70 (PTEN-Deficient)Vehicle LDE 225 BKM 120 Combo

Actin

cyclinD1

Actin

cyclinD1


a b

Supplementary Figure 3

0

0.2

0.4

0.6

0.8

1

1.2

pS6

S6

Actin

pS6 kinase

S6 kinase



BDNF

PION

TNFR S1B

KCNH2

RARRES2

ISG15

c

ed

hBT 70 (PTEN-Deficient)



Fold

cha

nge

rela

tive

to v

ehic

le

0.4

0.5

0.6

0.7

0.8

0.9

1.0

1.1

Vehicle ) Combo

gni

niats f

o ytisnet

ni evitaleR

phospho-S6

* **

f



5

from treated hBT70 and hBT112 cells. Actin: loading control. (f) hBT70 and

hBT112 cells were treated with vehicle, LDE225 (1 µM), BKM120 (500 nM), or

combination. mRNA expression was determined with Affymetrix Human Genome

U133A 2.0 chips. Results are shown as heat map generated with unsupervised

clustering. Data represent means of two independent experiments per line.

These genes are significantly altered in combination therapy in both GBM lines.

The full data set is in Table1.

Supplementary Table 1

Affymetrix data for hBT70 and hBT112 cells treated with vehicle control, LDE225

(1 µM), BKM120 (500 nM), or the combination of both for 5 d.


images.nature.com100 nm) bkm 120 (500 nm) ehicle leq 506 0 bkm 120 (100 nm) leq+bkm (100 nm) (500...

Documents