farnesyl transferase inhibitor (r115777)–induced inhibition of...farnesyl transferase inhibitor...

Farnesyl Transferase Inhibitor (R115777)–Induced Inhibition of

STAT3(Tyr705) Phosphorylation in Human Pancreatic Cancer

Cell Lines Require Extracellular Signal-Regulated Kinases

Kolaparthi Venkatasubbarao,1Ahsan Choudary,

1and James W. Freeman

1,2,3

1Department of Medicine, Division of Medical Oncology and 2Department of Cellular and Structural Biology, University of Texas HealthCenter and 3Research and Development, Audie Murphy Veterans Administration Hospital, San Antonio, Texas

Abstract

In this study, we report that R115777, a nonpeptidomimeticfarnesyl transferase inhibitor, suppresses the growth ofhuman pancreatic adenocarcinoma cell lines and that thisgrowth inhibition is associated with modulation in thephosphorylation levels of signal transducers and activatorsof transcription 3 (STAT3) and extracellular signal-regulatedkinases (ERK). Treatment of cells with R115777 inhibited thetyrosine phosphorylation of STAT3(Tyr705), while increasingthe serine phosphorylation of STAT3(Ser727). We found thedifferential phosphorylation of STAT3 was due to anincreased and prolonged activation of ERKs. The biologicalsignificance of ERK-mediated inhibition of STAT3(Tyr705)

phosphorylation was further assessed by treating the cellswith an inhibitor (PD98059) of mitogen-activated proteinkinase kinase (MEK) or by transfecting the cells with a vectorthat expresses constitutively active MEK-1. Expression ofconstitutively active MEK-1 caused an increase of ERKactivity and inhibited STAT3(Tyr705) phosphorylation. Con-versely, inhibition of ERK activity by PD98059 reversed theR115777-induced inhibition of STAT3(Tyr705) phosphorylation.R115777 also caused the inhibition of the binding of STAT3 toits consensus binding element. An increase in the activationof ERKs either by overexpressing MEK-1 or treatment of cellswith R115777 caused an up-regulation in the levels of acyclin-dependent kinase (cdk) inhibitor, p21cip1/waf1. Theseobservations suggest that R115777-induced growth inhibitionis partly due to the prolonged activation of ERKs thatmediates an inhibition of STAT3(Tyr705) phosphorylation andan increase in the levels of p21cip1/waf1 in human pancreaticadenocarcinoma cell lines. (Cancer Res 2005; 65(7): 2861-71)

Introduction

Carcinoma of the human exocrine pancreas represents thefourth most common cause of cancer deaths in the United States.There has been little improvement in patient survival because the5-year survival rate for these patients is <5%, with a mediansurvival of <6 months (1). A number of genetic mutations occur inpancreatic cancer cells (2). The most predominant (>90%) geneticalterations are mutations in codon 12 of the K-RAS gene and thecell cycle regulator p16(Ink4) gene. More than 50% of pancreatictumors show inactivation of DPC4 (deleted in pancreatic cancerlocus 4/Smad4) due to homozygous deletions and intragenic

mutations and mutations of p53 (3). In addition, overexpression ofseveral tyrosine kinase receptors and their respective ligands leadsto enhanced mitogenesis (4). Together, these genetic modificationscontribute to a selective growth advantage over normal cells,suppression of apoptotic pathways, and facilitation of invasion andmetastasis of pancreatic cancer.Farnesyl transferase inhibitors represent a new class of agents

that were originally developed to specifically inhibit the activity ofoncogenic RAS, which is mutated in many cancer types and isassociated with poor prognoses (5–8). RAS proteins are guaninenucleotide binding proteins that play a major role in cancerdevelopment. RAS activates several downstream effectors afterstimulation by various growth factors and cytokines (9). RASundergoes several post-translational modifications that facilitateits attachment to the inner surface of the plasma membrane. Themost crucial modification is the covalent attachment of a farnesylisoprenoid lipid to a cysteine residue in the COOH-terminal of RASproteins that is catalyzed by the enzyme farnesyl transferase andhas become an important target for anticancer drug discovery.However, subsequent studies suggest that another closely relatedenzyme, geranyl geranyl transferase is also involved in RASprenylation when farnesylation is inhibited by farnesyl transferaseinhibitors (10). A large number of preclinical studies using variousfarnesyl transferase inhibitors have examined their spectrum ofefficacy in an effort to understand the biological mechanism oftheir antitumor activity. Although a large number of preclinicalstudies show that farnesyl transferase inhibitors have antitumoractivity, the mechanisms of action of these drugs seem complexand still remain largely unknown (11). Other studies show thatfarnesyl transferase inhibitors have antitumor activity regardless ofRAS mutation suggesting that these compounds may target otherpotential prenylated proteins involved in growth regulation (11).Signal transducers and activator of transcription (STAT) were

originally discovered as latent cytoplasmic transcription factorsthat mediate pivotal cellular responses to diverse cytokines andgrowth factors (12–14). STATs are Src homology 2 (SH2) domain–containing transcription factors that exist as latent cytoplasmicproteins until stimulated by growth factors or cytokines. Apartfrom the cytokine receptor gp130, initial studies led to theidentification of receptor-associated Janus kinase family of proteinsas mediators for the activation of STATs (12–14). Further studiesrevealed that growth factor receptor tyrosine kinases or cytoplas-mic tyrosine kinases such as Src also mediate activation of STATs.The STAT family consists of seven members designated as STAT1,STAT2, STAT3, STAT4, STAT5a, STAT5b, and STAT6. Followingligand binding and activation of cytokines or growth factorreceptors, STATs are phosphorylated at their tyrosine residues.The tyrosine phosphorylated STAT monomers dimerize andsubsequently translocate to the nucleus and bind to STAT-specific

Requests for reprints: James W. Freeman, Department of Medicine, Division ofMedical Oncology, University of Texas Health Science Center, San Antonio, TX 78229-3900. Phone: 210-567-5298; Fax: 210-567-6687; E-mail: [email protected].

I2005 American Association for Cancer Research.

www.aacrjournals.org 2861 Cancer Res 2005; 65: (7). April 1, 2005

Research Article

Research. on October 12, 2020. © 2005 American Association for Cancercancerres.aacrjournals.org Downloaded from

http://cancerres.aacrjournals.org/

DNA-response elements of target genes and induce geneexpression (13). Among the STAT family, only STAT3 ablationleads to embryonic lethality (15). The tyrosine phosphorylation ofSTAT3 occurs at residue 705. In addition to the phosphorylation attyrosine residues, STAT3 also undergoes phosphorylation at serineresidue 727. The role of STAT3 serine phosphorylation intranscriptional regulation is not well defined because it mayenhance or inhibit the ability of tyrosine phosphorylated STAT3 tobind to DNA (16).Besides its normal functions, emerging evidence strongly

suggests that aberrant constitutive activation of STAT3 can leadto cellular transformation and tumorigenesis (17, 18). Many cancercell lines and primary tumors are reported to possess constitu-tively activated STAT3 (18). Other investigations have shown thatinactivation of STAT3 leads to an inhibition of cell proliferationin various cancer cell lines (19–23), including pancreatic cancer(24–28). A recent study from our laboratory shows thatSTAT3(Tyr705) phosphorylation depends on ErbB2 tyrosine kinaseactivity and is partially responsible for the mediation of autocrinegrowth factor–independent phenotype in human pancreaticcancer cells (26). We further showed in that study that functionalinhibition of STAT3 signaling by expression of dominant-negativeSTAT3 vector reduced the growth of human pancreatic cancercells (26). Thus, due to its intrinsic oncogenic potential, STAT3may be an attractive target for therapeutic intervention.This led us to investigate the mechanism of farnesyl transferase

inhibitor–induced growth inhibition in pancreatic cancer cells and

whether any tumor-promoting signaling pathways apart from RAS

signaling are affected by this compound. In this study, we used the

nonpeptidomimetic farnesyl transferase inhibitor, R115777 that

was shown to have antitumor effects both in vivo and in vitro

(29, 30). Treatment of pancreatic cancer cells with R115777 caused

a dose-dependent growth inhibition due to G1 arrest. While

analyzing various oncogenic signaling pathways, we found that

R115777 caused a differential modulation of STAT3(Tyr705) and

STAT3(Ser 727) phosphorylation. Although previous studies have

been attempted to analyze the effects of R115777 on STAT3

phosphorylation, the underlying mechanism has not been

established. Whereas a recent phase I clinical study (31) suggests

no correlation between clinical response and variations in

STAT3(Tyr705) phosphorylation exists in myelodysplastic syndrome,

an in vitro study showed that higher concentrations of R115777

inhibited the ability of interleukin 6 (IL-6) to induce STAT3(Tyr705)

