A

B

p-AKT S473

ERK

p-ERK

p-AKT T308

AKT

H630 CaR-1

sh KRAS

KRAS

SW1463 Lovo Gp5d HCT-116 LS-174TT84

KRAS/PIK3CA mutant

KRAS wild type

KRAS mutant

SW948HCT-15

p-AKT S473

ERK

p-ERK

p-AKT T308

AKT

sh KRAS

KRAS

HT55

Supplemental Figure 1. KRAS is required for activation of MEK-ERK in KRAS mutant

colorectal cancers

KRAS mutant and KRAS/PIK3CA double mutant (A) and KRAS wt (B) colorectal cancer cells

were infected with two shRNA targeting KRAS (B, C) or control shRNA. 3 days following

infection, protein lysates were prepared and probed with the indicated antibodies.

Gp5dLS-174T T84

KRAS/PIK3CA mutant

p-AKT S473

ERK

p-ERK

p-AKT T308

AKT

S6

p-S6

p-EGFR

EGFR

HCT-116 SW948HCT-15

SW837 SW1463 Lovo

KRAS mutant

p-AKT S473

ERK

p-ERK

p-AKT T308

AKT

S6

p-S6

p-EGFR

EGFR

SW620

N.D

N.D

Supplemental Figure 2. EGFR does not regulate PI3K-AKT signaling in KRAS mutant

colorectal cancers

The indicated cell lines, grouped according to mutational status, were treated with Gefitinib

(Gef) 1μM or Cetuximab (Cet) 10 μg/mL for 6 hours. Lysates were probed with the indicated

antibodies.

H630 CaR-1

p-AKT S473

ERK

p-ERK

p-AKT T308

AKT

p-EGFR

EGFR

KRAS wild type

HT55 CoCM-1 CCK-81 Difi

Supplemental Figure 3. EGFR inhibition suppresses MEK-ERK signaling pathway in KRAS

wildtype cancers

KRAS wildtype cancer cell lines were treated with 1μM of gefitinib (Gef) or 10 μg/mL of

cetuximab (Cet) for 6 hours, and lysates were probed with the indicated antibodies.

p-AKT S473

AKT

sh KRAS

Cet (-)

ERK

p-ERK

Cet (+)

SW1463

KRAS

-

SW837

Dox - + +

- + - +Cet

p-AKT S473

AKT

ERK

p-ERK (short)

KRAS

p-ERK (long)

Supplemental Figure 4. KRAS exerts a more dominant regulation on ERK signaling than

EGFR in SW1463 and SW837 cell lines. (Left) SW1463 cells were infected with one of two

independent shRNA targeting KRAS (B, C) or control shRNA. 3 days following infection, cells were

treated with vehicle or cetuximab (10 μg/mL) for 6 hours. The cells were lysed and western blots

were probed with the indicated antibodies. (Right) SW837 cells infected with lentivirus expressing

a doxcycyline-inducible KRAS shRNA were cultured in the presence or absence of doxycycline (20

ng/mL) for 72 hours. Then cells were treated with cetuximab for 6 hours. The cells were lysed and

western blots were probed with the indicated antibodies.

p-AKT S473

p-ERK

KRAS

ERK

AKT

- +Dox - +

SW620 (0.5%) SW620 (5%)

- + - + Dox

HCT-116 (0.5%) HCT-116 (5%)

- + - + - + - +

p-AKT S473

p-ERK

KRAS

ERK

AKT

Supplemental Figure 5. Restored expression of KRAS rescues the effects of KRAS

knockdown on downstream signaling.

SW620 and HCT-116 cells harboring doxycyline-inducible KRAS or control shRNA (Figure

1F) were infected with GFP-expressing lentivirus encoding KRAS(12V) or control cDNA and

sorted for low GFP fluorescence intensity by FACS. Since the KRAS shRNA targets the 3’UTR,

the KRAS expression was restored in the cells infected with the KRAS(12V) expressing

constructs. Stable polyclonal cell lines were then cultured in the presence or absence of

doxycycline (10 ng/mL) in either full serum (5% FBS) or low serum (0.5% FBS) for 72 hours.

Lysates were probed with the indicated antibodies.

