1996;56:5369-5374.Cancer Res Christian Simon, Jose Juarez, Garth L. Nicolson, et al. Invasion

in VitroProtein Kinase Kinase, on Urokinase Expression and Effect of PD 098059, a Specific Inhibitor of Mitogen-activated  

  

Updated Version http://cancerres.aacrjournals.org/content/56/23/5369

Access the most recent version of this article at:

  

Citing Articles http://cancerres.aacrjournals.org/content/56/23/5369#related-urls

This article has been cited by 13 HighWire-hosted articles. Access the articles at:

  

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

SubscriptionsReprints and

.pubs@aacr.orgDepartment atTo order reprints of this article or to subscribe to the journal, contact the AACR Publications

Permissions.permissions@aacr.orgDepartment at

To request permission to re-use all or part of this article, contact the AACR Publications

American Association for Cancer Research Copyright © 1996 on March 12, 2013cancerres.aacrjournals.orgDownloaded from

ICANCER RESEARCH56. 5369-5374. December 1, 19961

Advances in Brief

Effect of PD 098059, a Specific Inhibitor of Mitogen-activated Protein KinaseKinase, on Urokinase Expression and in Vitro Invasjon'

Christian Simon, Jose Juarez, Garth L. Nicolson, and Douglas Boyd2

Departments of Head and Neck Surgery and Tumor Biology, University of Texas M.D. Anderson Cancer Center, Houston. Texas 77030 (C. S., J. J., D. B.J, and Institute forMolecular Medicine, irvine California 92619-2470 1G. L N.J

Abstract

The elevated expression of the urokinase-type plasminogen activatorgene, which is necessary for the invasive phenotype of several types ofcancers, is controlled by growth factors such as epidermal growth factor,transforming growth factor a, and fibroblast growth factor which bind toand activate protein tyrosine kinase transmembrane receptors. Since theseactivated receptors communicate with the nucleus via a signaling pathwayin which c-Raf-1, mitogen-activated protein kinase kinase 1 (MEK1), andthe extracellular signal-regulated kinases are sequentially activated, wedetermined the effect of a specific MEK1 inhibitor (PD 098059) on urokinase expression in two squamous cell carcinoma cell lines (UM-SCC-1 andMDA-TU-138) characterized as avid secretors of the plasminogen activator. PD 098059 treatment ofeither cell line reduced the amount of secretedurokinase in a dose-dependent manner. In contrast, a compound (daidzein) chemically unrelated to PD 098059 had little effect on urokinasesecretion. The effect of PD 098059 on urokinase secretion in UM-SCC-1cells was reversible and correlated with decreased extracellular signalregulated kinase 1 activIty. PD 098059 caused a dose-dependent reductionin the in vitro invasiveness of UM.SCC-1 cells whereas It had little effecton proliferation rates Transient transfection assays with a chloramphenicol acetyl transferase reporter driven by the urokinase promoter indicated that diminished secretion of the protease was largely a consequenceof reduced promoter activity. These findings suggest that interfering withMEK1 may provide a novel means of controlling the invasiveness of

tumors In which this signaling cascade is activated by autocrine and/orparacrine growth factors.

Introduction

The u-PA3 converts the inert zymogen plasminogen into the widelyacting serine protease plasmin and has been implicated in the invasiveness of several tumor types by virtue of its ability to control thesynthesis of the extracellular matrix-degrading plasmin (1). Althoughurokinase-dependent proteolysis can be modulated by endogenousinhibitors (2) as well as by the number of cell surface urokinasereceptors (3), plasminogen activator synthesis is nonetheless a majordeterminant of urokinase-dependent proteolysis, and the ability tocontrol the expression of this gene could have a profound effect on theinvasive phenotype.

Several studies have demonstrated that urokinase synthesis is regulated by a diverse set of growth factors including fibroblast growthfactor (4), EGF (5), and hepatocyte growth factor/scatter factor (6).

Received 8/26/96; accepted 10/17/96.The costs of publication of this article were defrayed in part by the payment of page

charges.Thisarticlemustthereforebe herebymarkedadvertisementin accordancewith18 U.S.C. Section 1734 solely to indicate this fact.

