1

Deoxycholic acid differentially regulates focal adhesion kinase phosphorylation: Role of tyrosine phosphatase ShP2

Sharad Khare†, Cory Holgren, and Allen M. Samarel*. Department of Gastroenterology and *Cardiovascular institute, Stritch School of Medicine, Loyola University Chicago.

Running Title: Regulation of FAK by Deoxycholic Acid

Key Words: colon cancer, deoxycholic acid, focal adhesion kinase, tyrosine phosphatase

†To whom correspondence should be addressed:

Department of Medicine, Division of Gastroenterology, Hepatology and NutritionLoyola University Medical Center 2160 S. First Ave Maywood, IL 60153.

Telephone (708) 216-4783.FAX (708) 216-4113E-mail skhare@lumc.edu

Acknowledgments:These studies were funded in part by the NCI grant CA097540 to S Khare and NIH RO1 HL34328 to AM Samarel. The Department of Medicine, Section of Gastroenterology and Pathology Research Core Facilities of the Loyola University provided additional funding.

FOOTNOTES1The abbreviations used are: DCA, deoxycholic acid; FAK, focal adhesion kinase; ECM, extracellular matrix; MTT, tetrazolium 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide; ODN, oligodeoxynucleotide; PTPase, protein tyrosine phosphatase; PBS, Phosphate buffered saline; AOM, azoxymethane; Tyr, Tyrosine

Page 1 of 45Articles in PresS. Am J Physiol Gastrointest Liver Physiol (August 17, 2006). doi:10.1152/ajpgi.00008.2006

Copyright © 2006 by the American Physiological Society.

2

Abstract

Environmental factors, including dietary fats, are implicated in colonic carcinogenesis.

Dietary fats modulate secondary bile acids including deoxycholic acid (DCA) concentrations in

the colon, which are thought to contribute to the nutritional-related component of colon cancer

risk. Here we demonstrate, for the first time, that DCA differentially regulated the site-specific

phosphorylation of focal adhesion kinase (FAK). DCA decreased adhesion of HCA-7 cells to

the substratum and induced dephosphorylation of FAK at Tyrosine-576/577 (Tyr-576/577) and

Tyr-925. Tyrosine phosphorylation of FAK at Tyr-397 remained unaffected by DCA

stimulation. Interestingly, we found that c-Src was constitutively associated with FAK and

DCA actually activated Src, despite no change in FAK-397 and an inhibition of FAK-576

phosphorylation. DCA concomitantly and significantly increased association of tyrosine

phosphatase ShP2 with FAK. Incubation of immunoprecipitated FAK, in vitro, with GST-ShP2

fusion protein resulted in tyrosine dephosphorylation of FAK in a concentration dependent

manner. Antisense oligodeoxynucleotides directed against ShP2 decreased ShP2 protein levels

and attenuated DCA induced FAK dephosphorylation. Inhibition of FAK by adenoviral-

mediated over-expression of FRNK, and gene silencing of Shp2 both abolished DCA’s effect

on cell adhesion, thus providing a possible mechanism for inside-out signaling by DCA in

colon cancer cells. Our results suggest that DCA differentially regulates focal adhesion

complexes and that tyrosine phosphatase ShP2 has a role in DCA signaling.

Page 2 of 45

3

Introduction

Colorectal cancer is a leading cause of cancer-related deaths in the United States. In addition

to genetic factors, dietary constituents play important roles in tumor development (1, 52). In this

regard, diets high in animal fat predispose to experimental colon cancer (33). Endogenous bile

acids, secreted in response to dietary fats, are thought to contribute to this increased risk (1, 56,

69). The primary bile acids, cholic acid and chenodeoxycholic acid, are synthesized and

conjugated to glycine and taurine in the liver and secreted into the bile to facilitate dietary fat

absorption (29). Most of the bile acids are re-absorbed from the distal small intestine via the

enterohepatic circulation. A small fraction (less than 5%) escapes into the colon where they are

deconjugated and further metabolized by colonic bacteria. In the colon, cholic acid is converted

primarily to deoxycholic acid (DCA) 1 and chenodeoxycholic acid is metabolized to lithocholic

acid. Epidemiological studies have suggested that these secondary bile acids promote colon

cancer (9, 47). The notion that elevated fecal bile acids are linked to increased risk for colorectal

cancer is also supported by the observation that patients with colorectal adenomas have elevated

concentrations of bile acids in their blood and stool (3, 48). We have demonstrated previously

that feeding rats a diet containing cholic acid promoted the development of azoxymethane

(AOM)-induced colonic tumors (17).

A number of potential mechanisms have been proposed for bile acid-induced colonic tumor

promotion, including: (i) increased peroxidative damage to nucleic acids (15); (ii) increased

peroxidative damage to lipids and proteins (14); (iii) increased reactive nitrogen and oxygen

radicals that alter REDOX-sensitive cell signaling (34). Several specific nuclear targets for bile

acids recently have been identified, such as the pregnane X receptor, the farnesoid X receptor

and the nuclear vitamin D receptor permitting bile acid to control distinct patterns of gene

Page 3 of 45

4

expression (22, 38, 39). However, no cell surface receptor for bile acids has yet been identified

and it is presumed that these amphipathic molecules exert their effects by perturbing plasma

membrane structure and function. In this regard, DCA also causes ligand-independent

transactivation of epidermal growth factor (EGF) receptor in hepatocytes, leading to signaling

through mitogen-activated protein kinase pathways (55, 70).

Focal Adhesion Kinase (FAK), a cytoplasmic kinase localized in focal adhesions, at which

cells make close contacts with the extracellular matrix (ECM) through integrins, is one of the

most prominently tyrosine-phosphorylated proteins in many cell types in response to cell

adhesion (51). Engagement of integrins and other adhesion receptors can induce activation of

FAK (5), which leads to phosphorylation of several tyrosine residues through

autophosphorylation and recruitment of Src (63). Each of the FAK tyrosine residues is

implicated in generating a distinct signal, FAK Tyr-397 in recruiting Src, PI-3 kinase and

p130Cas to focal adhesions; FAK Tyr-576 and -577 in up-regulating FAK kinase activity and

FAK Tyr-925 in activating the Ras-MAPK pathway (61). FAK has been implicated in cell

motility (63), apoptosis and anchorage-independent growth (20). Because these are key

processes by which a transformed cell becomes invasive and metastatic, FAK may be intimately

involved in colonic carcinogenesis. Studies have demonstrated over-expression of FAK as colon

tissues become transformed and develop into primary, invasive carcinomas (7). Despite the well-

characterized roles of FAK in focal adhesion formation, the function of this signaling component

in the actions of DCA has never been studied. Here we report, for the first time, that DCA

regulates FAK by selectively dephosphorylating distinct tyrosine residues through the

recruitment of tyrosine phosphatase ShP2 and there is compelling functional evidence that

secondary bile acids may regulate membrane signaling.

Page 4 of 45

5

Materials and Methods

Materials: DCA, fibronectin, laminin, poly-l-lysine, collagen-I, BSA, crystal violet and anti-β-

actin antibodies were obtained from Sigma-Aldrich (St. Louis, MO). Protein A/G plus agarose,

polyclonal anti-FAK, anti-Src, monoclonal anti-ShP2, and anti- phosphotyrosine antibodies (PY-

20) were purchased from Santa Cruz Biotechnology (Santa Cruz, CA). Cell Signaling provided

phosphospecific FAK-925, -576/577 and Src-418 polyclonal antibodies. Phosphospecific FAK-

397 was obtained from Biosource (Camarillo, CA). Three 20-mer oligodeoxynucleotides were

synthesized by Bio Synthesis (Lewisville, TX). Fusion protein of the phosphatase domain,

comprising amino acids 224-529 of human ShP2, tagged at the N-terminus with GST was

obtained from Abcam (Cambridge, MA). For Western blotting studies, electrophoretic grade

acrylamide, bisacrylamide, Tris, SDS and prestained molecular weight markers were obtained

from Bio-Rad Laboratories. Kodak supplied X-OMAT blue film for xerography. Polyvinylidene

difluoride membranes (Immobilon-P) were purchased from Millipore, Inc. ECL Western blotting

detection reagents, as well as horseradish peroxidase coupled anti-mouse and anti-rabbit

secondary antibodies, were obtained from Amersham Biosciences. Protease inhibitor cocktail

tablets were purchased from Roche Diagnostics (Indianapolis, IN). Other reagents were of the

highest quality available and obtained from Sigma, unless otherwise noted.

Cell culture: Dr. Susan Kirkland (ICRF, London UK) generously provided the HCA-7 cells, that

were derived from a rectal carcinoma (42). Cells were grown in McCoy’s media containing 10%

fetal calf serum and penicillin/streptomycin mixture. Pre-confluent cells (70-80% confluent)

were used for experiments. Cells were treated with DCA or vehicle for the indicated times and at

the indicated concentrations.

Page 5 of 45

6

Cell viability and toxicity assays: Cell viability of DCA treated (100µM, 3hrs) HCA-7 cells was

assessed by the colorimetric conversion of tetrazolium 3-(4,5-dimethylthiazol-2-yl)-2,5-

diphenyltetrazolium bromide (MTT) to formazan as previously described (46). To assess cell

death as a consequence of exposure to bile acid, cultured cells in 24-well plates were incubated

with DCA (100µM, 3hrs). Cells were then incubated for 1 h in culture medium containing

5µg/ml propidium iodide, a cell membrane impermeant nucleic acid stain. Images of cells were

obtained on a microscope equipped for phase contrast and fluorescence imaging. Total number

of cells and the number of propidium iodide stained cells were counted. The average proportion

of dead cells per well, calculated from three images, was considered to be a single data point.

Cell adhesion assays: 96 well plates were coated with collagen-I, fibronectin, laminin and poly-

l-lysine in pre-coating buffer (15mM Na2CO3, 35 mM NaHCO3, pH 9.4) overnight at 4oC.

Nonspecific binding sites were blocked with 1% BSA in PBS for 1 hr at room temperature.

