1

B-RafV600E cooperates with alternative spliced Rac1b to sustain

colorectal cancer cell survival

Paulo Matos1, Carla Oliveira2,3, Sérgia Velho2, Vânia Gonçalves1, Luís Teixiera

da Costa2, Mary Pat Moyer4, Raquel Seruca2,3, Peter Jordan1

1Centre of Human Genetics, National Health Institute ‘Dr. Ricardo Jorge', 1649-

016 Lisbon, Portugal; 2Institute of Molecular Pathology and Immunology of the

University of Porto (IPATIMUP) 4200-465 Porto, Portugal; 3Faculty of Medicine

of the University of Porto, Portugal; 4INCELL Corporation; San Antonio Texas

This work was supported by the Portuguese Fundação para a Ciência e

Tecnologia (grants POCTI/47546/02 and POCTI/56921/2004, Programa de

Financiamento Plurianual do CIGMH)

The authors declare that no conflict of financial interest exists in this study.

Corresponding author: Peter Jordan

Centro de Genética Humana, Instituto Nacional de Saúde 'Dr. Ricardo Jorge'

Avenida Padre Cruz, 1649-016 Lisboa, Portugal

Tel: +351-217519380, Fax: +351-217526410;

e-mail: peter.jordan@insa.min-saude.pt

Short Title: Rac1b and B-RafV600E cooperate in colorectal cancer

2

ABSTRACT

Background & Aims: In colorectal tumors, activating BRAF mutations occur

alternative to KRAS oncogenic mutations, but in cell culture possess a much

lower transforming capacity. Rac1b, an hyperactive Rac1 spliced variant, is

over-expressed in some colorectal tumors and activates the transcription factor

NF-kB, which initiates a transcriptional response that promotes cell cycle

progression and inhibits apoptose. The aim of this study was to determine

whether Rac1b overexpression is associated with B-RafV600E in primary

colorectal tumors and whether a functional cooperation between these two

proteins exists in colorectal cells with a wild-type KRAS genotype. Methods:

Screening of BRAF and KRAS mutations by direct sequencing and Rac1b

mRNA expression analysis by quantitative Real Time PCR were conducted in

74 samples (13 normal colonic mucosa, 45 primary colorectal tumors and, 16

colorectal cancer cell lines). RNA interference and focus formation assays were

used to assess the cooperation between Rac1b and B-RafV600E in cancer cell

viability. Results: Rac1b overexpression and B-RafV600E are significantly

associated in primary colorectal tumors (P=0.008) and colorectal cell lines. The

simultaneous suppression of both proteins dramatically decreased colorectal

cancer cell viability through impaired cell cycle progression and increased

apoptosis. Conclusions: Our data demonstrate that Rac1b and B-RafV600E

functionally cooperate to sustain colorectal cell viability and suggest they

constitute an alternative survival pathway to oncogenic K-Ras. These results

reveal a novel molecular characteristic of colon tumors containing B-Raf

mutations and should help in defining novel targets for cancer therapy.

3

Activating mutations in the K-Ras gene are detected in about 30-40% of

colorectal tumors1,2 and promote cell proliferation by stimulating the

Raf/MEK/ERK-MAP kinase cascade. However, Ras-induced transformation is

dependent on its capacity to activate further downstream signaling pathways

that protect transformed cells from apoptosis3,4. Rac signaling has been shown

essential for Ras-induced transformation5-8, involving the activation of the NF-kB

transcription factor, which initiates an anti-apoptotic transcriptional response

and further promotes cell cycle progression by an increase in cyclin D1

expression9-11.

Activating mutations in B-Raf were recently described12 as an alternative

oncogenic event in colorectal tumors without K-Ras mutations13. The presence

of either mutation characterizes molecularly distinct subclasses of mainly

proximal colorectal tumors: B-Raf mutations occur in tumors with high levels of

microsatellite instability and an intense methylation of multiple genes, the CpG

island methylator phenotype (CIMP) 1 group14. In contrast, K-Ras mutations are

found in microsatellite-stable tumors with occasional gene methylation (CIMP 2

group)14. At the protein level, the alternative oncogenic mutation V600E

increases B-Raf activity stimulating the MEK/ERK kinase cascade independent

of Ras. However, it exhibited a much lower transforming activity than oncogenic

Ras12, indicating that tumors with mutant B-Raf require additional mechanisms

for progression that activate Rac1 or NF-kB 3,15.