phosphorylation in human myeloma cells (32). The present study

shows that R115777 causes a differential modulation of STAT3

phosphorylation in pancreatic cancer cell lines. Treatment of cells

with R115777 inhibited STAT3(Tyr705) phosphorylation but caused

an increase in the phosphorylation of STAT3(Ser 727). We further

show that this differential modulation of STAT3(Tyr705) and

STAT3(Ser 727) phosphorylation is mediated by prolonged activation

of ERKs. The role of ERKs in this phenomenon was further verified by

either blocking or increasing ERK activity in these cells. Inhibition of

ERK activity reverses the R115777-induced inhibition of STAT3(Tyr 705)

activation. On the other hand, increasing the activation of ERKs

inhibited the phosphorylation of STAT3(Tyr 705). Treatment of cells

with R115777 also caused an increase in expression of the cyclin-

dependent kinase (cdk) inhibitor p21cip1/waf1. Thus, the growth

inhibitory effects of R115777 were associated with the inhibition of

STAT3(Tyr705)-mediated tumorigenic signaling pathway and an

increase in the levels of p21cip1/waf1 mediated by prolonged

activation of ERKs. This study suggests that although farnesyl

transferase inhibitors are developed to target RAS signaling

pathways, these compounds may lead to inhibition of other

important tumorigenic pathway(s) in pancreatic cancer.

Materials and Methods

Cell culture. The human pancreatic cancer cell lines MIA PaCa-2 andPANC-1 were purchased from American Type Culture Collection (Manassas,

VA) and the cell line UK Pan-1 was established in our laboratory (33). Cells

were grown in DMEM (Cell Grow-Mediatech, Herndon, VA) containing 10%

fetal bovine serum. The medium was supplemented with 2 mmol/L glu-tamine and 100 units/mL penicillin/streptomycin. Cell cultures were main-

tained under 5% CO2.

R115777 treatment and cell culture stimulation. The farnesyl tran-sferase inhibitor used in the study is R115777 from Janssen Pharmaceuticals

(Titusville, NJ). The compound was solubilized in DMSO and working dilu-

tions were made as described previously (29). Unless otherwise mentioned,

for all R115777 treatments conducted in this study, exponentially growingcells were plated in medium supplemented with 10% serum. Forty-eight

hours after plating the cells, serum-containing media was replaced with

serum-free medium. Cells were maintained under serum-free conditions for

48 hours. R115777 dissolved in DMSO was added at indicated concen-trations during the last 24 hours of serum starvation. For the untreated

control set of cells, only vehicle (DMSO) was added. The cells were

maintained under low serum conditions to arrest cell growth and for

partial synchronization. At the end of serum starvation, cells werestimulated to reenter the cell cycle with medium containing transferrin

(20 Ag/mL), insulin (4 Ag/mL), and epidermal growth factor (5 ng/mL),

referred to as growth factor medium (34) with or without freshlysupplemented R115777. Cells were harvested after 24 hours of growth

factor medium stimulation and used for various analyses. When mitogen-

activated protein kinase (MAPK) inhibitors were used, cells were plated,

serum starved as mentioned above, and the specific MAPK inhibitors wereadded at indicated concentrations 6 hours before the beginning of

pretreatment with R115777. At the end of serum starvation, cells were

stimulated with fresh growth factor medium with or without freshly added

R115777 and/or MAPK inhibitors. To analyze the effects of IL-6 on STAT3phosphorylation, cells were plated, serum starved as mentioned above,

treated with R115777, and IL-6 (10 ng/mL) was added 6 hours before

harvesting cells for Western blot analysis.Cell growth assays.The growth rate of cells was determined by standard

3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assays

as described previously (35). Briefly, exponentially growing cells (1 � 103)

were plated in 96-well dishes. After 48 hours of plating, cells were treatedwith indicated concentrations of R115777. MTT assays were done after 1, 3,

and 5 days of R115777 treatment. At the time of assay, cells were stained

with 0.5 mg/mL MTT (Sigma Chemical Co., St. Louis, MO) at 37jC for

2 hours and cells were solubilized in 200 AL of DMSO. Colorimetric deter-mination was done with a FLUOstar Optima plate reader (Durham, NC).

The data are represented as the mean value of eight wells per treatment

group and the experiments were repeated a minimum of three times.Mitogenesis assays. [3H]-Thymidine incorporation assay that deter-

mines the rate of DNA synthesis, which is reflective of proliferation of cells,

was done as described previously (36). Briefly, cells were plated and treated

with R115777 as described for MTT assays. After 2 days of treatment withR115777, cells were pulsed with [3H]-thymidine (1 ACi; 46 Ci/mmol;

Amersham Pharmacia Biotech, Inc., Piscataway, NJ) for 2 hours. DNA was

then precipitated with 10% trichloroacetic acid and the amount of [3H]-

thymidine incorporated was analyzed by liquid scintillation counting in aBeckman LS3801 Scintillation Counter. The data represented as the mean

value of quadruplicate wells per treatment group and the experiments were

repeated a minimum of two times.

Cell cycle analysis by flow cytometry.To determine the percentage of cellsat various phases of cell cycle, serum-starved cells were stimulatedwith growth

factor medium as described in cell culture stimulation section. Untreated

Cancer Research

Cancer Res 2005; 65: (7). April 1, 2005 2862 www.aacrjournals.org



control or R115777-treated cells (1� 106) cells were fixed in 70% ethanol for 30minutes at room temperature. Fixed cells were stainedwith PBSwith 50 Ag/mL

propidium iodide and 300 Ag/mL RNase A at 37jC. Flow cytometric analysis

was done in FACSCalibur (Becton Dickinson Immunocytometry Systems, Inc.,

San Jose, CA). Data collected from 20,000 cells for each experiment wasanalyzed using ModFit software (Verity Software House, Topsham, ME).

Western immunoblots. Western immunoblots were prepared to anal-

yze the protein levels of phosphorylated or total STAT3, ERKs, ErbB2,

human h-actin, MEK-1, c-Myc, and p21cip/waf1. Cells were harvested atindicated time points after treatment with R115777, and total cellular

proteins were extracted using radioimmunoprecipitation assay buffer (1�PBS containing 1% NP40, 0.5% sodium deoxycholate and EDTA-free

protease inhibitor cocktail, 2 mmol/L phenylmethylsulfonyl fluoride and2 mmol/L sodium orthovanadate). Fifty micrograms of protein lysates were

electrophoresed on an 8% SDS-polyacrylamide gel and then transferred to

Hybond-P, polyvinylidene difluoride membrane (Amersham PharmaciaBiotech, Inc.). The blots were probed with specific antibodies for the above

mentioned proteins and detected by enhanced chemiluminescence

methods (NEN Life Science Products, Boston, MA). Wherever total cellular

levels of STAT3, ERKs, MEK-1, and h-actin were to be detected, theexposure time was minimized to avoid saturation effects of these proteins

due to their abundant amount present in the cells.

STAT3-DNA binding studies. Preparation of nuclear extracts, EMSA,

and competitive analyses were carried out as described previously (37, 38).The wild-type and mutant EMSA oligonucleotide probes for the STAT3

consensus element were the same as reported previously (37, 38). The wild

type probe sequence is 5V-GATCTCCTTCCCGGAAGCA-3V. The bold faceletters represent the STAT3 binding sequence. The sequence of mutant

STAT3 SIE is as follows: 5V-GATCTCCAAGCTTGAAGCA-3V. The underlined

sequence represents the mutated STAT3 binding sequence. Densitometric

analysis was done for the autoradiograms to determine the variations in thebinding of STAT3 to its consensus element using AlphaImager software.