Serum starved HCA-7 cells were treated with DCA (100µM, 1hr), rinsed with serum free

medium and harvested with 2mM EDTA in PBS. In some experiments cells were pre-incubated

with serum free media containing 1mM MnCl2 for 1 hr. followed by DCA treatment as described

above. Recovered cells (5x104/well) were plated in triplicate onto pre-coated 96 well plates in

serum free medium containing 0.1% BSA and allowed to adhere for 30 min at 37oC. Adherent

cells were washed twice with PBS and fixed with 3.7% paraformaldehyde. Cells were then

stained with 0.1% crystal violet in 20% methanol for 20 min. The dye was aspirated and wells

washed three times with reagent grade water. The dye was extracted with 50% ethanol in 50mM

sodium citrate, pH 4.5, and quantified by measuring absorbance at 550nm on ELISA reader.

Page 6 of 45

7

Immunolocalization: HCA-7 cells were plated on glass slides. At 50% confluency medium was

aspirated and cells were washed twice in 1 ml of sterile PBS, fixed for 15 min with 2%

paraformaldehyde in PBS at 4oC. Cells were then washed 3 times with PBS and covered with

1ml of 1% glycerol in PBS for 15 min. Cells were permeabilized with 1 ml 0.2% Triton X-100 in

PBS. To prevent non-specific staining, slides were blocked with 1% goat serum and incubated

overnight at 4oC with phosphospecific FAK antibodies (1:1000). Next morning slides were

washed twice with PBS and blocked again as described above. Cells were then incubated with

goat anti-mouse rhodamine secondary antibodies (1:30) in block solution for 1 hr. After washing

FITC phalloidin (1:40) in PBS was added for 30 min. Slides were mounted with coverslips using

Slow Fade and Airvol and dried overnight in the dark. Fluorescent-labeled cells were then

viewed with the use of a laser scanning confocal microscope. (Model LSM 510, Zeiss).

Preparation of cytosolic cell extracts and immunoprecipitation: Pre-confluent cells were serum

starved for 18hr and then stimulated with DCA (100µM) for indicated times. Cells were lysed in

a buffer containing 25 mM HEPES pH 7.5, 150 mM NaCl, 1% Igepal CA-630, 1mM EDTA, 2%

glycerol, 1mM Na3VO4, 50mM NaF and Roche protease inhibitors cocktail. For study of

phosphatase activity, Na3VO4 and NaF were excluded from lysis buffer. After incubation on ice

for 15 min, the lysates were passaged several times through a 25-gauge needle, and insoluble

materials were removed by centrifugation at 4oC at 10,000g. The cell lysates (200µg protein)

were precleared by incubation with protein A/G agarose plus for 1 hr. The lysates were then

incubated with appropriate antibody (2µg for each sample) overnight at 4oC. To collect specific

immune complexes, 50µl of protein A/G agarose plus beads were added and sample was

incubated on a rotator at 4oC for 2 hr. The beads were washed three times with lysis buffer. In

some experiments serum starved HCA-7 cells were washed twice with PBS and detached gently

Page 7 of 45

8

with PBS containing 2 mM EDTA. Cells were maintained in suspension in PBS for 30 min at

37oC before the addition of DCA (100µM) for the indicated time periods. Cells were then lysed

and immunoprecipitated with anti-FAK antibodies as described above.

Western blotting: Proteins were extracted in SDS-containing Laemmli buffer and subjected to

quantitative Western blotting as described previously (4). Briefly, proteins (50µg) were separated

by SDS-PAGE using a 10% resolving polyacrylamide gel and electroblotted. Blots were

incubated overnight at 4°C with specific primary antibodies anti-FAK (1:5000), anti-ShP2

(1:3000), and anti-Src (1:5,000), followed by 1 hr incubation with appropriate peroxidase-

coupled secondary antibodies and subsequent detection by enhanced chemiluminescence.

In vitro FAK activity assay: FAK was immunoprecipitated from DCA treated cells with anti-

polyclonal FAK antibody and washed twice with lysis buffer and once with kinase buffer (10mM

Tris, pH 7.4, 10mM MnCl2, 2mM MgCl2, 0.02% Triton X-100). Immunoprecipitates were

incubated at 30oC for 30 min with 10 µM ATP (10µci/assay) alone (FAK autophosphorylation)

or 50µg poly (E4Y1) (Mr 20K-50K, Sigma) in 30 µl kinase buffer. In autophosphorylation

experiments, the reaction was terminated by the addition of SDS buffer. Samples were separated

on a 10% resolving gel, and the FAK band was identified by radioautography. In experiments

employing the synthetic peptide substrate, the reactions were terminated by spotting the

supernatant on p81 phosphocellulose strips. Strips were washed three times in 75 mM

phosphoric acid and the radioactivity was quantified by Cherenkov counting. Parallel

immunoprecipitates were probed for FAK abundance to ensure comparable kinase mass in

samples from control and DCA-treated cells.

Page 8 of 45

9

Immune complex phosphatase assay: HCA-7 cells were treated with DCA for indicated times

and immunoprecipitates of ShP1 and ShP2 were prepared as described above and washed three

times with cold lysis buffer. Phosphatase activity was assayed by resuspending the final pellet in

a total volume of 80 µl of reaction buffer (phosphatase buffer, pH 5.5, and containing 1mg/ml

bovine serum albumin, 5 mM EDTA, 10mM dithiothreitol). The reaction was initiated by the

addition of p-nitrophenyl phosphate (10mM, final concentration) for 30 min at 30oC. The

reaction was stopped by the addition of 0.9 ml of 1N NaOH, and the absorbance of the samples

was measured at 410 nm.

Tyrosine phosphatase activity and FAK tyrosine dephosphorylation assay: The Tyrosine

phosphatase activity of GST-ShP2 (catalytic domain) was determined by hydrolysis of p-

nitrophenyl phosphate in 80µl reaction mixture containing 50mM Hepes, pH 7.0, 20mM NaCl, 1

mM DTT, 10mM p-nitrophenyl phosphate and various amounts of the enzyme at 30oC for 10

min. The reaction was stopped by the addition of 80µl of 1M NaOH and the absorbance was

measured at 410 nM. One unit of tyrosine phosphatase was defined as the amount of enzyme that

produces 1 A410 unit in this assay.

For FAK-dephosphorylation assay, the FAK was immunoprecipitated from preconfluent

serum starved HCA-7 cells. Immunoprecipitates were incubated with different amounts of GST-

ShP2 in 50mM Hepes, pH 7.0, 20mM NaCl, 1mM DTT in 80µl reaction volume at 30oC for 10

min as described above. After the reaction, immunoprecipitates were washed three times with

lysis buffer and resolved by SDS-PAGE. FAK tyrosine dephosphorylation was then examined by

Western blotting with an antiphosphotyrosine antibody (Py-20).

Page 9 of 45

10

Antisense oligodeoxynucleotides transfection of HCA-7 cells: Biosynthesis, TX, synthesized

specific antisense oligodeoxynucleotides (ODNs) for human ShP2. Sequences were as follows:

Antisense, CTCCGCGATGTCATGTTCCT; Sense, GAGGAACATGACATCGCGGA;

Scrambled antisense TGGGTGTGTCCAAGAGAACT. The phosphorothioate modified ODNs

were resuspended in sterile water at 100µM concentrations. To transfect, cells were grown to 50-

70% confluency and then incubated with 1µM of oligodeoxynucleotides with oligofectamine

(Invitrogen) transfection reagent for 24 hours. A day after transfection, cells were serum starved

overnight and then treated with DCA.

FRNK Transfection of HCA-7 cells: A replication-deficient adenovirus (Adv) expressing GFP-

tagged FRNK was constructed as previously described (27). The Adv was propagated in human

embryonic kidney-293 cells (HEK-293) and purified by double CsCl centrifugation. As a

control for nonspecific effects of adenoviral infection, and Adv expressing GFP alone (Adv-

GFP) was used. Multiplicity of viral infection (MOI) was assayed by viral dilution in HEK-293

cells grown on 96 well plates. HCA-7 cells (50-70% confluent) were infected (10-100 MOI,

24hr) with either Adv-GFP or Adv-FRNK. Twenty-four hours after transfection cells were serum

starved overnight and then treated with or without DCA (100µM, 1hr). Fluorescent microscopy

and Western blotting was used to confirm expression of these constructs in colon cells.

ShP2 siRNA Transfection of HCA-7 cells: Validated synthetic siRNAs for ShP2 were purchased

from Qiagen (Valencia, CA). HCA-7 cells were grown to attain approximately 50-70%

confluence. Transfection of SiRNA was performed using HiPerFect transfection reagent

(Qiagen) to achieve a final RNA concentration of 5 nM. Cells were either treated with HiPerFect

(mock transfection) or siRNA. Twenty-four hours after transfection cells were serum starved

Page 10 of 45

11

overnight and then treated with or without DCA (100µM, 1hr). Cells were also lysed in lysis

buffer and Western blot analysis was performed using ShP2 antibodies.

Statistics: Data were expressed as means ±SD. All statistical analyses were performed by

student’s t-test. Values of p < 0.05 were considered statistically significant.

Results

DCA inhibits integrin-mediated cell adhesion: DCA inhibited cell adhesion to collagen-I (Fig.

1a), fibronectin (Fig. 1b) and laminin (Fig. 1c). Maximum inhibition was observed at 100µM

concentration. Further, adhesion of untreated HCA-7 cells on collagen-I (Fig. 1d), fibronectin

(Fig. 1e), and laminin (Fig. 1f) was up regulated with dose-dependent increase in matrix

concentration. However, similar adhesion kinetics was not observed in DCA-treated cells.

Adhesion to poly-L-lysine, which does not require integrins, was not affected by DCA-treatment

(Fig. 1g). We chose to study these matrix proteins because collagen-I represents the dominant

collagen of the interstitial matrix beneath the basement membrane, laminin is the dominant non-

collagenous constituent of the basement membrane itself, and serum fibronectin is a primary

constituent of blood clots, which often form on bleeding mucosal lesions.