Previously, we reported increased expression of alternative spliced Rac1b in

some colorectal tumor samples16. Rac1b is hyperactivated but differs from Rac1

by favoring reactive oxygen species production and stimulation of the NF-kB

pathway in detriment of other classical Rac signaling pathways17-21. These

4

properties of Rac1b prompted us to determine whether Rac1b overexpression

was associated to B-RafV600E in primary colorectal tumors and validate in vitro

whether a cooperation between these two proteins may constitute a K-Ras-

independent pathway to deregulate proliferation and survival of colorectal cells.

MATERIALS AND METHODS

Samples

We collected 45 primary colorectal tumors, 13 of which paired with their normal

colonic mucosas, and 15 paraffin-embedded tumor samples which were

classified and macrodissected using Hematoxylin and Eosin (HE)-stained

sections. Tissue sample collection was carried out in accordance with

previously established ethical protocols.

Cell lines

Sixteen colorectal, 15 melanoma and 6 thyroid tumor cell lines were analysed

(see Supplementary Tables 1 and 2). NIH 3T3 cells were used for focus

formation assays and maintained in DMEM supplemented with 10% (v/v) new

born calf serum (CS). HT29, Co115, FM3D, FM3P, FM45, FM87, Mewo,

UACC62, M14, Hth74, C643-1, NPA and 8509C cells were maintained in RPMI,

SKmel2, SKmel5 and SKmel28 cells in MEM and the remaining cell lines in

DMEM, all supplemented with 10% (v/v) fetal calf serum (FCS) (all reagents

from Gibco Invitrogen Corporation; Barcelona, Spain). The normal colonocyte

cell line NCM46022 was received by a licensing agreement with INCELL

5

Corporation, San Antonio, Tx. The cells were routinely propagated under

standard conditions in M3:10TM medium (INCELL).

KRAS and BRAF genotyping

Total DNA was isolated from 58 sporadic colorectal samples either from frozen

samples or from paraffin-embedded tissue, as described2,16. DNA was also

isolated from 37 cell lines (16 colorectal carcinoma, 15 melanoma and 6 thyroid

carcinoma). Mutational analysis of hotspot KRAS exon 1 and BRAF exon 15

was performed using PCR/Sequencing as described2.

RT-PCR and Real Time PCR

Total RNA from paraffin-embedded tumors was isolated using manufactures

instructions (Purescript- Gentra Systems) and 200 ng were reverse transcribed

using random primers and SuperScript II (Invitrogen). Total RNA from frozen

tumours, normal colonic mucosas and cell lines was extracted with the

RNAeasy kit (Qiagen, Hilden, Germany) and 1 µg reverse transcribed using

random primers (Invitrogen) and Ready-to-Go you prime beads (GE Healthcare,

Buckinghamshire, UK). Semi-quantitative, ARMS-based PCRs to amplify

mutant B-RafV600E and control RNA polymerase II (Pol2) transcripts were carried

out in parallel from a common master mix, containing all components except

primers. Following equal distribution between 2 PCR tubes, either BRafVE 5’-

AGG TGA TTT TGG TCT AGC TAC TGA and BRaf-R2 5’-GAA TAA GGT AAC

TGT CCA GTC or Pol2-F 5’-GAG CGG GAA TTT GAG CGG ATG C and Pol2-

R 5’-GAA GGC GTG GGT TGA TGT GGA AGA 10 µM primers were added to

6

each tube and samples subjected to standard PCR programs with annealing

temperatures of 60º C in 25 amplification cycles for Pol2, or 54º C with 30

cycles for BRafVE. Rac1b amplification was as described17.

Real Time relative quantification was performed on an ABI Prism 7000

Sequence Detection System. Two primers were designed using the ABI Primer

Express software that amplified amplicons specific for Rac1b (78 bp) or for total

Rac1 (75 bp) transcripts (i.e., Rac1+Rac1b). Primer sequences were 5’-GGG

CAA AGA CAA GCC GAT TG and 5’-CGG ACA TTT TCA AAT GAT GCA GG

(Rac1b detector) or 5’-CCT GCA TCA TTT GAA AAT GTC CG and 5’- GAT

GAT GGG AGT GTT GGG ACA GT (total Rac1 detector). Each cDNA sample

was diluted 5 times and 5 µl transferred to each Real Time reaction together

with 300 nM primers and SYBR Green Master Mix (Applied Biosystems). For

standardization, all samples were analyzed against HT29 cDNA (reference)

using the 7000 SDS 1.1 RQ Software (∆∆CT method - Applied Biosystems).