Transient transfection of mitogen-activated protein kinase kinase-1vector. To determine the biological role of increased expression of ERKs,

on the modulation of STAT3 phosphorylation, PANC-1 and UK Pan-1 cellswere transiently transfected with a vector that expresses constitutively

active MEK-1. Exponentially growing cells were transfected with 1 and 2 Agof the MEK-1 vector per mL of medium in a 6-cm dish using commerciallyavailable transfection reagent Fugene (Roche, Indianapolis, IN). As a

control, 2 Ag of an empty vector was transfected under identical conditions.

Forty-eight hours after transfection, whole cellular lysates were extracted

and subjected to Western blot analysis.

Results

R115777 causes a dose-dependent decrease of growth ofpancreatic cancer cell lines. Exponentially growing PANC-1, UKPan-1, and MIA PaCa-2 cells were treated with different concen-trations of R115777. The growth was measured by MTT assays after1, 3, and 5 days of treatment. As shown in Fig. 1A , R115777 caused adose-dependent growth inhibition of these cells. The highestconcentration used in this study (5 Amol/L) inhibited growth byabout 50% in the cell lines tested. Furthermore, [3H]-thymidineassays, which measure DNA synthesis, that reflect the rate ofproliferation of cells was determined in PANC-1 cells (Fig. 1B). Cellswere treated either with DMSO alone or with different concen-trations of R115777, followed by determination of [3H]-thymidineincorporation. The data shows that [3H]-thymidine incorporationwas inhibited in cells treated with R115777 (Fig. 1B) in a dose-dependent manner. We further examined the effect of R115777 oncell cycle progression in these cells. PANC-1 cells were growtharrested by serum deprivation and stimulated with a growth factorenriched medium (34) to reenter the cell cycle. Serum-deprived,control and R115777-treated cells were mostly in G1 phase, 79.6%and 84.9%, respectively. After 24 hours of stimulation of cells with

growth factor medium, control cells showed reentry into the cellcycle as indicated by a decrease of cells in G1-phase cells ( from 79.6to 57.9) and an increase in S-phase cells ( from 11.8 to 35.3; Fig. 1C).However, cells treated with R115777 and stimulated with growthfactor medium failed to show reentry into the cell cycle and weremostly maintained in G1 phase (86.1; Fig. 1C). Taken together, theseobservations indicate that R115777 causes a dose-dependentsuppression of pancreatic cancer cell growth.R115777 causes a differential modulation of STAT3(Tyr705)

and STAT3(Ser727) phosphorylation that is associated with anincrease in the activation of extracellular signal-regulatedkinases in pancreatic cancer cells. It has become increasinglyevident that activated STAT3(Tyr705) plays a role in tumorigenesis inmany types of cancer. Recent studies from our lab (26) and others(24, 25, 27, 28) implicate a growth-promoting role of activatedSTAT3(Tyr705) in human pancreatic cancer cells. While analyzingvarious oncogenic pathways in R115777-treated cells, we alsodetermined whether treatment of pancreatic cancer cells withR115777 could interfere with the activation of STAT3(Tyr705). Todetermine the levels of various proteins analyzed in this study,pancreatic cancer cells were serum starved and stimulated withgrowth factor medium in the presence or absence of R115777.Western immunoblot analysis revealed that treatment of pan-creatic cancer cells (MIA PaCa-2, PANC-1, and UK Pan-1) withR115777 inhibited STAT3(Tyr705) phosphorylation (Fig. 2A). It hasbeen reported that STAT3 also undergoes phosphorylation of itsserine residues, especially STAT3(Ser727), and it was thought to beinvolved in the transcriptional regulation of target genes (39). Weascertained whether R115777 treatment causes alterations toSTAT3(Ser727) phosphorylation. Western immunoblots revealed thatR115777 treatment caused an increase of STAT3(Ser727) phos-phorylation (Fig. 2A). This suggests that R115777 treatment maylead to differential phosphorylation of STAT3. While inhibitingSTAT3(Tyr705) phosphorylation, R115777 resulted in an increase ofSTAT3(Ser727) phosphorylation.ERKs are phosphorylated via several signal transduction path-

ways, including signaling through protein-tyrosine kinases or byactivated RAS through RAF/MEK-1 kinases. Because farnesyltransferase inhibitors affect functionality of the RAS signalingpathway, we sought to determine whether this pathway wasinhibited by R115777 in pancreatic cancer cells, and analyzed thelevels of ERK activation. Surprisingly, we found that the levels ofactivated ERKs were increased in the pancreatic cancer cells treatedwith R115777 (Fig. 2A). These findings suggest that R115777-mediated effects are not restricted to one specific cell line. Similarobservations were also seen in experiments conducted withexponentially growing cells (data not shown). The total cellularlevels of STAT3 and ERKs, however, were not altered by R115777treatment. Cells treated with R115777, however, showed accumula-tion of unfarnesylated RAS in the cell lines studied (data not shown).Next, we determined the dose-dependent effects of R115777 in

the activation of ERKs and STAT3(Tyr705). Cells were serumstarved and treated with indicated concentrations of R115777.Total cellular lysates were subjected to Western immunoblotanalysis. A representative Western blot analysis of PANC-1 cellstreated with R115777 is shown in Fig. 2B . Treatment withR115777 caused an increase of the phosphorylation of ERKs(Fig. 2B). As shown in Fig. 2B , the levels of phosphorylatedSTAT3(Tyr705) in R115777-treated cells were reduced in a dose-dependent manner. On the other hand, the phosphorylated levelsof STAT3(Ser727) increased in R115777-treated cells. Total cellular

Farnesyl Transferase Inhibitor Targets STAT3




levels of STAT3 and ERKs, however, remain unchanged (Fig. 2B).Treatment of cells with R115777 did not cause any changes in theactivated levels of a separate MAPK member, p38 (data notshown). Thus, treatment with R115777 specifically caused aprolonged and sustained activation of ERKs in pancreatic cancercells.To determine the time course of phosphorylation of ERKs and

STAT3(Tyr705), serum-deprived PANC-1 cells were stimulated withgrowth factor medium and harvested at various time points. Incontrol cells, accumulation of phosphorylated STAT3(Tyr705) oc-

curred from 12 hours after stimulation with growth factor mediumand these levels were maintained through 24 hours. However, inR115777-treated cells, the levels of phosphorylated STAT3(Tyr705)

were low when compared with untreated control cells (data notshown).R115777 prevents the ability of interleukin-6 to induce

phosphorylation of STAT3(Tyr705). IL-6 activates the Janus kinase/STAT3 pathway (14, 40). To determine whether IL-6 can abrogateR115777-mediated inhibition of STAT3(Tyr705) phosphorylation,cells were treated with R115777 and stimulated with growth

Figure 1. Farnesyl transferase inhibitor R115777 causes a dose-dependent decrease of pancreatic cancer cell growth by causing a G1 arrest. A, exponentially growingPANC-1, UK Pan-1, and MIA PaCa-2 cells were exposed to indicated concentrations of R115777 and MTT assays were performed as explained in Materialsand Methods to determine the growth of cells. Growth was measured 1, 3, and 5 days after exposure to R115777. Points, mean of three separate experiments; bars,FSE. B, rate of DNA synthesis was measured in PANC-1 cells untreated control or treated with R115777 by [3H]-thymidine incorporation as explained in Materialsand Methods. Columns, mean % [3H]-thymidine incorporated in untreated control cells of two separate experiments; bars, FSE. The increasing concentrationof R115777 caused a dose-dependent decrease of DNA synthesis. C, flow cytometric analysis of control and R115777-treated PANC-1 cells to assess the variousphases of the cell cycle. DNA histograms (top) and % of each phase of cell cycle (table).