Viability of HCA-7 cells after DCA treatment: Cell viability of DCA-treated cells was assessed

with the MTT colorimetric assay, and propidium iodide staining. In MTT assay, mitochondrial

dehydrogenases of viable cells convert MTT to purple formazan crystals. There was no change

in MTT conversion in cells treated with 100µM DCA for 3 hours (negative data not shown).

Total cell death was also measured using propidium iodide, which indicates membrane integrity.

Viable cells exclude the dye. There was no significant cell death after DCA (100µM, 3 hrs)

treatment (negative data not shown).

Page 11 of 45

12

DCA regulates FAK phosphorylation on specific residues: Focal adhesion complexes play an

important role in the modulation of cell adhesion, which involves integrins. To investigate the

mechanism by which DCA inhibits cell adhesion, we examined the effects of DCA stimulation

on the phosphorylation status of the FAK, which is implicated in cell adhesion. After DCA

(100µM) treatment, cell lysates were immunoprecipitated with anti-FAK antibody and further

analyzed by Western blotting with an antiphosphotyrosine antibody. FAK was rapidly

dephosphorylated in a time dependent manner (Fig. 2a). The membranes were stripped and re-

probed with the FAK antibody to ensure that similar amounts of total FAK had been precipitated

from each sample. In some experiments serum starved HCA-7 cells were rinsed with PBS and

harvested with 2mM EDTA in PBS. Cells were gently detached, maintained in suspension for 30

min, and subsequently stimulated with DCA (100µM) for the indicated time periods. Cell lysates

were immunoprecipitated with anti-FAK antibody and further analyzed by Western blotting with

an antiphosphotyrosine antibody. In these conditions we observed only a weak phosphorylation

of FAK in unstimulated cells compared to adherent cells. Further, DCA treatment, in suspension,

had no significant effect on FAK phosphorylation (Fig. 2b). In order to determine whether DCA

effect on FAK dephosphorylation is not restricted to a single cell line we utilized HT-29 cells,

which also express high levels of FAK. DCA treatment significantly reduced FAK

phosphorylation in a time dependent manner (Fig. 2c). As the kinase activities of FAK are

modulated by its degree of tyrosine phosphorylation, dephosphorylation of FAK may lead to

reduction of its kinase activities. However, FAK immunoprecipitated from cells treated with

DCA (100µM, 30 minutes) showed no reduction in vitro autophosphorylation activity compared

with FAK from untreated cells (negative data not shown). However, FAK precipitated from cells

Page 12 of 45

13

stimulated with DCA showed reduction in kinase activity towards an exogenous substrate, poly

(E4Y1) (Fig. 2d).

FAK can be tyrosine phosphorylated on a number of tyrosine residues, including Tyr-397, -

925 and -576/577 in response to various stimuli. To map DCA-responsive phosphorylation sites

on FAK, we utilized a series of phosphospecific antibodies. DCA rapidly dephosphorylated Tyr-

925 and -576 /577. However, DCA had little or no effect on the phosphorylation status of Tyr-

397 (Fig. 3).

Tyr-397 is the major autophosphorylation site of FAK that associates with c-Src during

integrin activation. We investigated c-Src association with FAK by utilizing phospho-active Src

and native Src antibodies (Fig. 4). HCA-7 cells were treated with DCA (100µM, 30 minutes),

FAK was immunoprecipitated and probed for phospho-active FAK (Tyr-397) or native Src or

phospho-active Src (Tyr-418). DCA treatment significantly activated c-Src as assessed by

increased tyrosine phosphorylation of Src at Tyr-418 (Fig. 4a). Src was constitutively associated

with FAK as detected by co-immunoprecipitation analyses and there was no further increase in

Src recruitment after DCA treatment (Fig. 4a). Src-family kinase inhibitor PP2 decreased the

tyrosine phosphorylation of Src (Tyr-418) and also decreased the tyrosine phosphorylation of

FAK (Tyr-397), detected by co-immunoprecipitation experiments (Fig. 4b).

FAK localizes to focal adhesions: To confirm the significance of DCA mediated changes in the

tyrosine phosphorylation of FAK, we examined the existence of these events in vivo. HCA-7

cells were treated with DCA (100µM, 30 min) and FAK Tyr-397, and FAK Tyr-925

(phosphorylated forms of FAK) were analyzed by dual labeling immunofluorescence using

polyclonal antibodies against phosphorylated forms of FAK (red color) and FITC-conjugated

Page 13 of 45

14

phalloidin (green color) to identify f-actin filaments. In control cells, immunostaining of FAK

Tyr-925 was predominantly co-localized with actin containing focal adhesion complexes at the

cell-substratum interface. However, DCA (100µM, 30 min) caused a dramatic loss of staining

intensity at focal adhesions (Fig. 5). Similar results were obtained by staining cells with FAK

Tyr-576/577 (data not shown). Analysis of morphology of focal contacts revealed that control

cells were anchored to the substratum by mature focal adhesion complexes that encircled the

entire cell. In contrast, when the cells were activated with DCA small focal adhesion points were

either accumulated at one pole of the cell or partially lost. However, DCA treatment had no

effect on immunostaining of FAK Tyr-397 (Fig. 5).

DCA stimulates tyrosine phosphatase activity: To explore the mechanism and significance of

FAK dephosphorylation by DCA, we treated HCA-7 cells with phenylarsine oxide (PAO), an

inhibitor of protein tyrosine phosphatases (PTPases) (23). Cells were then treated with DCA (100

µM) for the indicated times. FAK was immunoprecipitated and blotted for phosphotyrosine as

described above. Pretreatment of cells with PAO (10µM, 10 min) blocked the dephosphorylation

of FAK by DCA (Fig. 6a). To determine the potential involvement of PTPases in DCA-mediated

dephosphorylation of FAK, we analyzed the effect of DCA on phosphorylation status of ShP1 and

ShP2, the SH2 domain containing PTPases. DCA treatment significantly increased tyrosine

phosphorylation of ShP2 but not ShP1 in a time dependent manner (Fig. 6b). This suggests a

relatively early involvement of ShP2 in DCA signaling.

Tyrosine phosphorylation of ShP2 has been correlated with its activation (67). Next, we

assessed whether DCA stimulated the enzyme activity of ShP1 and ShP2. Immunoprecipitated

ShP1 and ShP2 from control or DCA treated cells were assayed for phosphatase activity using

pNPP as a substrate. A two to three fold increase in phosphatase activity was consistently observed

Page 14 of 45

15

in ShP2 immunoprecipitates and significant activity was detectable as late as 30 min after DCA

stimulation (Fig. 6b). However, ShP1 activity was not significantly different.

Increased association of ShP2 with FAK upon DCA stimulation: The rapid and specific

dephosphorylation of FAK upon DCA treatment and its inhibition by PTPase inhibitors are

consistent with the involvement of PTPases in this process. ShP2 is a PTPase consisting of two Src

homology (SH2) domains followed by a catalytic phosphatase domain. To determine whether

ShP2 is associated with FAK, FAK was immunoprecipitated and immunoprecipitates were

subjected to Western blotting with anti-ShP2 antibody. Stimulation with DCA significantly

increased the association of ShP2 with FAK in a time dependent manner (Fig. 7).

ShP2 dephosphorylates FAK in vitro: To evaluate whether ShP2 was responsible for FAK

tyrosine dephosphorylation in DCA-stimulated cells, we immunoprecipitated FAK from serum

starved HCA-7 cells and incubated with various amounts of a GST fusion protein of the ShP2

catalytic domain. Incubation of FAK with GST-ShP2 fusion protein resulted in tyrosine

dephosphorylation of FAK in a concentration dependent manner (Fig. 8a). The GST control

protein had no apparent effect on FAK dephosphorylation in vitro. These results demonstrate that

FAK is a substrate of ShP2.

We next examined whether FAK tyrosine dephosphorylation by ShP2, in vitro, resulted in loss

of FAK-associated Src. FAK immunoprecipitates from serum starved HCA-7 cells were incubated

with GST-ShP2 or with GST as a control. The immune complex was washed to remove proteins

that had been dissociated from the immune complex. The immune complex was then analyzed by

Western blotting for total Src and FAK. Incubation of FAK with ShP2, in vitro, did not affect Src

binding to FAK (Fig. 8b). These results demonstrate that binding of Src to FAK is independent of

FAK tyrosine dephosphorylation by ShP2.

Page 15 of 45

16

Antisense ODN directed against ShP2 decreased ShP2 protein levels and attenuated DCA induced

FAK dephosphorylation: To determine whether ShP2 was necessary for DCA induced FAK

dephosphorylation we used anti-sense oligonucleotides approach to inhibit endogenous levels of

ShP2 protein (49). A conventional pharmacological inhibitor that selectively targets ShP2 is

currently not available. The use of an antisense ODN that selectively down-regulates the levels of

ShP2 has been reported. HCA-7 cells were transfected with specific ShP2 antisense ODNs for 24

hours without serum to protect stability of ODNs. The ShP2 antisense ODN but not the nonsense

ODN sequence significantly decreased the expression of ShP2 protein in HCA-7 cells (Fig. 9a).

Antisense ODNs did not alter expression of a closely related phosphatase ShP1 (negative data not

shown). The viability of cells assessed at this time always exceeded 90% and was not different

from control cells, indicating that at concentration of 1µM the ODNs were non-toxic to the cells.

For FAK-dephosphorylation experiments, cells were transfected with antisense ODNs as described

above. After 24 hours, cells were treated with DCA (100µM, 60 min). FAK remained

phosphorylated for up to 1 hr after stimulated with DCA in cells transfected with ShP2 antisense

compared to control (Fig. 9b).

Manganese partially reversed the DCA-response on cell adhesion: To determine whether the

DCA effect on cell adhesion is solely on integrin confirmation, integrins were fully activated by

incubating cells with manganese. The Mn+2 cells adhered more than the cells without Mn+2.