Tumor and cell line relative Rac1b expression values were then normalized to

the mean expression level of control tissue samples. Samples were analyzed,

as duplicates, for each detector and repeated in at least two independent

experiments. Low yield RNAs obtained from paraffin tumor samples were

primarily validated using the total Rac1 detector and a series of 5 log10 dilutions

of HT29 cDNA (standard curve method). Only those tumor samples that

exhibited an amplification CT earlier than that of the 10-4 dilution of HT29 were

included. This was found in titration experiments as the minimal concentration

for accurate determination of the amount of Rac1b.

7

Plasmids constructs and siRNA oligos

Rac1 and Rac1b cDNAs and their Q61L mutants were cloned into pcDNA3-Myc

or pcDNA3-GFP vectors as described17. The coding sequences of the normal

and V600E mutant forms of human B-Raf were amplified by PCR, fused to a

Kozak consensus translation initiation sequence and cloned into a pCMV

promoter-containing expression vector. B-Raf, B-RafV600E and K-RasV12 were

further subcloned into pEGFP vector (Clontech, Mountain View, CA, USA).

SiRNA oligos were ordered from MWG-Biotech AG with the following

sequences: siRac1b’A 5’-CAG UUG GAG AAA CGU ACG GTT; siRac1b’B 5’-

CGU ACG GUA AGG AUA UAA CTT; siBRafVE 5’-GCU ACA GAG AAA UCU

CGA UTT (described as MuA in ref. 22). SiCtrl was a mixture of standard

control siRNAs against GFP 5’-GGC UAC GUC CAG GAG CGC ACC TT and

Luciferase GL2 5’-CGU ACG CGG AAU ACU UCG ATT.

siRNA transfection

For RNA interference experiments, cells at 30-40% confluence were transfected

using LipofectAMINE 2000 (Invitrogen) and the indicated siRNAs, according to

the manufacturer’s instructions, and analyzed 24 and 48h later. Total amounts

of siRNA used were: 200 pmol for SW480, Caco2 and Co115 cells and 400

pmol for HT29. When required, the total amount of siRNA in the individual

knockdowns was adjusted with control siRNA (siCtrl). In ectopic protein

expression knockdown, SW480 cells were first transfected with the indicated

siRNA for 24h, followed by a second transfection with 500 ng of the indicated

constructs, and analyzed following 16-20h. Depletion of endogenous Rac1b

8

protein reached approximately 70% in all targeted cell lines and that of the

mutant B-Raf allele 80%. The simultaneous transfection of Rac1b and B-

RafV600E siRNAs in the same cell lines reached efficiencies equivalent to those

of the individual knockdowns.

SDS-PAGE, Western blotting and CRIB-peptide pull down assays

Samples were prepared and detected as described21. The antibodies used in

this study were rabbit anti-c-Myc A14 and mouse anti-BRaf F7 from Santa Cruz

Biotechnology (Santa Cruz, CA, USA), rabbit anti-GFP ab290 from Abcam,

(Cambridge, UK), mouse anti-�-tubulin clone Tub2.1 from Sigma-Aldrich

(Madrid, Spain) (as a loading control), mouse anti-cyclin D1 clone DCS6 from

BD-Pharmingen, rabbit anti-cleaved PARP from New England Biolabs, rabbit

anti-Rac1b polyclonal serum as described17. For densitometric analysis film

exposures from at least three independent experiments were digitalized and

analyzed using ImageJ software (NIH).

Viability, cell death and G1/S-progression assays

Cells were seeded in 35 mm dishes containing six 10 mm cover slips,

transfected as described above and analyzed at 0 h, 24 h and 48 h. At each

time point, two cover slips were retrieved and cells either stained with trypan

blue (Sigma) or fixed and processed similarly to what was described20. Briefly,

cell death was detected by fluorescence TUNEL assay (In-situ cell death kit,

Roche Applied Science, Indianapolis, IN, USA) and progression into S-phase

was monitored by incorporation of 5-bromo-2-deoxyuridine (BrdU), followed by

detection with mouse anti-BrdU (Roche) and goat-anti mouse-TexasRed

9

(Jackson Immunoresearch Laboratories). For each time point, phase contrast

and fluorescence images were digitally recorded in 10 randomly chosen fields

(200x magnification) and both positive and negative cells were scored. Total

number of cells counted is indicated. All results were confirmed by at least three

independent experiments.

Focus formation assay

Low passage NIH 3T3 cells seeded on 35 mm dishes at 60-80% confluence

were transfected using LipofectAMINE Plus (Invitrogen) with a total of 1 µg of

plasmid DNA, according to the manufacturer’s instructions. Equivalent amounts

of ectopic protein expression were achieved using 100-500 ng of the indicated

GFP-tagged constructs. When required, the amount of DNA was adjusted with

empty vector. Twenty four hours later cells were trypsinised, split into two 100

mm dishes and maintained in Dulbecco’s modified Eagle’s medium plus 5%

(v/v) new born calf serum (CS) (Invitrogen). The medium was changed every 3

days thereafter. Fourteen-21 days after cells were fixed with methanol/acetic

acid [1/3 (v/v)] and GFP fluorescence of the foci confirmed under an inverted

fluorescence microscope. Cells were then stained with 0.4% crystal violet in

methanol/acetic acid in order to count the foci and photograph the dishes.