Dox-KRAS

KRAS-withdrawal

48hr

p-ERKp-AKT

B

p-ERK

ERK

AKT

p-AKT S473

p-AKT T308

24 48

Dox –

withdrawal

(hr)

A

Supplemental Figure 6. KRAS knockdown does not suppress PI3K signaling

(A, B) Tet-op-mutant Krasmice were induced to develop lung tumors with doxycycline. The

doxycycline was removed and lungs were harvested 24 and 48 hours later. The lungs were

assessed by western blot analyses (A) and immunohistochemical analyses (B) with the

indicated antibodies. Scale bar, 100μM.

p-RSK

RSK

p-ERK

ERK

Gp5d SW1463

p-S6

S6

SW837

Supplemental Figure 7. MEK-ERK signaling regulates S6 phosphorylation in multiple KRAS

mutant cancer cell lines. Cells were treated with AZD6244 1μM for 6 hours. Protein lysates were

blotted with the indicated antibodies.

Gp5d T84LS-174T

p-AKT S473

ERK

p-ERK

p-AKT T308

AKT

S6

p-S6

KRAS/PIK3CA mutant

HCT-15 SW948

SW620 SW1463

p-AKT S473

ERK

p-ERK

p-AKT T308

AKT

S6

p-S6

Lovo

KRAS mutant

p-AKT S473

ERK

p-ERK

p-AKT T308

AKT

S6

p-S6

T84

p-EGFR

EGFR

B

A

Supplemental Figure 8. Mechanisms of PI3K

activation in KRAS mutant colorectal cancers

(A) Cells were treated with either DMSO (-) or the

indicated drug(s) for 6 hours. Drug concentrations are the

same as in Figure 4A. Protein lysates were immunoblotted

and probed with the indicated antibodies.

(B) PI3K suppression requires IGF-IR and EGFR inhibition

in the presence of MEK inhibitors in T84 cells. Cells were

treated with either DMSO (-) or the indicated drug

combinations for 6 hours. Drugs were used at 1μM.

Protein lysates were probed with indicated antibodies.

SW1463

GP5d

SW620SW837

KRAS mutant

KRAS/PIK3CA mutant

B

A

Sub-G

1

popula

tion (

%)

0

5

10

15

20

25

30

T84LS-174T HCT-116

Supplemental Figure 9. Decreased cell viability is induced by combined inhibition of

IGF-IR (or MET) and MEK in KRAS mt colorectal cancers

(A) Cell viability data for the individual KRASmutant and KRAS/PIK3CA mutant cell lines

using the indicated drugs and combinations. Each concentration was assessed in sixplicate,

and the averages and standard deviations are shown. All cell lines except LS-174T showed

synergy with IGF-IR and MEK inhibitors.

(B) A combination of MET and MEK inhibitors is the most effective in Lovo cells. (left) Lovo

cells were assessed by a 72-hour survival assay in increasing concentrations of the indicated

drugs and combinations. Combined inhibition of MET and MEK showed synergy comparing

to single treatments. (right) Lovo cells were incubated in the presence of the indicated

drug(s), and the percentage of cells with sub-G0/G1 DNA content was determined by

propidium iodine staining and fluorescence-activated cell sorting analyses.

0

0.5

1

1.5

2

2.5

3

3.5

4

0 1 2 3>50% <50%

IC50 (mM

)Gef/NVP

AZD/NVP

p < 0.001

Supplemental Figure 10. The effects of inhibitor combinations on the activation MEK/ERK

and PI3K/AKT pathways correlates with their efficacy. The levels of phosphorylated ERK and

AKT after treatment with gefitinib and NVP-AEW541 or AZD6244 and NVP-AEW541 were

quantified for each cell line examined (raw data shown in Figure 4A and Supplemental Figure 8A).

The average of suppression of AKT and ERK phosphorylation was calculated in each cell lines.

Using a cutoff of 50% reduction of phosphorylation, cells were plotted versus IC50s.

Average reduction of p-AKT and

p-ERK following treatment

*

Supplemental Figure 11. Tumor volume at the end of the treatment. Tumor volumes at the

beginning and the end of the experiment shown in Figure 5 are presented as the average + SEM.

Combination treatment showed significant reduction in tumor volume (* indicates p <0.05 by

paired t-test).