Though, as compared to control, there was a partial reversal of the DCA effect on cell adhesion,

DCA treatment of Mn+2 cells still demonstrated a significant reduction in cell adhesion (Fig. 10).

GFP-tagged FRNK decreased the FAK expression, activation and attenuated the DCA effect on

cell adhesion: To determine whether FAK expression and activation are important determinants

of DCA induced loss of adhesion, we utilized adenoviral (Adv) FRNK constructs. Infection of

Page 16 of 45

17

HCA-7 cells with Adv-GFP-tagged FRNK (10-100 MOI) demonstrated a dose-dependent

decrease in FAK expression and FAK Tyr-397 activation, with a maximum effect observed at

100 MOI (Fig. 11b) Further, FRNK-infection significantly decreased adhesion of HCA-7 cells

on collagen-I and fibronectin (Fig. 11c) compared to GFP-expressing cells. As predicted, DCA

significantly decreased cell adhesion in Adv-GFP-infected cells (control). However, DCA had no

effect on cell adhesion of FRNK-infected cells (Fig. 11c).

ShP2 siRNA abolished the DCA effect on cell adhesion: Finally, we were prompted to determine

whether ShP2 was necessary for the DCA induced loss of adhesion. siRNA directed gene

silencing in colon cells decreased the ShP2 protein expression (Inset, Fig. 12). Total FAK

expression (unrelated gene) was not altered in these cells (Inset, Fig. 12). Decrease in ShP2 level

resulted in slight increase in basal cell adhesion of control cells. However, cell adhesion was not

inhibited by DCA in ShP2 reduced cells compared to mock transfected cells (Fig. 12).

Discussion

The genesis of these studies was an initial observation that exposure of cultured HCA-7 colon

cancer cells to DCA reduced the cell adhesion to extra cellular matrices. In an effort to better

understand this phenomenon, we examined several aspects of FAK signaling that are believed to

be involved in cell adhesion. Our results indicate that physiological concentrations of DCA

dephosphorylated FAK on specific tyrosine residues and PTPase inhibitors block such

dephosphorylation. Our data indicate that ShP2 may be one PTPase involved in this process. These

results also demonstrated that DCA stimulation triggers FAK/ShP-2 association and FAK

dephosphorylation was differentially regulated. Bile acids, end product of cholesterol metabolism,

participate in fat digestion in the gastrointestinal tract. Majority of bile acids present in millimolar

ranges during digestion are efficiently absorbed via both active and passive mechanisms by the

Page 17 of 45

18

small intestine, and only a small fraction of bile acid escapes from enterohepatic circulation

resulting in the bile acid concentrations in the cecum being in micromolar ranges (43). Previous

reports have indicated no evidence of cytotoxicity when gastric mucosal cells were exposed to

DCA (50-300µM)(57). Further, total concentrations of bile acids in the aqueous phase of stool

ranges from 50-300µM and DCA is the main soluble bile acid ranging from 10-200µM (68). We

first established what DCA treatments would not be toxic to the cells. As measures of toxicity, we

assayed the cells ability to oxidize MTT and to exclude propidium iodide. For all these measures,

doses of DCA employed in these studies caused no detectable toxicity to the cells, indicating that

the results observed were not due to overt cytotoxicity.

Accumulating evidence suggests that the secondary bile acids are involved in the progression

of colon cancer cells to a more invasive phenotype and that this may involve reorganization of

cytoskeleton architecture. Here we investigated the effects of DCA-induced early signaling on

the focal adhesion proteins FAK. Our findings suggest that DCA regulates the tyrosine

phosphorylation of focal adhesion proteins, but not all tyrosine sites are targets of DCA

signaling. DCA-induced dephosphorylation of FAK depends on integrin arrangement, since it

does not occur in cells that are in suspension.

FAK co-localizes with integrins in focal adhesions. Tyrosine phosphorylation of FAK and its

catalytic activity stimulation is dependent upon integrin mediated cell adhesion. FAK contains

large N- and C-terminal domains that flank the catalytic domain. At the amino-terminus of FAK

is a FERM domain (41). Between the FERM and catalytic domains is Tyr-397, which is the

major autophosphorylation site of FAK. On its activation, Tyr-397 is a docking site for the SH2

domains of a number of proteins including Src (12), p85 subunit of PI3K (10), phospholipase C-γ

Page 18 of 45

19

and Grb7 (25). In DCA treated cells phosphorylation at FAK autophosphorylation site Tyr-397

was preserved, however, it induced dephosphorylation of FAK at Tyr-576, -577 and -925.

Further, FAK immunoprecipitated from cells treated with DCA showed no change in

autophosphorylation activity compared with FAK from untreated cells. We further investigated

co-association of Src with FAK in DCA stimulated cells. Results demonstrated that Src was

constitutively associated with FAK. DCA treatment significantly activated c-Src as assessed by

increased tyrosine phosphorylation of Src at Tyr-418 with no further increase in Src recruitment

after DCA treatment. Previous studies have demonstrated that the interaction of FAK with the

Src SH2 domain renders the autophosphorylation site of FAK resistant to the action of tyrosine

phosphatases (12). In our studies in vitro experiments also demonstrated that binding of Src to

FAK is independent of FAK tyrosine dephosphorylation by ShP2. In HCA-7 cells, therefore,

constitutive association of Src with FAK and activation of Src upon DCA treatment may explain

preservation of the Tyr-397 phosphorylation despite Tyr-576/577 and Tyr-925

dephosphorylation in the presence of DCA, a unique FAK/Src regulation by tumor promoter bile

acid. Further, the c-Src/FAK relationship appears to be reciprocal as FAK phosphorylation on

Tyr-397 is required for c-Src binding to FAK SH domain. In hepatic cells, FAK/c-Src

association is adhesion dependent (8), while in human colon cancer cells, increased c-Src activity

disrupts adhesion (31). Similarly in our studies, increased c-Src activation by DCA may be

responsible for decreased cell adhesion.

Two phosphorylation sites in the activation loop of the catalytic domain, tyr-576, and -577,

are regulatory sites of phosphorylation that enhance catalytic activity. At the carboxyl-terminus

of FAK is a 150-amino acid focal adhesion-targeting (FAT) domain. This domain is essential for

localization of FAK to sites of cell-ECM adhesion (13). Tyr-925, in the FAT domain of FAK,

Page 19 of 45

20

can bind to SH2 domain of Grb-2 and presents one of the mechanisms of activation of the

Ras/MAPK pathway (60). DCA selectively dephosphorylated FAK at Tyr-576, -577 and -925

and reduced FAK kinase activity towards a model, exogenous substrate (E4Y1). However, FAK

tyrosine phosphorylation at the Tyr-397 autophosphorylation site was preserved. The

dissociation of FAK autocatalytic activity from its activity against exogenous substrates is

consistent with what is known about the role of the Tyr-576/-577 phosphorylation sites in the

regulation of FAK-dependent signaling (59). FAK phosphorylates itself at Tyr-397 during

integrin engagement and clustering, which provides a docking site for Src family kinases to bind

to FAK via their SH2 domains. Src then phosphorylates FAK at multiple sites, including Tyr-

576 and -577 within the catalytic subdomain VIII present in many different protein tyrosine

kinases. FAK autophosphorylation at Tyr-397 is thus required for adhesion-dependent Tyr-576/-

577 phosphorylation of FAK by Src, which in turn increases FAK catalytic activity at both the

Tyr-397 site, as well as towards exogenous substrates (58). Calalb et al. (6) have previously

shown that Y→F mutations at Tyr-576 and -577 reduced FAK kinase activity approximately

50% against a similar, model substrate (E4Y1), which is similar to the degree of inhibition in

extrinsic kinase activity observed in immunoprecipitates following mutation of the Tyr-397

autophosphorylation site. Although the physiological significance of decreased Tyr-576/-577

phosphorylation on the phosphorylation of endogenous substrates (other than FAK itself)

remains unknown, ShP2-dependent dephosphorylation of Tyr-576/-577 would be expected to

reduce downstream signaling, in a fashion similar to the effect of dephosphorylation or mutation

of catalytic subdomain VIII sites on other protein tyrosine kinases (19).

Since tyrosine phosphorylation of FAK regulates its catalytic activity and association with

other signaling molecules, dephosphorylation of tyrosine residues is potentially a very important

Page 20 of 45

21

mechanism in regulating FAK signaling in DCA treated cells. The fact that DCA differentially

dephosphorylated tyrosine residues of FAK raises the possibility that distinct protein tyrosine

phosphatases may be responsible for dephosphorylation of different sites. Since different

phosphorylation sites function to regulate catalytic activity and protein-protein interactions, site-

specific dephosphorylation of FAK by DCA may be an effective mechanism to modulate some

aspects of FAK signaling independently from others. A number of PTPase have been suggested

as regulators of FAK, including PTP-PEST (62), PTP1B (35), ShP2 (44), and PTEN (24).

Experiments with the tyrosine phosphatase inhibitor PAO support the involvement of tyrosine

phosphatases in DCA-induced cytoskeleton signaling. DCA triggered tyrosine phosphorylation

of ShP-2 and its association with FAK. To test whether ShP2 was necessary for DCA-stimulated

FAK dephosphorylation, we decreased ShP2 protein levels via the use of antisense methodology.

We found that antisense ODN but not control ODN significantly attenuated DCA-stimulated

FAK dephosphorylation.

Conformational changes in integrins have been suggested to underlie the modulation of their

affinity for ECM components. If DCA treatment did indeed inhibit cell adhesion by converting

integrins into an inactive confirmation, then treatment of cells with manganese, which uniformly

activates integrins by inducing an active confirmation (21), should be able to reverse the

inhibitory effects. Manganese increased the adhesion of control cells, presumably by binding

extracellular domains directly and altering integrin-binding affinity. Previous studies have shown

that Ca+2 decreased colon cancer cell adhesion and that both Mg+2 and Mn+2 increased cell

adhesion (18, 66). Earlier studies in several other cell types have reported that Mn+2 enhances the

adhesive properties of many integrins (16, 64). Small changes in divalent cation concentration

can substantially stimulate or inhibit integrin binding under physiological conditions (66).