Statistical analysis

The statistical analysis was performed using unpaired two-tailed Student’s T

test or Fisher’s statistical test when appropriated. Differences were taken to be

significant at P<0.05.

10

RESULTS

Initially, we analyzed 16 colorectal cancer cell lines and found that all lines

carrying B-RafV600E had elevated levels of Rac1b expression, whereas none of

the cell lines with mutant K-Ras and only 1 cell line wild type for both genes

showed Rac1b over-expression (Supplementary Table 1). Moreover, and in

agreement with Rac1b being a hyperactivated variant, we found that the roughly

2-fold increase in Rac1b expression observed in wild type or B-RafV600E cells

was sufficient to yield a high level of active variant, comparable to the level of

GTP-Rac1 in a cell line with mutant K-Ras (Figure 1A).

We then extended our quantitative analysis of Rac1b expression to 13 normal

mucosa and 45 primary tumors samples genotyped for KRAS and BRAF

mutations. Sixteen of the tumors (35.6%) showed clearly elevated levels of

Rac1b expression, ranging from 1.9 to 4.1 times the mean expression level

found in the normal mucosa samples (Figure 1B and data not shown). We then

performed association studies between Rac1b expression and KRAS or BRAF

genotypes and a significant association (P=0.008 by F test) between the

presence of B-RafV600E and Rac1b over-expression emerged (Figure 1C). In

fact, up to 82% of the tumors with mutant B-Raf had elevated levels of Rac1b

expression. In contrast, wild type and mutant K-Ras genotypes showed no

association with Rac1b expression (P�0.50 by F tests – Figure 1C and data not

shown).

A high incidence of B-Raf mutations has also been detected in melanoma and

thyroid tumors23-25. A series of melanoma and thyroid cancer cell lines was

therefore characterized for presence of B-Raf mutations and Rac1b over-

11

expression but revealed no association (Supplementary Table 2). In fact, the

levels of Rac1b found in melanoma and thyroid cancer cells were similar to the

levels found in normal mucosa. These data strongly suggest that the

association between Rac1b over-expression and B-RafV600E reflects a colorectal

cancer cell-specific molecular context.

To investigate the impact of the observed association on colorectal cancer cell

viability, we selectively depleted Rac1b and mutant B-RafV600E expression in

representative cell lines using specific RNA interference methodology (Figures

2A-B and Supplementary Figure 1A). Remarkably, the simultaneous repression

of both proteins revealed a dramatic reduction of cell viability, reaching less

than 20% viable cells after 48h of siRNA treatment. Moreover, the double

knockout effect was clearly more pronounced than that achieved with the

individual repression of either Rac1b or B-Raf (Fig. 2C-D and Supplementary

Figs. 1B and 2A). This strongly indicates a functional cooperation between the

two proteins, which is consistent with previously described data that B-Raf

mutation alone is not sufficient for cancer and requires activation of additional

pathways to fully induce its oncogenic potential3,12,15.

As additional controls, neither the B-RafV600E (not shown) nor the Rac1b siRNA

oligos affected the viability of KRAS mutant SW480 cells (Figure 2E-F), which

do not express Rac1b nor harbor a B-RafV600E, demonstrating that no unspecific

cytotoxic effect was observed. Moreover, the viability of Caco2 cells, which

express elevated levels of Rac1b with a wild type BRAF genotype, was not

affected in the presence of B-RafV600E siRNAs (not shown) but was also

reduced to about 50% after 48h post-transfection of Rac1b-specific oligos

12

(Figure 2E-F), a value consistent with the effect observed in HT29 and Co115

cells.

As previously described, Ras promotes transformation by stimulating cell

proliferation through both the Raf/MEK/ERK cascade and Rac signaling,

whereas protection from apoptosis occurs mainly through Rac signaling. The

contribution of Rac1b and mutant B-Raf to cell cycle progression and apoptosis

was therefore determined. Suppression of B-RafV600E and Rac1b revealed that

both proteins provide an independent but additive contribution to cell cycle

progression in HT29 colorectal cells, with their simultaneous suppression

producing a 3.3-fold reduction in BrdU-positive cells that correlated with

decreasing cyclin D1 levels (Figure 3A and Supplementary Figure 3A).