Pre-treatment

End of treatment

0

500

1000

1500

2000

R1507 AZD6244 Comb

Volu

me (

mm

3)

p-Tyr

p85

IRS-1

(short exposure)

IRS-1

(long exposure)

KRAS mutation

BRAF mutation

-

-

-

-

-

-

-

-

-

-

-

-

-

-

+

-

+

-

+

-

-

+

-

+

-

+

IRS-1

ERBB3

IP

p-Tyr

p85

No. 7

IRS-1

IRS-1

A B

C

p-EGFR

p-IGFR

EGFR

IGFR

p-ERBB3

ERBB3

Supplemental Figure 12. IRS-1 associates with PI3K in human colorectal cancers

(A) Lysates from the human colorectal cancer specimens. Whole cell lysates from nine

different human colorectal cancer specimens used for the immunoprecipitations in Figure

6 were immunoblotted for the indicated antibodies. Phospho-MET was not detected in any

samples.

(B) The protein lysate from patient #7 (7th lane of Figure 6A) was immunoprecipitated with

p85 (lane 1) and IRS1 (lane 3). Supernatant from the IRS-1 IP was subsequently followed

by p85 IP (lane 2). Please note that IP of IRS-1 cleared the pTyr band associated with the

p85 IP.

(C) p85 was immunoprecipitated from additional 13 human colorectal cancer specimens

followed by western blotting with the indicated antibodies. The KRAS and BRAF mutational

status of each cancer specimen is indicated.

0

2

4

6

8

10

% a

popto

sis

SW620

Supplemental Figure 13. Concomitant inhibition of PI3K-AKT and MEK-ERK signaling does

not lead to apoptosis in SW620 cells

SW620 cells were treated with indicated drugs and combinations for 72 hours. The percent of

cells undergoing apoptosis, as measured by annexin V positivity, is shown relative to untreated

cells. The average + SD of three independent experiments is shown. GDC-0941 (GDC) is a pan-PI3

kinase inhibitor.

P-Tyr

p85

250

150

100

ERBB3

IRS1/2

A B C

p-AKT S473

ERK

p-ERK

p-AKT T308

AKT

S6

p-S6

NCI-H630

% p

-AK

T f

ollo

win

g tr

ea

tme

nt

0

20

40

60

80

100

120

p < 0.01

p < 0.01

Supplemental Figure 14. Regulation of the PI3K-AKT signaling pathway in KRAS

wildtype cancers

(A) Cell extracts from the indicated cell lines were immunoprecipitated with an antibody to

p85 and blotted with an anti-pTyr antibody.

(B) NCI-H630 cells were treated with control, gefitinib (1 μM), NVP-AEW541 (NVP) (1 μM),

a MEK inhibitor AZD6244 (AZD) (1 μM), and gefitinib and NVP-AEW541 (Gef/NVP) for 6

hours. Cell lysates were probed with the indicated antibodies.

(C) The normalized levels of phosphorylated AKT were quantified for each of the KRAS

wildtype cell lines 6 hours following the indicated treatments (blots are shown in

Supplemental Figure 3 and 15B). Each point represents the results from a single cell line.

Bars indicate mean values.

A

B

C

HT55 CaR-1 H630 CoCM-1

0

20

40

60

80

0

10

20

30

40

CoCM-1CCK-81 Difi

0

10

20

30

40

% a

popto

tic

% a

popto

tic

% a

popto

tic

p-AKT S473

ERK

p-ERK

p-AKT T308

AKT

S6

p-S6

HT55 CoCM-1H630 CaR-1

*

CCK-81 Difi

Supplemental Figure 15. Combined EGFR and IGF-IR inhibition downregulate both AKT

and ERK phosphorylation leading to apoptosis in KRAS wt cancers

(A) Cell viability data for the individual KRASwt cell lines using the indicated drugs and

combinations. Each concentration was assessed in sixplicate, and the averages and standard

deviations are shown.

(B) Cells were treated with either DMSO (-) or the indicated drug or drug combinations for 6

hours. Drugs were used at the same concentrations as in Supplemental Figure 8A. Protein lysates

were probed with the indicated antibodies. *, non-specific bands.

(C) KRAS wt cells were treated with indicated drugs and combinations for 72 hours. The percent

of cells undergoing apoptosis, as measured by annexin V positivity, is shown relative to untreated

cells. The average + SD of three independent experiments is shown.