Page 21 of 45

22

Further, the affinity of integrins for their extracellular ligands can also be regulated from within

the cells through intracellular signaling that may be mediated through integrin cytoplasmic

domains (36). In our investigations, addition of manganese to medium didn’t completely reverse

the inhibitory effects of DCA on cell adhesion, indicating a possible inside-out signaling upon

DCA treatment.

Although outside-in signaling pathways from integrin to the nucleus have been well studied,

inside-out pathways for cell adhesion have received little attention. Many intracellular pathways

can control inside-out regulation of integrin functions (30). Such inside-out signaling could be

mediated by FAK activation (32, 50). More recently, extracellular pressure promoted the colon

cancer cell adhesion via actin dependent inside-out FAK and Src signals and FAK inhibition by

FRNK transfection prevented the effect of pressure on cell adhesion (65). We also demonstrated

that GFP-FRNK, the C-terminal non-catalytic domain of FAK that competes with FAK for

localization in focal adhesions and interferes with FAK signaling, significantly inhibited FAK

expression and tyrosine phosphorylation and reduced cell adhesion. However, the inhibitory

effect of DCA on cell adhesion was not observed in FRNK infected cells, suggesting that FRNK

and DCA interfere with cell adhesion via the same signal transduction pathway. These results

also point towards a possible inside-out signaling mechanism by DCA in these cells. Further,

tyrosine phosphatase ShP2 was responsible for DCA induced FAK dephosphorylation in colon

cells. Moreover, SiRNA directed ShP2 reduction in these cells activated colon cancer cell

adhesion and abolished the DCA effect on cell adhesion. The increased cell adhesion in ShP2

reduced cells and the absence of inhibitory effect of DCA could be due to elevated levels of

phosphorylated FAK in the absence of ShP2, as observed in antisense experiments.

Page 22 of 45

23

Several receptor tyrosine kinases have been reported to reduce tyrosine phosphorylation of

FAK, including the EGFR (37), Insulin receptor (2), IGF-1R (40) and EphA2 (44). Since bile

acids have no known transmembrane receptors, trans-activation of receptor tyrosine kinases by

DCA may be responsible for reduced tyrosine phosphorylation and catalytic activity of FAK. In

this regard, DCA has been implicated in ligand independent activation of EGFR and insulin

receptor but not IGF-1R in primary rat hepatocytes (26). Further, bile acid transporters have not

been identified in the colon and labeled bile acids are not taken up by colon cancer cells, in vitro,

(53). It is presumed that these amphipathic molecules exert their effects by perturbing plasma

membrane structure and function. DCA is known to deplete cholesterol from plasma membranes

(28). A number of key signaling molecules, including EGFR and Ras, partition into cholesterol-

rich lipid rafts within the plasma membrane (45, 54). Cholesterol depletion has been shown to

activate EGFR and Ras signaling in other cells (11).

Our results define a new signaling pathway, originating with DCA-induced FAK

dephosphorylation. In summary, our results suggest that DCA differentially regulates signaling

from focal adhesion complexes through selective phosphorylation/dephosphorylation or through

association of participating components and that these regulatory events may have distinct roles

in diverse physiological and disease processes including inflammation, wound healing and

cancer progression.

Page 23 of 45

24

References:

1. Augenlicht LH, Anthony GM, Church TL, Edelmann W, Kucherlapati R, Yang K, Lipkin M, and Heerdt BG. Short-chain fatty acid metabolism, apoptosis, and Apc-initiated tumorigenesis in the mouse gastrointestinal mucosa. Cancer Res 59: 6005-6009, 1999.2. Baron V, Calleja V, Ferrari P, Alengrin F, and Van Obberghen E. p125Fak focal adhesion kinase is a substrate for the insulin and insulin-like growth factor-I tyrosine kinase receptors. J Biol Chem 273: 7162-7168, 1998.3. Bayerdörffer E, Mannes GA, Ochsenkuhn T, Dirschedl P, Wiebecke B, and Paumgartner G. Unconjugated secondary bile acids in the serum of patients with colorectal adenomas. Gut 36: 268-273, 1995.4. Bissonnette M, Khare S, von Lintig FC, Wali RK, Nguyen L, Zhang Y, Hart J, Skarosi S, Varki N, Boss GR, and Brasitus TA. Mutational and nonmutational activation of p21ras in rat colonic azoxymethane-induced tumors: effects on mitogen-activated protein kinase, cyclooxygenase-2, and cyclin D1. Cancer Res 60: 4602-4609, 2000.5. Burridge K and Chrzanowska-Wodnicka M. Focal adhesions, contractility, and signaling. Annu Rev Cell Dev Biol 12: 463-518, 1996.6. Calalb MB, Polte TR, and Hanks SK. Tyrosine phosphorylation of focal adhesion kinase at sites in the catalytic domain regulates kinase activity: a role for Src family kinases. Molecular and Cellular Biology 15: 954-963, 1995.7. Cance WG, Harris JE, Iacocca MV, Roche E, Yang X, Chang J, Simkins S, and Xu L. Immunohistochemical analyses of focal adhesion kinase expression in benign and malignant human breast and colon tissues: correlation with preinvasive and invasive phenotypes. Clin Cancer Res 6: 2417-2423, 2000.8. Carloni V, Defranco RM, Caligiuri A, Gentilini A, Sciammetta SC, Baldi E, Lottini B, Gentilini P, and Pinzani M. Cell adhesion regulates platelet-derived growth factor-induced MAP kinase and PI-3 kinase activation in stellate cells. Hepatology 36: 582-591, 2002.9. Chaplin MF. Bile acids, fibre and colon cancer: the story unfolds. J R Soc Health 118: 53-61., 1998.10. Chen HC and Guan JL. Association of focal adhesion kinase with its potential substrate phosphatidylinositol 3-kinase. Proc Natl Acad Sci U S A 91: 10148-10152, 1994.11. Chen X and Resh MD. Cholesterol depletion from the plasma membrane triggers ligand-independent activation of the epidermal growth factor receptor. J Biol Chem 277: 49631-49637, 2002.12. Cobb BS, Schaller MD, Leu TH, and Parsons JT. Stable association of pp60src and pp59fyn with the focal adhesion-associated protein tyrosine kinase, pp125FAK. Mol Cell Biol14: 147-155, 1994.13. Cooley MA, Broome JM, Ohngemach C, Romer LH, and Schaller MD. Paxillin binding is not the sole determinant of focal adhesion localization or dominant-negative activity of focal adhesion kinase/focal adhesion kinase-related nonkinase. Mol Biol Cell 11: 3247-3263, 2000.14. Craven PA, Pfanstiel J, and DeRubertis FR. Role of reactive oxygen in bile salt stimulation of colonic epithelial proliferation. Journal of Clinical Investigation 77: 850-859, 1986.15. Debruyne PR, Bruyneel EA, Li X, Zimber A, Gespach C, and Mareel MM. The role of bile acids in carcinogenesis. Mutat Res 480-481: 359-369., 2001.

Page 24 of 45

25

16. Dransfield I, Cabanas C, Craig A, and Hogg N. Divalent cation regulation of the function of the leukocyte integrin LFA-1. J Cell Biol 116: 219-226, 1992.17. Earnest DL, Holubec H, Wali RK, Jolley CS, Bissonnette M, Bhattacharyya AK, Roy H, Khare S, and Brasitus TA. Chemoprevention of azoxymethane-induced colonic carcinogenesis by supplemental dietary ursodeoxycholic acid. Cancer Research 54: 5071-5074, 1994.18. Ebert EC. Mechanisms of colon cancer binding to substratum and cells. Dig Dis Sci 41: 1551-1556, 1996.19. Ellis L, Clauser E, Morgan DO, Edery M, Roth RA, and Rutter WJ. Replacement of insulin receptor tyrosine residues 1162 and 1163 compromises insulin-stimulated kinase activity and uptake of 2-deoxyglucose. Cell 45: 721-732, 1986.20. Gabarra-Niecko V, Schaller MD, and Dunty JM. FAK regulates biological processes important for the pathogenesis of cancer. Cancer Metastasis Rev 22: 359-374, 2003.21. Gailit J and Ruoslahti E. Regulation of the fibronectin receptor affinity by divalent cations. J Biol Chem 263: 12927-12932, 1988.22. Goodwin B and Kliewer SA. Nuclear receptors. I. Nuclear receptors and bile acid homeostasis. Am J Physiol Gastrointest Liver Physiol 282: G926-931, 2002.23. Guvakova MA and Surmacz E. The activated insulin-like growth factor I receptor induces depolarization in breast epithelial cells characterized by actin filament disassembly and tyrosine dephosphorylation of FAK, Cas, and paxillin. Exp Cell Res 251: 244-255, 1999.24. Haier J and Nicolson GL. PTEN regulates tumor cell adhesion of colon carcinoma cells under dynamic conditions of fluid flow. Oncogene 21: 1450-1460, 2002.25. Han DC and Guan JL. Association of focal adhesion kinase with Grb7 and its role in cell migration. J Biol Chem 274: 24425-24430, 1999.26. Han SI, Studer E, Gupta S, Fang Y, Qiao L, Li W, Grant S, Hylemon PB, and Dent P. Bile acids enhance the activity of the insulin receptor and glycogen synthase in primary rodent hepatocytes. Hepatology 39: 456-463, 2004.27. Heidkamp MC, Bayer AL, Kalina JA, Eble DM, and Samarel AM. GFP-FRNK disrupts focal adhesions and induces anoikis in neonatal rat ventricular myocytes. Circ Res 90: 1282-1289, 2002.28. Heuman DM and Bajaj R. Ursodeoxycholate conjugates protect against disruptions of cholesterol-rich membranes by bile salts. Gastroenterology 106: 1333-1341, 1994.29. Hofmann AF. Bile acid secretion, bile flow and biliary lipid secretion in humans. Hepatology 12: 17S-22S; discussion 22S-25S, 1990.30. Hughes PE and Pfaff M. Integrin affinity modulation. Trends Cell Biol 8: 359-364, 1998.31. Irby RB and Yeatman TJ. Increased Src activity disrupts cadherin/catenin-mediated homotypic adhesion in human colon cancer and transformed rodent cells. Cancer Res 62: 2669-2674, 2002.32. Kharbanda S, Saleem A, Yuan Z, Emoto Y, Prasad KV, and Kufe D. Stimulation of human monocytes with macrophage colony-stimulating factor induces a Grb2-mediated association of the focal adhesion kinase pp125FAK and dynamin. Proc Natl Acad Sci U S A 92: 6132-6136, 1995.33. Kulkarni N, Zang E, Kelloff G, and Reddy BS. Effect of the chemopreventive agents piroxicam and D,L-alpha- difluoromethylornithine on intermediate biomarkers of colon carcinogenesis. Carcinogenesis 13: 995-1000., 1992.