However, when apoptosis was assessed under the same double knockdown

conditions a striking 7.8-fold, synergistic increase in TUNEL-positive cells was

observed (Fig. 3C-D). Moreover, Rac1b shows a predominant role in this effect

since its isolated depletion resulted in two times more apoptotic cells than that

achieved by single suppression of B-RafV600E (Fig. 3C-D). TUNEL results were

confirmed by Western Blot showing an increase in the 89 kDa fragment of

PARP-1, a well defined, caspase 3/7-cleaved apoptosis marker (Figure 3C).

Comparable results were obtained for Co115 cells that also carry a mutant

BRAF allele (Supplementary Figures 1C-D and 2B-C).

These results were again validated in Caco2 cells (wild-type KRAS and BRAF)

and SW480 cells (mutant KRAS). Consistent with the results on cell viability,

depletion of Rac1b in Caco 2 cells decreased S-phase cells and increased cell

death, whereas transfection of either Rac1b or B-RafV600E oligos had no effect

13

on G1/S progression and apoptosis induction in SW480 cells (Figure 3B,E,F

and Supplementary Figure 3B).

Since over-expressed Rac1b associated with activating BRAF mutations but not

with KRAS mutations, we hypothesised that the observed synergy in sustaining

cell viability could be functionally substituting the classical oncogenic K-Ras

signaling. We made use of a classical fibroblast focus formation assay to test

this. As shown in Figure 4, co-expression of wild type Rac1b, but not of wild

type Rac1, produces a synergistic increase in the focus forming ability of mutant

B-RafV600E, similar to the co-expression of constitutively active Rac1-L61 and

raising efficiency to levels that approximate those of oncogenic K-Ras.

DISCUSSION

Colorectal cancer is a leading cause of mortality in the Western world. A

hallmark characteristic in more than 30% of colorectal tumors is the presence of

oncogenic mutations in the KRAS gene that promote both proliferation and

survival of cells. Alternative BRAF mutations were reported in a significant

number of remaining tumors12,14, however, by itself, oncogenic BRAF is not

sufficient for cancer and must cooperate with other signaling processes15.

Our data reveal that mutant B-Raf cooperates with overexpression of

hyperactive Rac1b in the survival of colorectal tumor cells. Rac1b is a rarely but

naturally expressed splice variant that exists predominantly in the active

conformation17,18,26,27 and preferentially stimulates reactive oxygen species

production and NF-kB activity17,19,21,28. Overexpressed Rac1b provided a

predominant pro-survival signal but also a substantial contribution to cell cycle

stimulation. Oncogenic K-Ras is able to prevent apoptosis by activating

14

canonical Rac1 signaling5-8 but B-RafV600E lies downstream of K-Ras and by

itself cannot activate Rac1, thus requiring cooperation with hyperactive Rac1b.

Consistent with this model, we demonstrate that combined targeting of both

molecules is highly efficient in killing these colorectal tumor cells (80% of cell

death) and this may be therapeutically relevant to improve the modest efficacy

that existing Raf/MEK/ERK pathway inhibitors revealed in clinical trials29. Our

results thus characterize a molecular pathway sustaining cancer cell

proliferation in a subtype of colorectal tumors that complements the previously

identified CIMP1 sub-classification14. An interesting question is whether Rac1b

overexpression is already involved in the initial transformation of colon cells. We

have tried to test this in the only available primary colorectal cell line model,

NCM46022. Unfortunately, these cells revealed no contact-inhibition, thus

precluding focus formation assays and in our hands no viable cells could be

selected with an ectopic B-RafV600E expression level that would reproduce the

heterozygous B-Raf expression in tumors (data not shown). Finally, our findings

may have yet broader implications in cancer research and treatment because

Rac1b over-expression has also been detected in breast cancer19,26.

SUPPLEMENTARY MATERIAL

Supplementary Material is available at Gastroenterology Online. The single pdf-

file contains Supplementary Tables 1 and 2 listing the colorectal and melanoma

or thyroid cell lines, respectively, which were analyzed for Rac1b expression

and the KRAS or BRAF genotypes in this study. Supplementary Figure 1 shows

that depletion of Rac1b and B-RafV600E also affected cell viability, BrdU

15

incorporation and apoptosis in a mismatch-repair deficient (MSI) colorectal

tumor cell line, namely Co115. Supplementary Figure 2 shows representative

microscopic images of Co115 cells following depletion of Rac1b and/or B-

RafV600E and analysis of cell viability, BrdU incorporation and TUNEL staining.