Cancer Research




factor medium supplemented with IL-6 (10 ng/mL) as explainedunder cell culture stimulation section. As expected, IL-6

stimulation caused an increase in the phosphorylation of

STAT3(Tyr705) (Fig. 3A , lane 3). However, in R115777-treated cells,IL-6 did not cause appreciable increase in the phosphorylation of

STAT3(Tyr705) (Fig. 3A , lane 4). This suggests that the treatment of

cells with IL-6 did not completely abrogate the R115777-inducedinhibition of STAT3(Tyr705). Thus, R115777 can partially inhibit thephosphorylation of STAT3(Tyr705), even in the presence of a strongactivator of STAT3(Tyr705) phosphorylation.R115777-induced STAT3(Tyr705) inhibition and extracellular

signal-regulated kinase activation are not transient. Todetermine whether R115777-induced effects are transient andreversible, cells were serum starved and treated with R115777. Atthe end of treatment, cells were allowed to grow in the growthfactor medium without R115777 for an additional 72 hours.Western immunoblots revealed that R115777-induced inhibition ofSTAT3(Tyr705) phosphorylation is not reversed at 72 hours afterwithdrawal of treatment (Fig. 3B). The sustained activation ofERKs also was not reversed to basal levels due to treatmentwithdrawal suggesting that the effects caused by R115777 are nottransient and not reversible up to 72 hours after treatmentwithdrawal.Functional inhibition of extracellular signal-regulated

kinases by PD98059 reverses the R115777-induced inhibitionof STAT3(Tyr705) phosphorylation. Because we observed

a correlation between ERK activation and inhibition ofSTAT3(Tyr705) phosphorylation in R115777-treated cells, next wedetermined the biological significance of ERKs in this process. Toassess this phenomenon, we blocked the activation of ERKs byusing the chemical inhibitor of MEK, PD98059. Serum-starved cellswere treated with R115777 and stimulated with growth factormedium. Western immunoblots were analyzed for phosphorylatedSTAT3(Tyr705) and ERKs (Fig. 4A). As loading controls, total cellularlevels of STAT3 and ERKs were also measured. Treatment of cellswith PD98059 alone (Fig. 4A , lane 3) did not cause appreciablealteration in the phosphorylation levels of STAT3(Tyr705). However,PD98059 treatment partially reversed the R115777 inducedinhibition of STAT3(Tyr705) phosphorylation (Fig. 4A , lane 5)compared with R115777 treatment alone (Fig. 4A , lane 2). Asexpected, PD98059 treatment inhibited ERK activation (Fig. 4A ,lane 3). To determine the specificity of requirement of ERKs toinhibit STAT3(Tyr705), cells were also treated with the p38 MAPKinhibitor, SB20358. Treatment of cells with SB20358 alone (Fig. 4A ,lane 4) did not cause any changes in the levels of STAT3(Tyr705).Furthermore, treatment of cells with SB20358 in combination withR115777 (Fig. 4A , lane 6) did not prevent the R115777-inducedinhibition of STAT3(Tyr705) phosphorylation indicating that SB20358did not change the ability of R115777 to induce ERK activation(Fig. 4A , lane 6). Thus, this suggests that R115777-inducedinhibition of STAT3(Tyr705) phosphorylation is specifically mediatedby ERK activation, and not mediated by p38 MAPK.

Figure 2. Differential modulation ofSTAT3 phosphorylation by R115777 isassociated with prolonged activation ofERKs in pancreatic cancer cells. PANC-1,UK Pan-1, and MIA PaCa-2 cells wereserum starved and stimulated with growthfactor medium in the presence or absenceof R115777 as explained in Materialsand Methods. Total cellular lysates wereisolated from control and R115777-treatedcells and subjected to Western immunoblotanalysis to determine the levels ofphosphorylated STAT3 and ERKs. A,Western immunoblot showing thatR115777 caused an inhibition ofSTAT3(Tyr705) phosphorylation and anincrease of STAT3(Ser727) phosphorylationassociated with an increase of ERKactivation of pancreatic cancer cell lines,MIA PaCa-2, PANC-1 and UK Pan-1. As aloading control, total cellular levels ofSTAT and ERKs were measured. B,representative Western immunoblot ofphosphorylated and total levels of STAT3and ERKs. PANC-1 cells treated withindicated concentrations of R115777.Treatment with R115777 caused adose-dependent inhibition of STAT3(Tyr705)

phosphorylation, while increasing thephosphorylation of STAT3(Ser727)

associated with an increase of ERKactivation. The total cellular levels of bothSTAT3 and ERKs remain unchanged.





Functional activation of extracellular signal-regulatedkinases by mitogen-activated protein kinase kinase-1 over-expression inhibits STAT3(Tyr705) phosphorylation. We furtherdetermined whether increasing ERK signaling pathway wouldprevent STAT3(Tyr705) phosphorylation. PANC-1 and UK Pan-1 cellswere transiently transfected with an empty vector, Neo, or with avector that express constitutively activated MAPK kinase (MEK-1).The protein lysates obtained from these transiently transfectedcells were analyzed by Western immunoblotting for STAT3(Tyr705)

phosphorylation levels and activated ERKs and shown in Fig. 4Band C for PANC-1 and UK Pan-1 cells, respectively. As loadingcontrols, total cellular levels of STAT3 and ERKs were measured.Transient transfection of cells with MEK-1 expression vector, asexpected, showed an increase in the total cellular levels of MEK-1and also caused an increase of ERK activation (Fig. 4B and C) incomparison with cells transfected with an empty vector, Neo(Fig. 4B and C). With increasing amounts of MEK-1 vector, a dose-dependent decrease of STAT3(Tyr705) was also observed (Fig. 4Band C , lanes 2-3). This data further supports the notion that ERKactivation negatively regulates STAT3(Tyr705) phosphorylation.Similar to R115777-treated cells, MEK-1 transfection also increasedthe STAT(Ser727) activity. These observations suggest that ERKactivation induced by either MEK-1 expression or R115777 resultsin the differential modulation of STAT3(Tyr705) versus STAT3(Ser727)

phosphorylation levels.Binding of STAT3 to its consensus element is inhibited by

R115777 treatment in pancreatic cancer cells. Enhanced DNAbinding activity of activated STAT3 has been reported in severalcell systems (18, 41). Hence, we determined whether R115777causes functional inhibition of STAT3-DNA binding. Nuclearextracts were prepared from control and R115777-treated cellsand then analyzed by electrophoretic mobility shift assay.

The STAT3-DNA protein complex is indicated by an arrow(Fig. 5). The specificity of STAT3 was verified by competitionanalysis using either a nonspecific oligonucleotide probe (lane 2)or with an oligonucleotide probe with a mutation in STAT3binding site (lane 3). These probes did not affect the STAT3binding to its consensus element. However, competition with 25and 50 mol/L excess of unlabeled wild-type oligonucleotide probescaused a reduction in the intensity of the bands (lanes 4 and 5 ,respectively), suggesting the specificity of STAT3 binding. Treat-ment of cells with R115777 caused a dose-dependent decrease ofSTAT3 binding to its consensus element (lanes 7-10) comparedwith nuclear extracts from untreated control cells (lane 6).Treatment of cells with the ERK inhibitor PD98059 did not affectthe STAT3-DNA binding ability (lane 11). However, inhibition ofERK activity by PD98059 partially reversed R115777-inducedinhibition of STAT3-DNA binding (lane 12). Similar to PANC-1cells, UK Pan, and MIA PaCa-2 cells showed a diminished bindingof STAT3 in R115777-treated cells (lanes 13-16).Activation of extracellular signal-regulated kinases by either

expression of mitogen-activated protein kinase kinase-1 ortreatment of cells with R115777 up-regulated the cyclin-dependent kinase inhibitor p21cip1/waf1. The cdk inhibitorp21cip1/waf1 plays a crucial role in several physiologic functions in-cluding controlling G1-S progression, differentiation, apoptosis, senes-cence, and affects DNA replication (42). ERK activation was previouslyreported to be involved in the up-regulation of p21cip1/waf1 (43, 44).In this contextwe sought to determine the levels of p21cip1/waf1, sinceR115777 caused a prolonged activation of ERKs. Westernimmunoblots were analyzed for p21cip1/waf1 levels (Fig. 6A and B).Transient transfection of cells with MEK-1 also caused a similarincrease of p21cip1/waf1 suggesting that prolonged ERK activationmay lead to an increase of the cdk inhibitor p21cip1/waf1 (Fig. 6A).