Page 25 of 45

26

34. Lechner S, Muller-Ladner U, Renke B, Rüschoff J, Schölmerich J, and Kullmann F.Gene expression profiling of aberrant crypts in colon carcinogenesis. Gastroenterology 122: A-123, 2002.35. Liu F, Sells MA, and Chernoff J. Protein tyrosine phosphatase 1B negatively regulates integrin signaling. Curr Biol 8: 173-176, 1998.36. Longhurst CM and Jennings LK. Integrin-mediated signal transduction. Cell Mol Life Sci 54: 514-526, 1998.37. Lu Z, Jiang G, Blume-Jensen P, and Hunter T. Epidermal growth factor-induced tumor cell invasion and metastasis initiated by dephosphorylation and downregulation of focal adhesion kinase. Mol Cell Biol 21: 4016-4031, 2001.38. Makishima M, Lu TT, Xie W, Whitfield GK, Domoto H, Evans RM, Haussler MR, and Mangelsdorf DJ. Vitamin D Receptor As an Intestinal Bile Acid Sensor. Science 296: 1313-1316., 2002.39. Makishima M, Okamoto AY, Repa JJ, Tu H, Learned RM, Luk A, Hull MV, Lustig KD, Mangelsdorf DJ, and Shan B. Identification of a nuclear receptor for bile acids. Science284: 1362-1365, 1999.40. Manes S, Mira E, Gomez-Mouton C, Zhao ZJ, Lacalle RA, and Martinez AC.Concerted activity of tyrosine phosphatase SHP-2 and focal adhesion kinase in regulation of cell motility. Mol Cell Biol 19: 3125-3135, 1999.41. Mangeat P, Roy C, and Martin M. ERM proteins in cell adhesion and membrane dynamics. Trends Cell Biol 9: 187-192, 1999.42. Marsh KA, Stamp GW, and Kirkland SC. Isolation and characterization of multiple cell types from a single human colonic carcinoma: tumourigenicity of these cell types in a xenograft system. J Pathol 170: 441-450, 1993.43. McJunkin B, Fromm H, Sarva RP, and Amin P. Factors in the mechanism of diarrhea in bile acid malabsorption: fecal pH--a key determinant. Gastroenterology 80: 1454-1464, 1981.44. Miao H, Burnett E, Kinch M, Simon E, and Wang B. Activation of EphA2 kinase suppresses integrin function and causes focal-adhesion-kinase dephosphorylation. Nat Cell Biol2: 62-69, 2000.45. Mineo C, James GL, Smart EJ, and Anderson RG. Localization of epidermal growth factor-stimulated Ras/Raf-1 interaction to caveolae membrane. J Biol Chem 271: 11930-11935., 1996.46. Morgan DM. Tetrazolium (MTT) assay for cellular viability and activity. Methods Mol Biol 79: 179-183, 1998.47. Nagengast FM, Grubben MJ, and van Munster IP. Role of bile acids in colorectal carcinogenesis. European Journal of Cancer 31A: 1067-1070, 1995.48. Ochsenkuhn T, Bayerdorffer E, Meining A, Schinkel M, Thiede C, Nussler V, Sackmann M, Hatz R, Neubauer A, and Paumgartner G. Colonic mucosal proliferation is related to serum deoxycholic acid levels. Cancer 85: 1664-1669., 1999.49. Pazdrak K, Adachi T, and Alam R. Src homology 2 protein tyrosine phosphatase (SHPTP2)/Src homology 2 phosphatase 2 (SHP2) tyrosine phosphatase is a positive regulator of the interleukin 5 receptor signal transduction pathways leading to the prolongation of eosinophil survival. J Exp Med 186: 561-568, 1997.50. Pichard V, Honore S, Kovacic H, Li C, Prevot C, Briand C, and Rognoni JB.Adhesion, actin cytoskeleton organisation and the spreading of colon adenocarcinoma cells

Page 26 of 45

27

induced by EGF are mediated by alpha2beta1 integrin low clustering through focal adhesion kinase. Histochem Cell Biol 116: 337-348, 2001.51. Playford MP and Schaller MD. The interplay between Src and integrins in normal and tumor biology. Oncogene 23: 7928-7946, 2004.52. Potter JD, Slattery ML, Bostick RM, and Gapstur SM. Colon cancer: a review of the epidemiology. Epidemiologic Reviews 15: 499-545, 1993.53. Powell AA, LaRue JM, Batta AK, and Martinez JD. Bile acid hydrophobicity is correlated with induction of apoptosis and/or growth arrest in HCT116 cells. Biochem J 356: 481-486., 2001.54. Puri C, Tosoni D, Comai R, Rabellino A, Segat D, Caneva F, Luzzi P, Di Fiore PP, and Tacchetti C. Relationships between EGFR signaling-competent and endocytosis-competent membrane microdomains. Mol Biol Cell 16: 2704-2718, 2005.55. Qiao L, Studer E, Leach K, McKinstry R, Gupta S, Decker R, Kukreja R, Valerie K, Nagarkatti P, Deiry WE, Molkentin J, Schmidt-Ullrich R, Fisher PB, Grant S, Hylemon PB, and Dent P. Deoxycholic acid (dca) causes ligand-independent activation of epidermal growth factor receptor (egfr) and fas receptor in primary hepatocytes: inhibition of EGFR/mitogen-activated protein kinase- signaling module enhances dca-induced apoptosis. Mol Biol Cell 12: 2629-2645., 2001.56. Reddy BS, Mangat S, Sheinfil A, Weisburger JH, and Wynder EL. Effect of type and amount of dietary fat and 1,2-dimethylhydrazine on biliary bile acids, fecal bile acids, and neutral sterols in rats. Cancer Res 37: 2132-2137., 1977.57. Redlak MJ, Dennis MS, and Miller TA. Apoptosis is a major mechanism of deoxycholate-induced gastric mucosal cell death. Am J Physiol Gastrointest Liver Physiol 285: G870-879, 2003.58. Ruest PJ, Roy S, Shi E, Mernaugh RL, and Hanks SK. Phosphospecific antibodies reveal focal adhesion kinase activation loop phosphorylation in nascent and mature focal adhesions and requirement for the autophosphorylation site. Cell Growth Differ 11: 41-48, 2000.59. Schaller MD. Biochemical signals and biological responses elicited by the focal adhesion kinase. Biochim Biophys Acta 1540: 1-21, 2001.60. Schlaepfer DD, Hanks SK, Hunter T, and van der Geer P. Integrin-mediated signal transduction linked to Ras pathway by GRB2 binding to focal adhesion kinase. Nature 372: 786-791, 1994.61. Schlaepfer DD and Hunter T. Focal adhesion kinase overexpression enhances ras-dependent integrin signaling to ERK2/mitogen-activated protein kinase through interactions with and activation of c-Src. J Biol Chem 272: 13189-13195, 1997.62. Shen Y, Schneider G, Cloutier JF, Veillette A, and Schaller MD. Direct association of protein-tyrosine phosphatase PTP-PEST with paxillin. J Biol Chem 273: 6474-6481, 1998.63. Sieg DJ, Hauck CR, Ilic D, Klingbeil CK, Schaefer E, Damsky CH, and Schlaepfer DD. FAK integrates growth-factor and integrin signals to promote cell migration. Nat Cell Biol2: 249-256, 2000.64. Smith JW, Piotrowicz RS, and Mathis D. A mechanism for divalent cation regulation of beta 3-integrins. J Biol Chem 269: 960-967, 1994.65. Thamilselvan V and Basson MD. Pressure activates colon cancer cell adhesion by inside-out focal adhesion complex and actin cytoskeletal signaling. Gastroenterology 126: 8-18, 2004.

Page 27 of 45

28

66. Thamilselvan V, Fomby M, Walsh M, and Basson MD. Divalent cations modulate human colon cancer cell adhesion. J Surg Res 110: 255-265, 2003.67. Vadlamudi RK, Adam L, Nguyen D, Santos M, and Kumar R. Differential regulation of components of the focal adhesion complex by heregulin: role of phosphatase SHP-2. J Cell Physiol 190: 189-199, 2002.68. van Munster IP, Tangerman A, de Haan AF, and Nagengast FM. A new method for the determination of the cytotoxicity of bile acids and aqueous phase of stool: the effect of calcium. European Journal of Clinical Investigation 23: 773-777, 1993.69. Weisburger JH, Reddy BS, Barnes WS, and Wynder EL. Bile acids, but not neutral sterols, are tumor promoters in the colon in man and in rodents. Environ Health Perspect 50: 101-107, 1983.70. Werneburg NW, Yoon JH, Higuchi H, and Gores GJ. Bile acids activate EGF receptor via a TGF-alpha-dependent mechanism in human cholangiocyte cell lines. Am J Physiol Gastrointest Liver Physiol 285: G31-36, 2003.