Supplementary Figure 3 shows representative microscopic images of BrdU

incorporation following depletion of Rac1b and/or B-RafV600E in HT29, SW480 or

Caco2 cells.

ACKNOWLEDGEMENTS

We thank Eric Chastre (INSERM U 773, Paris) for generously providing total

RNA extracted from colon tumors and mucosa samples and JG Collard (NKI,

Holland) for helpful suggestions. We also acknowledge FCT for fellowship BPD

11180/02 to P.M.

REFERENCES

1. Andreyev HJ, Norman AR, Cunningham D, Oates JR, Clarke PA. Kirsten

ras mutations in patients with colorectal cancer: the multicenter “RASCAL"

study. J. Natl. Cancer Inst. 1998; 90:675-684.

2. Oliveira C, Velho S, Moutinho C, Ferreira A, Preto A, Domingo E,

Capelinha AF, Duval A, Hamelin R, Machado JC, et al. KRAS and BRAF

oncogenic mutations in MSS colorectal carcinoma progression. Oncogene

2007; 26:158-163.

3. Downward J. Targeting ras signaling pathways in cancer therapy. Nat.

Rev. Cancer 2003; 3:11-22.

16

4. Cox AD, Der CJ. The dark side of Ras: regulation of apoptosis. Oncogene

2007; 22:8999-9006.

5. Khosravi-Far R, Solski PA, Clark GJ, Kinch MS, Der CJ. Activation of

Rac1, RhoA, and mitogen-activated protein kinases is required for Ras

transformation. Mol. Cell. Biol. 1995; 15:6443-6453.

6. Qiu RG, Chen J, Kirn D, McCormick F, Symons M. An essential role for

Rac in Ras transformation. Nature 1995; 374:457-459.

7. Joneson T, Bar-Sagi D. Suppression of Ras-induced apoptosis by the Rac

GTPase. Mol. Cell. Biol. 1999; 19:5892-5901.

8. Malliri A, van der Kammen RA, Clark K, van der Valk M, Michiels F,

Collard JG. Mice deficient for the Rac activator Tiam1 are resistant to Ras-

induced skin tumors. Nature 2002; 417:867-871.

9. Joyce D, Bouzahzah B, Fu M, Albanese C, D'Amico M, Steer J, Klein JU,

Lee RJ, Segall JE, Westwick JK, et al. Integration of Rac-dependent

regulation of cyclin D1 transcription through a nuclear factor-kappaB-

dependent pathway. J. Biol. Chem. 1999; 274:25245-25249.

10. Hinz M, Krappmann D, Eichten A, Heder A, Scheidereit C, Trauss M. NF-

kappaB function in growth control: regulation of cyclin D1 expression and

G0/G1-to-S-phase transition. Mol. Cell. Biol. 1999; 19:2690-2698.

11. Guttridge DC, Albanese C, Reuther JY, Pestell RG, Baldwin AS Jr. NF-

kappaB controls cell growth and differentiation through transcriptional

regulation of cyclin D1. Mol. Cell. Biol. 1999; 19:5785-5799.

17

12. Davies H, Bignell GR, Cox C, Stephens P, Edkins S, Clegg S, Teague J,

Woffendin H, Garnett MJ, Bottomley W, et al. Mutations in the BRAF gene

in human cancer. Nature 2002; 417:949-954.

13. Oliveira C, Pinto M, Duval A, Brennetot C, Domingo E, Espin E, Armengol

M, Yamamoto H, Hamelin R, Seruca R, Schwartz S Jr. BRAF mutations

characterize colon but not gastric cancer with mismatch repair deficiency.

Oncogene 2003; 22:9192-9196.

14. Shen L, Toyota M, Kondo Y, Lin E, Zhang L, Guo Y, Hernandez NS,

Chen X, Ahmed S, Konishi K, et al. Integrated genetic and epigenetic

analysis identifies three different subclasses of colon cancer. Proc. Natl.

Acad. Sci. USA 2007; 104:18654–18659.

15. Dhomen N, Marais R. New insight into BRAF mutations in cancer. Curr.

Opin. Genet. Dev. 2007; 17:31-39.

16. Jordan P, Brazão R, Boavida MG, Gespach C, Chastre E. Cloning of a

novel human Rac1b splice variant with increased expression in colorectal

tumors. Oncogene 1999; 18:6835-6839.

17. Matos P, Collard JG, Jordan P. Tumor-related alternative-spliced Rac1b is

not regulated by Rho-GDI and exhibits selective downstream signaling. J.

Biol. Chem. 2003; 278:50442-50448.