Figure 3. R115777 prevents the ability ofIL-6 to phosphorylate STAT3(Tyr705) inpancreatic cancer cells. A, untreated orR115777-treated cells were stimulated withIL-6 (10 ng/mL) for 6 hours before harvestingthe lysates. IL-6 caused an increase ofSTAT3(Tyr705) phosphorylation in untreatedcontrol cells (lane 1 versus lane 3).However, the presence of R115777prevented the IL-6-induced phosphorylationof STAT3(Tyr705) (lane 3 versus lane 4).Total cellular levels of both STAT3 andERKs were used as loading controls. B,R115777-induced modulations of STAT3and ERK activation are not transient inpancreatic cancer cells. Cells were treatedwith R115777 and stimulated with growthfactor medium for 72 hours as explained inMaterials and Methods. Cellular lysateswere isolated and analyzed by Westernimmunoblots which show that R115777treatment withdrawal does not reverse theR115777-induced inhibition of STAT3(Tyr705)

phosphorylation or ERK activation.

Cancer Research




Similarly, treatment of cells with R115777 caused a dose-dependentincrease of p21cip1/waf1 levels. The induction of p21cip1/waf1 may also beresponsible for G1 arrest of cells, suggesting that R115777 treatmentmay lead to alterations of multiple signaling pathways to suppressthe growth of cancer cells.R115777 causes inhibition of c-Myc, a transcriptional target

of STAT3. It has been shown earlier that c-Myc is atranscriptional target of STAT3. We sought to determine whethertreatment of pancreatic cancer cells with R115777 leads to aninhibition of c-Myc expression. Western blot analysis revealedthat treatment of cells with R115777 causes a decrease of c-Mycexpression (Fig. 7).

Discussion

Farnesyl transferase inhibitors were originally developed totarget tumors with mutated RAS genes. However, a majority of

tumor cell lines were found sensitive to farnesyl transferaseinhibitor–induced growth inhibition irrespective of the presenceof RAS mutation. Hence, it is not clear, whether farnesyltransferase inhibitor–induced growth inhibition was only due toinhibition of RAS (11). It is possible that the mechanismunderlying the farnesyl transferase inhibitor–induced growthinhibition may be due to the inhibition of prenylated proteins,besides RAS family of proteins, which may also be involved intumor promoting pathways. In this study, we show that a farnesyltransferase inhibitor, R115777 caused inhibition of pancreaticcancer cell growth apparently due to a cell cycle block in G1.Furthermore, we show this growth inhibition is associated with adecrease in phosphorylation of STAT3(Tyr705). R115777 treatmentalso led to a reduction of DNA binding of STAT3 to its consensuselement. Emerging evidence suggests that activated STAT3signaling is involved in tumorigenesis of several cancer types,

Figure 4. Down-modulation of STAT3(Tyr705) phosphorylation requires ERK activation. A, selective inhibition of ERK activity in PANC-1, UK Pan-1 and MIA PaCa-2cells reverses the R115777-induced inhibition of STAT3(Tyr705) phosphorylation. Cells were serum-starved and pretreated with R115777 as described in Materialsand Methods. For cell receiving combination treatments, the MAPK inhibitors were added to the medium 6 hours before the beginning of R115777 pretreatment. At theend of serum starvation and pretreatment, cells were stimulated with growth factor medium with or without R115777 and freshly supplemented MAPK inhibitors for24 hours. Total cellular lysates isolated were subjected to Western immunoblot analysis. Treatment of cells with MEK inhibitor PD98059 reversed the R115777-inducedinhibition of STAT3(Tyr705) phosphorylation. However, treatment with p38 MAPK inhibitor did not cause any alteration of R115777-induced down modulation ofSTAT3(Tyr705) phosphorylation. Treatment of cells with either of MAPK inhibitors alone did not alter the levels of STAT3(Tyr705) phosphorylation. As loading controls, totalcellular levels of both STAT3 and ERKs were used. Transient expression of constitutively active MAPK kinase (MEK-1) vector caused differential phosphorylation ofSTAT3(Tyr705) and STAT3(Ser727) in PANC-1 (B) and UK Pan-1 (C) cells. Exponentially growing PANC-1 and UK Pan-1 cells were transiently transfected with indicatedamounts of an empty vector, Neo, or a vector that expresses constitutively active MEK-1. Total cellular lysates were extracted after 48 hours of transfection and subjectedto Western immunoblot analysis. As expected, MEK-1 caused a dose-dependent increase of ERK activation associated with an inhibition of STAT3(Tyr705) phosphorylation.Constitutive MEK-1 expression, however, caused an increase of STAT3(Ser727) phosphorylation. Expression of the vectors, Neo or MEK-1, did not cause any alterationin the total cellular levels of STAT3 or ERKs. Expression of MEK-1 vector, however, caused an increase of the total cellular levels of MEK-1 as expected.





and inhibition of STAT3 signaling reverses tumor-promotingability (20–23). Activated STAT3 is also reported to play a role inpancreatic carcinogenesis (24–28).We found that the inhibition of STAT3(Tyr705) is causally linked to

a prolonged activation of ERKs. Initially, we speculated thatR115777 treatment would lead to a reduction of ERK activation,because many reports show that treatment with farnesyltransferase inhibitor causes a decrease in ERK signaling (45–47).Contrary to our expectation, we observed an increase of ERKactivation upon treatment of pancreatic cancer cells with R115777.IL-6 activates STAT3(Tyr705) in several cell systems (13, 14), and wetested whether IL-6 stimulation causes abrogation of R115777-induced hypophosphorylation of STAT3(Tyr705). Our observationsshow that IL-6 induced STAT3(Tyr705) phosphorylation wasdiminished in the presence of R115777. This is in agreement withan earlier study that shows higher concentrations of R115777caused a reduction of STAT3(Tyr705) phosphorylation in IL-6induced human myeloma cells (32). Furthermore, the R115777-induced modulations of ERKs and STAT3 phosphorylation werenot transient as these effects were still persistent after 72 hours oftreatment withdrawal. It is not clear at this time how the ERKs are

activated in response to R115777 treatment. Further studies areunder way to determine the specific pathways leading to theactivation of ERKs mediated by R115777. To our knowledge, thisstudy is the first to show that an inhibitor of farnesylation leads toactivation of ERKs. The findings of this study contradict the classicview that activated ERK signaling is involved in growth promotion.It is probable that prolonged activation of ERK, caused by R115777treatment as seen in this study, may have a detrimental effect ongrowth. This suggestion is in agreement with a recent study (48)that showed high intensity ERK signaling leads to growth arrest bymaintaining the hypophosphorylated forms of Rb protein inhuman hepatocellular carcinoma cell lines. Prolonged activation ofERKs is also reported to induce cell death mediated by syntheticvitamin K analogues in pancreatic cancer cells (49, 50). Activationof ERKs by treatment of cells with chemotherapeutic drugsinduced apoptosis of several cancer cells (51–53). Sustainedactivation of ERKs was also shown to play a role in oxidativestress-induced apoptosis of hydrogen peroxide–treated mousefibroblast cells (54). Thus, it is probable that sustained activation ofERKs can result in very different cellular effects than the effectscaused by transient induction upon growth factor stimulation.

Figure 5. DNA binding ability of STAT3 is inhibited in R115777-treated pancreatic cancer cells. Nuclear extracts isolated from cells either untreated or treated withR115777. EMSA performed to determine the DNA binding ability of STAT3 to its consensus element as explained in Materials and Methods. Representativeelectrophoretic mobility shift assay which indicates a dose-dependent decrease of STAT3 binding to its consensus element in R115777 treated cells (lanes 6-10).STAT3 nuclear protein complex (arrow). Incubation of nuclear extracts with radio-labeled STAT3 consensus element probe in the presence of 25 and 50 mol/L excess ofunlabeled wild-type (W) probe diminished the intensity of STAT3 protein-DNA complex (lanes 4 and 5). However, competition with a mutant STAT3 probe (M, lane 3)or a nonspecific probe (N, lane 2) did not affect the STAT3 protein-DNA complex, indicating the specificity of the STAT3-DNA complex. Nuclear extracts from cellstreated with MEK inhibitor PD98059 alone did not cause any alterations in the DNA binding ability of STAT3 protein (lane 11). However, treatment of cells withboth PD98059 and R115777 showed an increase in the binding of STAT3 to its consensus element (lane 12) suggesting a partial reversal of R115777-induced inhibitionof STAT3-DNA binding. R115777-treated cells also showed a decreased binding of STAT3 to its consensus element in UK Pan-1 (lane 13 versus lane 14) andMIA PaCa-2 cells (lane 15 versus lane 16). Inset, densitometric values of autoradiogram showing variations in the binding of STAT3 to its consensus element(lanes 6-12).