Page 28 of 45

29

Figure Legends

Fig. 1. DCA inhibits integrin-mediated cell adhesion. 96 well plates coated with 10µg/ml collagen-I

(a), fibronectin (b), and laminin (c). Serum starved HCA-7 cells were treated with DCA (10-125µM, 1hr)

and detached with 2mM EDTA. Cells (5x104/well) were then plated on matrix-coated 96 well plates and

after 30 min adherent cells were stained with crystal violet. Values represent mean A550 of the extracted

dye from triplicate wells. Results are the means ±SD of three independent experiments. *p<0.05

compared with control. Plates were also coated with serially diluted (0.05-5µg/ml) collagen-I (d),

fibronectin (e), laminin (f). Control untreated (filled square) and DCA treated (100µM, 1hr, open

diamond) cells (5x104/well) were plated and cell adhesion was calculated as described above. Results

represent mean ±SD of two independent experiments in triplicate. Plates were also coated with 10µg/ml

poly-l-lysine (g). Control and DCA treated (100µM, 1hr) cells (5x104/well) were plated and cell

adhesion was calculated as described above. Results are the means ±SD of three independent experiments

in triplicate.

Fig. 2. DCA induces FAK dephosphorylation. (a) Serum starved HCA-7 cells were stimulated with

vehicle or DCA (100µM) for the indicated times and lysed. FAK was immunoprecipitated and blotted

sequentially for phosphotyrosine (pY, upper panel) and, after stripping, for FAK (lower panel). The

results shown are representative of three independent experiments. (b) Serum starved HCA-7 cells were

detached, maintained in suspension for 30 min and subsequently stimulated with DCA (100µM) for the

indicated times and lysed. FAK was immunoprecipitated and blotted sequentially for phosphotyrosine

(pY, upper panel) and, after stripping, for FAK (lower panel). The results shown are representative of

three independent experiments. (c) Serum starved HT-29 cells were stimulated with DCA (100µM) for

the indicated times and lysed. FAK was immunoprecipitated and blotted sequentially for phosphotyrosine

(pY, upper panel) and, after stripping, for FAK (lower panel). The results shown are representative of two

independent experiments. (d) Reduced FAK kinase activity towards an exogenous substrate, poly (E4Y1).

Page 29 of 45

30

Cells were stimulated with DCA (100µM) for 30 min, lysed and immunoprecipitated. FAK kinase

activity towards an exogenous substrate, poly (E4Y1) was assayed as described in methods. Values were

expressed as % maximal and given as Means ±SD of three independent experiments. *p<0.05 compared

with control.

Fig. 3. DCA differentially regulates tyrosine phosphorylation of selective residues on FAK. HCA-7

cells were serum starved and treated with DCA (100µM) for the indicated times and cell lysates were

analyzed by immunoblotting with site-specific phosphotyrosine antibodies against FAK. Blots were also

stripped and reprobed with antibodies, which recognize total FAK. The results shown are representative

of three independent experiments.

Fig. 4. FAK is constitutively associated with c-Src. (a) DCA activates c-Src in HCA-7 cells. Serum

starved cells were treated with DCA (100µM, 30 min). FAK was immunoprecipitated and probed with

antibody specific to active FAK (FAK Tyr-397) or Src (Src Tyr-418) or native Src. Blots were stripped

and reprobed for total FAK. (b) PP2 inhibits Src and FAK phosphorylation. Serum starved cells were pre-

treated with PP2 (30µM, 15min) and then stimulated with DCA as described above. FAK was

immunoprecipitated and probed with antibody specific to active FAK (FAK Tyr-397) or c-Src (Src Tyr-

418). Blots were stripped and reprobed for total FAK. The results shown are representative of three

independent experiments.

Fig. 5. DCA affects the status and localization of FAK to focal adhesions. HCA-7 cells were treated

with DCA (100µM, 30 min). Cells were fixed and FAK was analyzed by confocal microscopy after dual-

labeling immunofluorescence using FITC-conjugated phalloidin (green color), as a marker actin filaments

and rabbit pAb against FAK Tyr-925 and FAK Tyr-397 (red color). Yellow color indicates co-

localization of actin with FAK. All images were taken with identical laser and microscope settings (1µm

optical sections obtained at the cell-substratum interface). Note that in control serum starved cells FAK

Tyr-925 and Tyr-397 staining co-localized predominantly to actin-containing dots. Control cells were

Page 30 of 45

31

anchored to the substrate by mature focal adhesion points. There was a dramatic loss of staining intensity

corresponding to phosphorylated form of FAK Tyr-925 and cells displayed dynamic, immature dot-like

focal adhesion site remnants. However, FAK Tyr-397 staining was relatively unaffected.

Fig. 6. DCA stimulates tyrosine phosphatase activity. (a) Inhibition of PTPase activity diminishes

FAK dephosphorylation. HCA-7 cells were pre-incubated for 10 min with DMSO (vehicle), or 10µM

Phenylarsine oxide (PAO), a PTPase inhibitor. Cells were then treated with 100 µM DCA for indicated

time periods. FAK was immunoprecipitated and blotted sequentially for phosphotyrosine (pY, upper

panel) and, after stripping, for FAK (lower panel). The results shown are representative of three

independent experiments. (b) DCA induces tyrosine phosphorylation of ShP2 in HCA-7 cells. Cells were

treated with DCA (100µM) for the indicated times and cell lysates were immunoprecipitated with anti-

ShP1 and -ShP2 antibodies and Western blotted for phosphotyrosine (insert). DCA stimulates enzyme

activity of ShP2. Cells were treated with 100µM DCA for 30 min as described above. ShP1 and ShP2 was

immunoprecipitated from total lysates and analyzed for phosphatase activity as described in methods.

Phosphatase activity was expressed as the percentage of activity in the control untreated cells. Results are

the means ±SD of three independent experiments. *p<0.05 compared with control.

Fig. 7. DCA treatment increases association of ShP2 with FAK. HCA-7 cells were stimulated with

DCA (100µM) for the indicated periods of time. FAK was immunoprecipitated and probed for ShP2. The

results shown are representative of three independent experiments.

Fig. 8. ShP2 dephosphorylates FAK in vitro. (a) FAK was immunoprecipitated from serum starved

HCA-7 cells and incubated with different amounts of GST fusion protein of the ShP2 PTPase domain

(ShP2-PTPase) as described under methods. After the reaction the immune complex was washed and

subjected to immunoblotting analyses with anti Py-20 antibodies for tyrosine dephosphorylation of FAK

(upper panel). The blot was reprobed with anti-FAK antibody (lower panel). The graph represents the

average of two experiments. (b) FAK was immunoprecipitated from serum starved HCA-7 cells and

Page 31 of 45

32

incubated with GST-ShP2 (8U ShP2-PTPase) as described above. The immune complex was washed and

the analyzed by immunoblotting for c-src (upper panel) and total FAK (lower panel). The results shown

are representative of three independent experiments. U, units.

Fig. 9. Antisense ShP2 prevent tyrosine dephosphorylation of FAK. (a) Antisense (AS)

oligodeoxynucleotides directed against ShP2 decreases ShP2 protein expression. Cells were

treated with 1µM ShP2 antisense (AS), sense (SS) and nonsense (NS) oligodeoxynucleotides or medium

(Control, Ct) for 24 hours. ShP2 protein levels were measured by Western blotting (upper panel). Blots

were also re-probed with β-actin (lower panel). (b) Antisense (AS) oligodeoxynucleotides directed

against ShP2 prevent tyrosine dephosphorylation of FAK by DCA. Cells were transfected with

oligodeoxynucleotides as described above and treated with DCA (100µM, 60 min). Cell lysates were

immunoprecipitated with anti-FAK and probed with anti-phosphotyrosine antibodies (upper panel). Blots

were also re-probed with anti-FAK (lower panel). The results shown are representative of three

independent experiments.

Fig. 10. Mn+2 partially prevents the DCA effect on cell adhesion. Figure represents the adhesion of

DCA treated HCA-7 cells compared with respective controls with or without 1 mM Mn+2. Serum starved

HCA-7 cells were treated with DCA (100µM, 1hr) with or without 1mM Mn+2. After EDTA detachment,

cells (5x104/well) were plated on 96 well plates coated with 10µg/ml collagen-I and after 30 min cell

adhesion was estimated as described before. Results are the means ±SD of three independent experiments

in triplicate. *p<0.05 compared with respective controls. **p<0.05 compared with cells without Mn+2

treatment.

Figure 11. FAK suppression prevents the DCA effect on cell adhesion. Replication-deficient

adenoviruses (Adv) expressing GFP or GFP-tagged FRNK were propagated in HEK-293 cells and

purified by double CsCl centrifugation. Multiplicity of viral infection (MOI) was assayed by viral

dilution in HEK-293 cells grown on 96 well plates. Then HCA-7 cells (50-60% confluent) were infected

Page 32 of 45

33

(10-100 MOI, 24hr) with either Adv-GFP or Adv-FRNK. (a), Fluorescent microscopy of GFP and GFP-

FRNK expressing HCA-7 cells (100 MOI, 24hr) (b), Cell lysates were separated by SDS-PAGE and

probed with a polyclonal antibody to the C-terminal domain of FAK (which recognizes both FAK and

FRNK, upper panel) and a polyclonal antibody specific to active FAK (FAK Tyr-397, lower panel). Note

a significant reduction of total FAK expression and activation at 100 MOI. (c), Twenty-four hours after

infection (100 MOI) cells were serum starved, treated with DCA (100µM, 1hr) and detached with 2mM

EDTA. Cells (5x104/well) were then plated on collagen-I or fibronectin coated 96 well plates and after

30 min adherent cells were stained with crystal violet. Results are the means ±SD of two independent

experiments in triplicate *p<0.05 compared with GFP alone.

Figure 12. ShP2 down-regulation by siRNA abolishes the DCA effect on cell adhesion. Transfection

of siRNA decreases ShP2 protein expression. Cells were treated with 5nM siRNA or HiPerFect

transfection reagent alone for 24 hours. ShP2 protein levels were measured by Western blotting (Inset).