18. Fiegen D, Haeusler LC, Blumenstein L, Herbrand U, Dvorsky R, Vetter IR,

Ahmadian MR. Alternative splicing of Rac1 generates Rac1b, a self-

activating GTPase. J. Biol. Chem. 2004; 279:4743-4749.

18

19. Radisky DC, Levy DD, Littlepage LE, Liu H, Nelson CM, Fata JE, Leake D,

Godden EL, Albertson DG, Nieto MA, et al. Rac1b and reactive oxygen

species mediate MMP-3-induced EMT and genomic instability. Nature

2005; 436:123-127.

20. Matos P, Jordan P. Expression of Rac1b stimulates NF-kB-mediated cell

survival and G1/S-progression. Exp. Cell Res. 2005; 305:292-299.

21. Matos P, Jordan P. Rac1, but not Rac1b, stimulates RelB-mediated gene

transcription in colorectal cancer cells. J. Biol. Chem. 2006; 281:13724-

13732.

22. Moyer MP, Manzano LA, Merriman RL, Stauffer JS, Tanzer LR. NCM460,

a normal human colon mucosal epithelial cell line. In Vitro Cell Dev. Biol.

Anim. 1996; 32:315-317.

23. Hingorani SR, Jacobetz MA, Robertson GP, Herlyn M, Tuveson DA.

Suppression of BRAF(V599E) in human melanoma abrogates

transformation. Cancer Res. 2003; 63:5198-5202.

24. Panka DJ, Atkins MB, Mier JW. Targeting the mitogen-activated protein

kinase pathway in the treatment of malignant melanoma. Clin. Cancer

Res. 2006; 12:2371s-2375s.

25. Wojciechowska K, Lewinski A. BRAF mutations in papillary thyroid

carcinoma. Endocr. Regul. 2006; 40:129-138.

26. Schnelzer A, Prechtel D, Knaus U, Dehne K, Gerhard M, Graeff H,

Harbeck N, Schmitt M, Lengyel E. Rac1 in human breast cancer:

19

overexpression, mutation analysis, and characterisation of a new isoform,

Rac1b. Oncogene 2000; 19:3013-3020.

27. Singh A, Karnoub AE, Palmby TR, Lengyel E, Sondek J, Der JC. Rac1b, a

tumor associated, constitutively active Rac1 splice variant, promotes

cellular transformation. Oncogene2004; 23:9369-9380.

28. Esufali S, Charames GS, Pethe VV, Buongiorno P, Bapat B. Activation of

tumor-specific splice variant Rac1b by Dishevelled promotes canonical

Wnt signaling and decreased adhesion of colorectal cancer cells. Cancer

Res. 2007; 67:2469-2479.

29. Fang JY, Richardson BC. The MAPK signaling pathways and colorectal

cancer. Lancet Oncol. 2005; 6:322-327.

20

LEGENDS TO FIGURES:

Figure 1. Overexpression of Rac1b in colorectal tumors and cell lines. (A)

Expression of endogenous Rac1b in normal colon mucosa and three colorectal

cell lines was detected by RT-PCR (top). Western blot analysis using an anti-

Rac1 antibody (middle) and a CRIB-domain pull-down assay (bottom)

demonstrate the amount of total versus active GTP-bound Rac1 and Rac1b in

the cell lines. Their KRAS and BRAF genotypes are indicated. (B) RT-PCR

analysis of Rac1b expression in representative colorectal tumors from freshly

frozen (T1-T5) or paraffin-embedded tumors (T6-T7) and correlation with the

indicated KRAS and BRAF genotypes. (C) Correlation between Rac1b over-

expression and the KRAS and BRAF genotypes in 45 primary colorectal

tumors, 13 of which existed as paired tumor/mucosa samples. Expression

levels of Rac1b were quantified by Real Time PCR as described in

Experimental Procedures. Below the graph, the percentage values are given, as

well as the number of positive cases versus the total tumor number in

parentheses. KRAS mutations were in codons 12 or 13, and of the eleven

mutated BRAF alleles nine were V600E and two were K601E, both

oncogenic12,29.