Cancer Research




The increase of ERK activation in cells treated with R115777caused a differential modulation of STAT3(Tyr705) and STAT3(Ser747)

phosphorylation. R115777 treatment while inhibiting STAT3(Tyr705)

phosphorylation caused an increase of STAT3(Ser727) phosphoryla-tion. The biological significance of ERK-mediated modulation ofSTAT3(Tyr705) phosphorylation was further verified in this study byusing synthetic compound PD98059 that inhibits MEK kinase andthus, ERK activity. Pretreatment of cells with PD98059 partiallyreversed the R115777-induced down-modulation of STAT3(Tyr705)

activity. However, inhibition of p38-MAPK signaling by SB20358did not alter the R115777-induced hypophosphorylation of

STAT3(Tyr705). Similarly, increasing the activity of ERKs byexpressing MEK-1 vector resulted in an inhibition of STAT3(Tyr705)

while increasing the STAT3(Ser727) phosphorylation. Thus, theseobservations show the role of ERKs in differential modulation ofSTAT3(Tyr705) and STAT3(Ser727) phosphorylation in R115777-treatedcells. Furthermore, our findings are supported by other studies thatsuggest ERKs are involved in the down-modulation of STAT3(Tyr705)

phosphorylation. Chung et al. (55) reported that ERK activationdecrease the rate of STAT3(Tyr705) phosphorylation. Their studiesfurther suggested that phosphorylation of STAT3(Ser727) may eitherinhibit phosphorylation of STAT3(Tyr705) or it may increase tyrosinedephosphorylation. Other in vitro studies show that ERKsincreased the phosphorylation of STAT3(Ser727) (55–57). Jain et al.(58) reported that binding of ERKs to STAT3 negatively regulates thephosphorylation of STAT3(Tyr705). Differentiation inducing factor-1also reported to inhibit the phosphorylation of STAT3(Tyr705) thatwas mediated by ERKs in gastric cancer cells (59). Collectively,these studies show a direct role of ERKs in the down-modulationof STAT3(Tyr705) phosphorylation.However, the in vivo role of STAT3(Ser727) is not clearly defined

because it can enhance or inhibit STAT3(Tyr705) phosphorylation,DNA binding, and transcriptional activity (16). For example,studies showed that STAT3(Ser727) activation affects the homo-dimer formation of STAT3 and inhibition of STAT3(Ser727) preventsthe DNA binding (60). Wen et al. (61) showed that STAT3(Ser727)

play a role in transcriptional activation. However, other studies(55, 56) show the negative effect of STAT3(Ser727) on STAT3(Tyr705)

phosphorylation and DNA binding. Thus, these studies suggestthat the role of STAT3(Ser727) is cell type specific. A recent reportsuggests that STATs do not require phosphorylation to formhomodimers (62). Thus, it is probable that differential phosphor-ylation of STAT3(Tyr705) versus STAT3(Ser727) mediated by ERKs, asobserved in this study, may lead to activation of a different set ofSTAT3 target genes. This differential trans-activation of targetgenes by STAT3 under different physiologic conditions in concertwith other transcriptional and cofactors may be the underlyingmechanism that either promotes or suppresses the growth ofcancer cells.Furthermore, we show that R115777 treatment caused an

increase of the cdk inhibitor p21cip1/waf1 in a dose-dependentmanner. p21cip1/waf1 plays a crucial role in controlling G1-Sprogression and also affects DNA replication (42). The inductionof p21cip1/waf1 occurs at the transcriptional and post-translationallevels in a p53-dependent (42, 63), or p53-independent (64),or transforming growth factor-h/Smad–dependent mechanisms(65, 66). ERK activation was previously reported to be involved inthe up-regulation of the cdk inhibitor p21cip1/waf1 (43, 44, 67,68). Several studies have suggested that the up-regulation ofp21cip1/waf1 in response to farnesyl transferase inhibitors occurred ina p53-dependent manner in cells with wild-type p53 (69, 70).

Figure 6. Overexpression of MEK-1 or R115777 treatment causes anup-regulation of the cdk inhibitor p21cip1/waf1 in human pancreatic cancer cells.A, representative Western immunoblot depicting a dose-dependent increase ofexpression of the cdk inhibitor p21cip1/waf1 in PANC-1 and UK Pan-1 cellstransiently transfected with a vector that expresses constitutively active MEK-1as explained in Materials and Methods and cellular lysates were isolated andsubjected to Western blot analysis. Human h-actin (loading control) levels. B,pancreatic cancer cells treated with indicated concentrations of R115777 causeda dose-dependent upregulation of the cdk inhibitor p21cip1/waf1. Cells weretreated or left untreated as explained in Materials and Methods and cellularlysates were subjected to Western blot analysis. Human h-actin (loading control)levels.

Figure 7. R115777 treatment causes a decrease in theexpression of c-Myc, a downstream transcriptional target ofSTAT3 in human pancreatic cancer cells. Cellswere untreated or treated with R115777 as explained inMaterials and Methods and cellular lysates were isolatedand subjected to Western blot analysis to measurethe levels of c-Myc, a downstream transcriptional target ofSTAT3. Treatment of cells with R115777 caused adecrease of the expression of c-Myc. h-Actin(loading control) levels.





However, the cell lines used in our study harbors a mutant p53gene. Thus, R115777-induced p21cip1/waf1 levels are independent ofp53 in these cells. It has been shown that oncogenic STAT3activation negatively regulates p21cip1/waf1 in cancer cells asopposed to normal cells (71). ERKs are also known to activatep21cip1/waf1 in some cell types (43, 44). It is of interest to note thatinhibition of STAT3 upstream activator Janus kinase 2 bytyrphostin also resulted in growth arrest by a delayed G1-S phaseprogression and an increase in the expression of p21cip1/waf1 inpancreatic tumor cells (25). Taken together, our study suggests thatinduction of p21cip1/waf1 that occurred in concert with inhibition ofSTAT3(Tyr705) and activation of ERKs , play a role in mediating theR115777-induced growth inhibition. Further studies are under wayto determine the mechanism of up-regulation of p21cip1/waf1 andwhether STAT3(Ser727) plays a role in transcriptional activation ofp21cip1/waf1 in pancreatic cancer cells. Furthermore, we showthat the levels of c-Myc, a well-known transcriptional downstreamtarget of STAT3 were found to be decreased as a consequence ofthe inhibition of STAT3(Tyr705) phosphorylation in R115777-treatedpancreatic cancer cells. It is of interest in this context to note thata recent report suggests that STAT3 can negatively regulatep21cip1/waf1 expression whereas increasing the levels of myc incancer cells (71). It is probable that treatment of pancreatic cancercells with R115777 caused a reversal of this phenomenonby increasing the levels of p21cip1/waf1, while reducing the levelsof c-Myc in pancreatic cancer cells.

In summary, this study suggests that inhibiting farnesylationactivities blocks phosphorylation of STAT3(Tyr705), an importanttumor promoting pathway in pancreatic cancer cells. Thisinhibition of STAT3(Tyr705) phosphorylation by R115777 is mediatedby prolonged ERK activation. The prolonged ERK activity alsoresulted in an increase in the levels of cdk inhibitor p21cip1/waf1 thatplay a role in G1-S cell cycle progression. Surmounting evidenceindicate that targeting STAT3 is of therapeutic importance (72, 73).Thus, this study identifies a novel target for farnesyl transferaseinhibitor–induced growth inhibition of pancreatic cancer cells.Understanding the biochemical basis for the growth inhibitoryactions of farnesyl transferase inhibitors will lead to a rationalclinical design for efficient therapy of several types of cancers. Alsoit may be possible that inhibition of oncogenic STAT3 signalingmay also increase the tumoral response to conventional chemo-therapy and radiotherapy.

Acknowledgments

Received 7/6/2004; revised 12/8/2004; accepted 1/17/2005.Grant support: San Antonio Area Foundation’s Beta and Melvin Leazar Memorial

fund (K. Venkatasubbarao).The costs of publication of this article were defrayed in part by the payment of page

charges. This article must therefore be hereby marked advertisement in accordancewith 18 U.S.C. Section 1734 solely to indicate this fact.