Blots were also re-probed with total FAK to confirm any nonspecific decrease of an unrelated gene. Cells

were transfected with ShP2 siRNA as described above and serum starved cells were treated with DCA

(100µM, 1hr). After EDTA detachment, cells (5x104/well) were plated on 96 well plates coated with

10µg/ml collagen-I and after 30 min cell adhesion was estimated as described before. Results are the

means ±SD of two independent experiments in triplicate. *p<0.05 compared with control.

Page 33 of 45

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

0 20 40 60 80 100 120 140DCA (µM)

Cell

Adhe

sion

(A55

0 )

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

0 20 40 60 80 100 120 140DCA (µM)

Cell

Adhe

sion

(A55

0 )

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

0 20 40 60 80 100 120 140DCA (µM)

Cell

Adhe

sion

(A55

0 )

0

0.1

0.2

0.3

0.4

0.5

0.01 0.1 1 10Collagen I (µg/ml)

Cell

Adhe

sion

(A55

0 )

0

0.1

0.2

0.3

0.4

0.5

0.6

0.01 0.1 1 10Fibronectin (µg/ml)

Cel

lAdh

esio

n(A

550 )

0

0.1

0.2

0.3

0.4

0.5

0.01 0.1 1 10Laminin (µg/ml)

Cel

lAdh

esio

n(A

550 )

(a) (b) (c)

(f)(e)(d)

*

* ***

**

* *

- + DCA

0

20

40

60

80

100

120

Poly-l-Lysine

Cel

lAdh

esio

n(%

Max

imal

)

Fig. 1

(g)

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

0 20 40 60 80 100 120 140DCA (µM)

Cell

Adhe

sion

(A55

0 )

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

0 20 40 60 80 100 120 140DCA (µM)

Cell

Adhe

sion

(A55

0 )

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

0 20 40 60 80 100 120 140DCA (µM)

Cell

Adhe

sion

(A55

0 )

0

0.1

0.2

0.3

0.4

0.5

0.01 0.1 1 10Collagen I (µg/ml)

Cell

Adhe

sion

(A55

0 )

0

0.1

0.2

0.3

0.4

0.5

0.6

0.01 0.1 1 10Fibronectin (µg/ml)

Cel

lAdh

esio

n(A

550 )

0

0.1

0.2

0.3

0.4

0.5

0.01 0.1 1 10Laminin (µg/ml)

Cel

lAdh

esio

n(A

550 )

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

0 20 40 60 80 100 120 140DCA (µM)

Cell

Adhe

sion

(A55

0 )

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

0 20 40 60 80 100 120 140DCA (µM)

Cell

Adhe

sion

(A55

0 )

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

0 20 40 60 80 100 120 140DCA (µM)

Cell

Adhe

sion

(A55

0 )

0

0.1

0.2

0.3

0.4

0.5

0.01 0.1 1 10Collagen I (µg/ml)

Cell

Adhe

sion

(A55

0 )

0

0.1

0.2

0.3

0.4

0.5

0.6

0.01 0.1 1 10Fibronectin (µg/ml)

Cel

lAdh

esio

n(A

550 )

0

0.1

0.2

0.3

0.4

0.5

0.01 0.1 1 10Laminin (µg/ml)

Cel

lAdh

esio

n(A

550 )

(a) (b) (c)

(f)(e)(d)

*

* ***

**

* *

- + DCA

0

20

40

60

80

100

120

Poly-l-Lysine

Cel

l Adh

esio

n (A

550)

Fig. 1

(g)

Page 34 of 45

020406080100120

Ct DCA

FAK

activ

atio

n(%

Max

imal

)

*

(a)

Ct 7.5 15 30 60

Fig. 2

Total FAK

pY - FAK

DCA (min)Ct 7.5 15 30 60

(b)

(d)(c)

Ct 7.5 15 30 60

pY - FAK

Total FAK

DCA (min)

*

020406080100120

Ct DCA

FAK

activ

atio

n(%

Max

imal

)

*020406080100120

Ct DCA

FAK

Act

ivat

ion

(%M

axim

al)

*

(a)

Ct 7.5 15 30 60 Ct 7.5 15 30 60

Fig. 2

Total FAK

pY - FAK

DCA (min)Ct 7.5 15 30 60

Total FAK

pY - FAK

DCA (min)DCA (min)Ct 7.5 15 30 60

(b)

(d)(c)

Ct 7.5 15 30 60

pY - FAK

Total FAK

DCA (min)Ct 7.5 15 30 60

pY - FAK

Total FAK

DCA (min)

*

Page 35 of 45

Fig. 3

Ct 7.5 15 30 60

FAK - 925

FAK - 576/577

FAK - 397

Total FAK

DCA (min)

Fig. 3

Ct 7.5 15 30 60

FAK - 925

FAK - 576/577

FAK - 397

Total FAK

DCA (min)DCA (min)Page 36 of 45

Ct Ct DCA DCA

(a)

(b)

Fig. 4

- - + PP2

Total FAK

FAK - 397Src - 418

Ct DCA DCA

Total FAK

FAK - 397

Src - 418

Total Src

DCA

Total Total

Page 37 of 45

FAK - 397

FAK - 397

FAK - 925

FAK - 925Control

DCA

Control

DCA

Fig. 5

Actin FAK Merge

.

Page 38 of 45

0

50

100

150

200

Ct DCA

Ph

osp

ha

tase

Acti

vit

y(%

of

Co

ntr

ol)

Ct 7.5 15 30

ShP1

DCA (min)

Fig. 6

(a)0 7.5 30 0 7.5 30

pY - FAK

Total FAK

- + + - + + DCA

Minutes

+ + + - - - PAO

(b)

0

50

100

150

200

250

300

350

400

Ct DCA

Ph

osp

ha

tase

Acti

vit

y(%

of

Co

ntr

ol)

Ct 7.5 15 30

ShP2

DCA (min)

*

0

50

100

150

200

Ct DCA

Ph

osp

ha

tase

Acti

vit

y(%

of

Co

ntr

ol)

Ct 7.5 15 30

ShP1

DCA (min)

Fig. 6

(a)0 7.5 30 0 7.5 30

pY - FAK

Total FAK

- + + - + + DCA

Minutes

+ + + - - - PAO

(b)

0

50

100

150

200

250

300

350

400

Ct DCA

Ph

osp

ha

tase

Acti

vit

y(%

of

Co

ntr

ol)

Ct 7.5 15 30

ShP2

DCA (min)

*

Page 39 of 45

Ct 2 7.5 15 30

ShP2

IgG

Fig.

DCA (min)Ct 2 7.5 15 30

ShP2

IgG

Fig 7

DCA (min)Page 40 of 45

020406080100120

0 0.5 1 2 4 8

ShP2 - PTPase (U)

Ph

osp

ho

tyro

sin

e(%

)PTPase (U)

Ct G 8 4 2 1 0.5

pY - FAK

Fig. 8

(a)

Total Src

Total FAK

- + GST - Shp2

020406080100120

0 0.5 1 2 4 8

Shp2 - PTPase (U)

Ph

osp

ho

tyro

sin

e(%

) Ct G 8 4 2 1 0.5

pY - FAK

Shp2

Total FAK

(b)

Page 41 of 45

Ct Ct SS SS NS NS AS AS

ShP2

β - Actin

(a)

(b)AS AS SS SS NS NS Ct Ct

Fig. 9

pY - FAK

Total FAK

- + - + - + - + DCA

Ct Ct SS SS NS NS AS AS

ShP2

β - Actin

(a)

(b)AS AS SS SS NS NS Ct Ct

pY - FAK

Total FAK

pY - FAK

Total FAK

- + - + - + - + DCA

Page 42 of 45

0

20

40

60

80

100

120

140

160

180

Ct DCA Mn Mn+DCA

Cel

lAdh

esio

n(%

Max

imal

)

*

**

Fig. 10

*

0

20

40

60

80

100

120

140

160

180

Ct DCA Mn Mn+DCA

Cel

lAdh

esio

n(%

Max

imal

)

*

**

Fig. 10

*

Page 43 of 45

0

20

40

60

80

100

120

GFP GFP+DCA FRNK FRNK+DCA

Cel

lAdh

esio

n(%

Max

imal)

Collagen I Fibronectin

** *

(c)

*

Adv-GFP

Adv-GFP-FRNK

(a) (b)

10 25 50 100 10 25 50 100

GFP GFP - FRNK

FAK

FRNK

MOI

FAK 397

Fig. 11 0

20

40

60

80

100

120

GFP GFP+DCA FRNK FRNK+DCA

Cel

lAdh

esio

n(%

Max

imal)

Collagen I Fibronectin

** *

(c)

*

0

20

40

60

80

100

120

GFP GFP+DCA FRNK FRNK+DCA

Cel

lAdh

esio

n(%

Max

imal)

Collagen I Fibronectin

** *

(c)

*

Adv-GFP

Adv-GFP-FRNK

(a)Adv-GFP

Adv-GFP-FRNK

(a) (b)

10 25 50 100 10 25 50 100

GFP GFP - FRNK

FAK

FRNK

MOI

FAK 397

Fig. 11

Page 44 of 45

0

20

40

60

80

100

120

140

160

Ct DCA Ct DCA

Cel

lAdh

esio

n(%

Max

imal

)Shp2

FAK

Ct Ct siRNA siRNA

Fig. 12- - + + siRNA

*

0

20

40

60

80

100

120

140

160

Ct DCA Ct DCA

Cel

lAdh

esio

n(%

Max

imal

)Shp2

FAK

Ct Ct siRNA siRNA

Fig. 12- - + + siRNA

0

20

40

60

80

100

120

140

160

Ct DCA Ct DCA

Cel

lAdh

esio

n(%

Max

imal

)Shp2

FAK

Ct Ct siRNA siRNA

Shp2

FAK

Ct Ct siRNA siRNA

Fig. 12- - + + siRNA

*

Page 45 of 45