Figure 2. Specific interference with Rac1b and B-RafV600E expression

decreases survival of colorectal tumor cells. (A) Specificity of the small

interfering RNA oligonucleotides (siRNAs) for the depletion of Rac1b and

mutant B-RafV600E. (Top) SW480 colorectal cells expressing either Myc-Rac1-wt

21

or Myc-Rac1b-wt were transfected with either control (siCtrl) or one of two

Rac1b-specific oligos (siRac1b'A or 'B), as indicated. (Bottom) SW480

colorectal cells expressing either GFP-B-Raf-wt or GFP-B-RafV600E were

transfected with either control (siCtrl) or a mutant-specific siBRafVE oligo22, as

indicated. Cells were lysed after 24 h and analyzed by Western blot with either

anti-tubulin (loading control) or anti-tag (target protein levels) antibodies. Note

that the two Rac1b siRNAs efficiently repressed Myc-Rac1b expression without

affecting the Myc-Rac1 protein and that the siBRafVE oligo strongly decreased

the expression of the mutant protein without depleting the wild type B-Raf

protein. (B) Specific depletion of endogenous Rac1b and B-RafV600E in HT29

colorectal cells. Cells were transfected with the indicated siRNAs and analyzed

after 24 h by Western blot with the indicated antibodies for the expression levels

of either B-Raf or Rac1b. Endogenous Rac1b protein was depleted by

approximately 70% whereas B-RafV600E depletion led to a decrease of

approximately 50% in the amount of endogenous B-Raf protein because HT29

cells are heterozygous for the B-RafV600E mutation. Transcript depletion from the

mutant B-Raf allele reached 80% in HT29 cells as documented by ARMS RT-

PCR (bottom panels) using the amplification of RNA polymerase II as internal

reference. Note that the simultaneous knockdown of Rac1b and B-RafV600E

reached depletion efficiencies equivalent to those of the individual knockdowns.

Also, the depletion of B-RafV600E shows no effect on the expression level of

Rac1b protein (or of its transcripts, not shown). As further controls, the

colorectal cells SW480 (do not express Rac1b) and Caco2 (express Rac1b but

no B-RafV600E) were transfected with the indicated siRNAs and analyzed after

24 h by Western blot with either anti-tubulin (loading control) or anti-Rac1b

22

antibodies. (C-F) � Depletion of Rac1b and B-RafV600E affects colorectal tumor

cell viability. (C) HT29 cells were transfected with the indicated siRNAs and

viable cells counted microscopically after 24 and 48 h. (D) Representative

phase contrast images taken 48 h after treatment with the indicated siRNAs .

(E) Depletion of Rac1b as in (C) and images as in (D) but in Caco-2 and SW480

colorectal tumor cells. Shown are mean values of three independent

experiments, with at least 1000 cells counted at each time point for each

sample. Bars indicate standard deviation. Note that SW480 cells, which do not

express Rac1b, are not affected in their viability.

Figure 3. Depletion of Rac1b and B-RafV600E impair cell cycle progression and

promote apoptosis in colorectal tumor cells. (A) HT29 cells or (B) SW480 and

Caco2 cells were transfected with the indicated siRNAs, pulse-labeled with

bromodeoxyuridine and S-phase cells counted microscopically. In a parallel

Western blot analysis the cyclin D1 and �-tubulin (loading control) levels were

determined. BrdU-positive HT29 cells decreased from 38% to 23% (1.6-fold)

upon depletion of Rac1b, to 21% (1.8-fold) following depletion of B-RafV600E,

and to 12% upon simultaneous suppression of both proteins (3.3-fold).

Depletion of Rac1b in Caco2 cells produced a 1.6-fold (42% to 28%) decrease

in BrdU incorporation but had no effect on SW480 cells. (C) � HT29 cells or (D)

SW480 and Caco2 cells were transfected with the indicated siRNAs and the

number of apoptotic cells determined after 48 h with the TUNEL assay. In a

parallel Western blot analysis the caspase-cleaved PARP and ��-tubulin

(loading control) levels were determined. Note the synergistic 7.8-fold increase

23

in apoptosis upon combined depletion of Rac1b and B-RafV600E. Depletion of

Rac1b also increased cell death in Caco2 cells. (E-F) Representative

fluorescence images of TUNEL-stained (E) HT29 or (F) SW480 and Caco2 cells

taken 48 h after treatment with the indicated siRNAs. All graphs show mean

values of three independent experiments, with at least 1000 cells counted at

each time point for each sample. Bars indicate standard deviation.

Figure 4. Synergistic effect of B-RafV600E and Rac1b on fibroblast survival. (A)

The cell survival promoting activities of B-RafV600E and Rac1b were tested in a

focus formation assay following transfection of NIH 3T3 cells with the indicated

GFP-tagged constructs. Following cell fixation, foci were microscopically

checked for the presence of a green fluorescent signal, then stained and

macroscopically counted. Shown are mean values of four independent

experiments together with photographs of representative dishes. Bars indicate

standard deviation. Note the synergy when B-RafV600E and wild type Rac1b are

simultaneously expressed. (B) Western blot showing the expression levels of

the indicated GFP-tagged proteins after transfection into NIH 3T3 cells.

24

Figure 1

25

Figure 2

26

Figure 3

27

Figure 4