We thank Dr. David End (Johnson and Johnson Pharmaceutical Research andDevelopment, Spring House, PA) for providing the compound R115777 used in thisstudy and for critical reading of this article and Dr. Kun-Liang Guan (University ofMichigan, Ann Arbor, MI) for kindly providing us the MEK-1 expression vector.

References1. Bardeesy N, DePinho RA. Pancreatic cancer biologyand genetics. Nat Rev Cancer 2002;2:897–909.

2. Hansel DE, Kern SE, Hruban RH. Molecular patho-genesis of pancreatic cancer. Annu Rev Genomics HumGenet 2003;4:237–56.

3. Goggins M, Kern SE, Offerhaus JA, Hruban RH.Progress in cancer genetics: lessons from pancreaticcancer. Ann Oncol 1999;10 Suppl 4:4–8.

4. Freeman JW, Dearmond D, Lake M, et al. Alterations ofcell signaling pathways in pancreatic cancer. FrontBiosci 2004;9:1889–98.

5. Rowinsky EK, Windle JJ, Von Hoff DD. Rasprotein farnesyltransferase: a strategic target foranticancer therapeutic development. J Clin Oncol1999;17:3631–52.

6. Prendergast GC, Rane N. Farnesyltransferase inhib-itors: mechanism and applications. Expert Opin Inves-tig Drugs 2001;10:2105–16.

7. Brunner TB, Hahn SM, Gupta AK, Muschel RJ,McKenna WG, Bernhard EJ. Farnesyltransferase inhib-itors: an overview of the results of preclinical andclinical investigations. Cancer Res 2003;63:5656–68.

8. Downward J. Targeting RAS signalling pathways incancer therapy. Nat Rev Cancer 2003;3:11–22.

9. Malumbres M, Barbacid M. RAS oncogenes: the first 30years. Nat Rev Cancer 2003;3:459–65.

10. Whyte DB, Kirschmeier P, Hockenberry TN, et al.K- and N-ras are geranylgeranylated in cells treatedwith farnesyl protein transferase inhibitors. J BiolChem 1997;272:14459–64.

11. Sebti SM, Der CJ. Opinion: searching for the elusivetargets of farnesyltransferase inhibitors. Nat Rev Cancer2003;3:945–51.

12. Darnell JE Jr. STATs and gene regulation. Science1997;277:1630–5.

13. Levy DE, Darnell JE Jr. Stats: transcriptional controland biological impact. Nat Rev Mol Cell Biol2002;3:651–62.

14. Aaronson DS, Horvath CM. A road map for thosewho don’t know JAK-STAT. Science 2002;296:1653–5.

15. Takeda K, Noguchi K, Shi W, et al. Targeteddisruption of the mouse Stat3 gene leads to earlyembryonic lethality. Proc Natl Acad Sci U S A 1997;94:3801–4.

16. Decker T, Kovarik P. Serine phosphorylation ofSTATs. Oncogene 2000;19:2628–37.

17. Bowman T, Yu H, Sebti S, Dalton W, Jove R. Signaltransducers and activators of transcription: noveltargets for anticancer therapeutics. Cancer Control1999;6:427–35.

18. Turkson J, Jove R. STAT proteins: novel moleculartargets for cancer drug discovery. Oncogene 2000;19:6613–26.

19. Bromberg JF, Horvath CM, Besser D, Lathem WW,Darnell JE Jr. Stat3 activation is required for cell-ular transformation by v-src. Mol Cell Biol 1998;18:2553–8.

20. Bromberg JF, Wrzeszczynska MH, Devgan G, et al.Stat3 as an oncogene. Cell 1999;98:295–303.

21. Turkson J, Bowman T, Garcia R, Caldenhoven E,De Groot RP, Jove R. Stat3 activation by Src inducesspecific gene regulation and is required for celltransformation. Mol Cell Biol 1998;18:2545–52.

22. Ni Z, Lou W, Leman ES, Gao AC. Inhibition ofconstitutively activated Stat3 signaling pathway sup-presses growth of prostate cancer cells. Cancer Res2000;60:1225–8.

23. Niu G, Heller R, Catlett-Falcone R, et al. Gene therapywith dominant-negative Stat3 suppresses growth of themurine melanoma B16 tumor in vivo . Cancer Res 1999;59:5059–63.

24. Greten FR, Weber CK, Greten TF, et al. Stat3 and NF-nBactivation prevents apoptosis in pancreatic carcinogen-esis. Gastroenterology 2002;123:2052–63.

25. Scholz A, Heinze S, Detjen KM, et al. Activated signaltransducer and activator of transcription 3 (STAT3)supports the malignant phenotype of human pancreaticcancer. Gastroenterology 2003;125:891–905.

26. DeArmond D, Brattain MG, Jessup JM, et al.Autocrine-mediated ErbB-2 kinase activation of STAT3is required for growth factor independence of pancre-atic cancer cell lines. Oncogene 2003;22:7781–95.

27. Toyonaga T, Nakano K, Nagano M, et al. Blockade ofconstitutively activated Janus kinase/signal transducerand activator of transcription-3 pathway inhibitsgrowth of human pancreatic cancer. Cancer Lett 2003;201:107–16.

28. Wei D, Le X, Zheng L, et al. Stat3 activation regulatesthe expression of vascular endothelial growth factorand human pancreatic cancer angiogenesis and metas-tasis. Oncogene 2003;22:319–29.

29. End DW, Smets G, Todd AV, et al. Characterization ofthe antitumor effects of the selective farnesyl proteintransferase inhibitor R115777 in vivo and in vitro .Cancer Res 2001;61:131–7.

30. Venet M, End D, Angibaud P. Farnesyl proteintransferase inhibitor ZARNESTRA R115777-history ofa discovery. Curr Top Med Chem 2003;3:1095–102.

31. Kurzrock R, Kantarjian HM, Cortes JE, et al.Farnesyltransferase inhibitor R115777 in myelodysplas-tic syndrome: clinical and biologic activities in thephase 1 setting. Blood 2003;102:4527–34.

32. Le Gouill S, Pellat-Deceunynck C, Harousseau JL,et al. Farnesyl transferase inhibitor R115777 inducesapoptosis of human myeloma cells. Leukemia2002;16:1664–7.

33. Fralix KD, Ahmed MM, Mattingly C, et al.Characterization of a newly established humanpancreatic carcinoma cell line, UK Pan-1. Cancer2000;88:2010–21.

34. Mulder KM, Brattain MG. Continuous maintenanceof transformed fibroblasts under reduced serum con-ditions: utility as a model system for investigatinggrowth factor-specific effects in nonquiescent cells.J Cell Physiol 1989;138:450–8.

35. Venkatasubbarao K, Ammanamanchi S, Brattain MG,Mimari D, Freeman JW. Reversion of transcriptionalrepression of Sp1 by 5 aza-2V deoxycytidine restoresTGF-h type II receptor expression in the pancreaticcancer cell line MIA PaCa-2. Cancer Res 2001;61:6239–47.

36. Sun L, Wu G, Willson JK, et al. Expression oftransforming growth factor h type II receptor leads toreduced malignancy in human breast cancer MCF-7cells. J Biol Chem 1994;269:26449–55.

Cancer Research




37. Chin YE, Kitagawa M, Su WC, You ZH, Iwamoto Y,Fu XY. Cell growth arrest and induction of cyclin-dependent kinase inhibitor p21 WAF1/CIP1 mediatedby STAT1. Science 1996;272:719–22.

38. Bellido T, O’Brien CA, Roberson PK, Manolagas SC.Transcriptional activation of the p21(WAF1,CIP1,SDI1)gene by interleukin-6 type cytokines. A prerequisitefor their pro-differentiating and anti-apoptotic effectson human osteoblastic cells. J Biol Chem 1998;273:21137–44.

39. Turkson J, Bowman T, Adnane J, et al. Requirementfor Ras/Rac1-mediated p38 and c-Jun N-terminal kinasesignaling in Stat3 transcriptional activity induced by theSrc oncoprotein. Mol Cell Biol 1999;19:7519–28.

40. Zhong Z, Wen Z, Darnell JE Jr. Stat3: a STAT familymember activated by tyrosine phosphorylation inresponse to epidermal growth factor and interleukin-6.Science 1994;264:95–8.

41. Yu CL, Meyer DJ, Campbell GS, et al. Enhanced DNA-binding activity of a Stat3-related protein in cellstransformed by the Src oncoprotein. Science 1995;269:81–3.

42. Sherr CJ. Cancer cell cycles. Science 1996;274:1672–7.43. Mulder KM, Morris SL. Activation of p21ras bytransforming growth factor h in epithelial cells. J BiolChem 1992;267:5029–31.

44. Hu PP, Shen X, Huang D, Liu Y, Counter C, Wang XF.The MEK pathway is required for stimulation ofp21(WAF1/CIP1) by transforming growth factor-h.J Biol Chem 1999;274:35381–7.

45. Reuveni H, Geiger T, Geiger B, Levitzki A. Reversal ofthe Ras-induced transformed phenotype by HR12, anovel ras farnesylation inhibitor, is mediated by theMek/Erk pathway. J Cell Biol 2000;151:1179–92.

46. Karp JE, Lancet JE, Kaufmann SH, et al. Clinical andbiologic activity of the farnesyltransferase inhibitorR115777 in adults with refractory and relapsed acuteleukemias: a phase 1 clinical-laboratory correlative trial.Blood 2001;97:3361–9.

47. Sun J, Ohkanda J, Coppola D, et al. Geranylgeranyl-transferase I inhibitor GGTI-2154 induces breastcarcinoma apoptosis and tumor regression in H-rastransgenic mice. Cancer Res 2003;63:8922–9.

48. Tsukada Y, Miyazawa K, Kitamura N. High intensityERK signal mediates hepatocyte growth factor-

induced proliferation inhibition of the human hepa-tocellular carcinoma cell line HepG2. J Biol Chem2001;276:40968–76.

49. Osada S, Carr BI. Critical role of extracellular signal-regulated kinase (ERK) phosphorylation in novelvitamin K analog-induced cell death. Jpn J Cancer Res2000;91:1250–7.

50. Adachi T, Kar S, Wang M, Carr BI. Transient andsustained ERK phosphorylation and nuclear translo-cation in growth control. J Cell Physiol 2002;192:151–9.

51. Nguyen TT, Tran E, Nguyen TH, Do PT, HuynhTH, Huynh H. The role of activated MEK-ERKpathway in quercetin-induced growth inhibition andapoptosis in A549 lung cancer cells. Carcinogenesis2004;25:647–59.

52. Wang X, Martindale JL, Holbrook NJ. Requirement forERK activation in cisplatin-induced apoptosis. J BiolChem 2000;275:39435–43.

53. Iguchi T, Miyakawa Y, Yamamoto K, Kizaki M, IkedaY. Nitrogen-containing bisphosphonates induce S-phasecell cycle arrest and apoptosis of myeloma cells byactivating MAPK pathway and inhibiting mevalonatepathway. Cell Signal 2003;15:719–27.

54. Lee YJ, Cho HN, Soh JW, et al. Oxidative stress-induced apoptosis is mediated by ERK1/2 phosphory-lation. Exp Cell Res 2003;291:251–66.

55. Chung J, Uchida E, Grammer TC, Blenis J. STAT3serine phosphorylation by ERK-dependent and -inde-pendent pathways negatively modulates its tyrosinephosphorylation. Mol Cell Biol 1997;17:6508–16.

56. Sengupta TK, Talbot ES, Scherle PA, Ivashkiv LB.Rapid inhibition of interleukin-6 signaling andStat3 activation mediated by mitogen-activatedprotein kinases. Proc Natl Acad Sci U S A 1998;95:11107–12.

57. Lim CP, Cao X. Regulation of Stat3 activation by MEKkinase 1. J Biol Chem 2001;276:21004–11.

58. Jain N, Zhang T, Fong SL, Lim CP, Cao X. Repressionof Stat3 activity by activation of mitogen-activatedprotein kinase (MAPK). Oncogene 1998;17:3157–67.

59. Kanai M, Konda Y, Nakajima T, et al. Differentiation-inducing factor-1 (DIF-1) inhibits STAT3 activityinvolved in gastric cancer cell proliferation via MEK-ERK-dependent pathway. Oncogene 2003;22:548–54.

60. Zhang X, Blenis J, Li HC, Schindler C, Chen-Kiang S.Requirement of serine phosphorylation for formation ofSTAT-promoter complexes. Science 1995;267:1990–4.

61. Wen Z, Zhong Z, Darnell JE Jr. Maximal activation oftranscription by Stat1 and Stat3 requires both tyrosineand serine phosphorylation. Cell 1995;82:241–50.

62. Braunstein J, Brutsaert S, Olson R, Schindler C. STATsdimerize in the absence of phosphorylation. J BiolChem 2003;278:34133–40.

63. el-Deiry WS, Tokino T, Velculescu VE, et al. WAF1, apotential mediator of p53 tumor suppression. Cell1993;75:817–25.

64. Zeng YX, el-Deiry WS. Regulation of p21WAF1/CIP1expression by p53-independent pathways. Oncogene1996;12:1557–64.

65. Datto MB, Li Y, Panus JF, Howe DJ, Xiong Y, Wang XF.Transforming growth factor h induces the cyclin-dependent kinase inhibitor p21 through a p53-indepen-dent mechanism. Proc Natl Acad Sci U S A 1995;92:5545–9.

66. Ravitz MJ, Wenner CE. Cyclin-dependent kinaseregulation during G1 phase and cell cycle regulationby TGF-h. Adv Cancer Res 1997;71:165–207.

67. Pumiglia KM, Decker SJ. Cell cycle arrest mediatedby the MEK/mitogen-activated protein kinase pathway.Proc Natl Acad Sci U S A 1997;94:448–52.

68. De Siervi A, Marinissen M, Diggs J, Wang XF, Pages G,Senderowicz A. Transcriptional activation of p21(waf1/cip1) by alkylphospholipids: role of the mitogen-activated protein kinase pathway in the transactivationof the human p21(waf1/cip1) promoter by Sp1. CancerRes 2004;64:743–50.

69. Sepp-Lorenzino L, Rosen N. A farnesyl-protein trans-ferase inhibitor induces p21 expression and G1 block inp53 wild type tumor cells. J Biol Chem 1998;273:20243–51.

70. Zhou JM, Zhu XF, Pan QC, Liao DF, Li ZM, Liu ZC.Manumycin induces apoptosis in human hepatocellularcarcinoma HepG2 cells. Int J Mol Med 2003;12:955–9.

71. Barre B, Avril S, Coqueret O. Opposite regulation ofmyc and p21waf1 transcription by STAT3 proteins.J Biol Chem 2003;278:2990–6.

72. Darnell JE Jr. Transcription factors as targets forcancer therapy. Nat Rev Cancer 2002;2:740–9.

73. Yu H, Jove R. The STATS of cancer: new moleculartargets come of age. Nat Rev Cancer 2004;4:97–105.





2005;65:2861-2871. Cancer Res Kolaparthi Venkatasubbarao, Ahsan Choudary and James W. Freeman KinasesCancer Cell Lines Require Extracellular Signal-Regulated

Phosphorylation in Human Pancreatic(Tyr705)of STAT3Induced Inhibition−Farnesyl Transferase Inhibitor (R115777)

Updated version

http://cancerres.aacrjournals.org/content/65/7/2861

Access the most recent version of this article at:

Cited articles

http://cancerres.aacrjournals.org/content/65/7/2861.full#ref-list-1

This article cites 73 articles, 37 of which you can access for free at:

Citing articles

http://cancerres.aacrjournals.org/content/65/7/2861.full#related-urls

This article has been cited by 6 HighWire-hosted articles. Access the articles at:

E-mail alerts related to this article or journal.Sign up to receive free email-alerts

Subscriptions

Reprints and

[email protected] at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

Permissions

Rightslink site. (CCC)Click on "Request Permissions" which will take you to the Copyright Clearance Center's

.http://cancerres.aacrjournals.org/content/65/7/2861To request permission to re-use all or part of this article, use this link



http://cancerres.aacrjournals.org/content/65/7/2861.full#ref-list-1

http://cancerres.aacrjournals.org/content/65/7/2861.full#related-urls

http://cancerres.aacrjournals.org/cgi/alerts

mailto:[email protected]



farnesyl transferase inhibitor (r115777)–induced inhibition of...farnesyl transferase inhibitor...

Documents