I This work was supported by NIH Grants ROl DE10845 and CA51539(D. B.), a grantfrom the National Foundation for Cancer Research (G. L N), and a fellowship (Si634/I-I) from the Deutsche Forschungsgemeinschaft (CS.).

2 To whom requests for reprints should be addressed, at Department ofTumor Biology,

M. D. Anderson Cancer Center, Box 108, 1515 Holcombe Boulevard, Houston, TX77030. Phone: (713) 792-8953; Fax: (713) 794-0209.

3 The abbreviations used are: u-PA, urokinase-type plasminogen activator, MiT,

3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide; CAT, chioramphenicolacetyl transferase; ERK, extracellular signal-regulated kinase; FBS, fetal bovine serum;MEKI, mitogen-activated protein kinase kinase 1; EGF, epidermal growth factor.

Indeed, a prevailing view is that the expression of urokinase in tumorcells and/or in surrounding stromal cells recruited by malignant cellsis driven, in part, by growth factor-activated signaling cascade(s)(7—9).Thus, we have hypothesized that interfering with the signalingevents connecting cell surface growth factor receptors with the muclear transcriptional machinery may represent a means of downregulating urokinase synthesis and tumor cell invasion.

It is now evident that altered gene expression brought about bygrowth factor-activated transmembrane receptors with intrinsic protein tyrosine kinases is Ras dependent (10), and the events occurringdownstream of this GTP/GDP-binding protein have been the subjectof intense investigation. Studies show that this stimulus can be propagated to the nucleus via a signaling module consisting of the sequential activation of the serine-threonine kinase c-Raf-l, MEK1 (alsoreferred to as MAPKK), and the ERKs (1 1—13).In fact, we recentlyreported that a squamous cell carcinoma cell line (UM-SCC-l), characterized by its high expression of this plasminogen activator, contamed a constitutively activated ERK1 (14). Consequently, we undertook the current study to determine whether interfering with thissignaling pathway using a specific inhibitor of MEKI (PD 098059;Refs. 15—17)was repressive for urokinase synthesis and in vitroinvasiveness of UM-SCC- 1 cells.

Materials and Methods

Tissue Culture. UM-SCC-l and MDA-TU-138 cells were obtained fromDrs. Thomas Carey (University of Michigan, Ann Arbor, MI) and Gary

Clayman (M. D. Anderson Cancer Center, Houston, TX), respectively. Cellswere maintained in McCoy's SA culture medium supplemented with 10%FBS. For the collection of conditioned medium, 80% confluent UM-SCC-lcells were washed extensively with serum-free medium (McCoy's 5A mediumsupplemented with 5 @xg/mlinsulin, 10 mg/mI EGF, and 4 .tg/ml transfemn)

and replenished with the same. The culture medium was collected 24 h later,centrifuged to remove cellular debris, and proliferation assessed with MU(18).

Growth curves were constructed as follows. UM-SCC-l cells (50,000) in10% FBS were plated into 35-mm dishes. After 12 h for cell attachment, thecultures were changed to fresh medium supplemented with varying amounts ofPD 098059 (day 0). At the indicated times, the cells were replenished withfresh medium containing 0.2 mg/mI of the vital stain MU. Water-insolubleformazan crystals were dissolved in DMSO and aliquots were read at 570 mm.

Western Blotting. Cell extracts (10 @tgprotein) generated in PBS containing 1% NP4O, 0.5% sodium deoxycholate, 0.1% SDS, and I mM phemylmethylsulfomyl fluoride (radioimmunoprecipitatiomassay buffer) or conditionedmedium from equal numbers of cells was denatured in the presence (ERK) orabsence (urokinase) of reducing agent, proteins resolved by SDS-PAGE, andthen transferred to a nitrocellulose filter. The filter was blocked with 3% BSAand incubated with a polyclonal antibody to u-PA (389; American Diagmostica,Greenwich, CT) or with a polyclonal antibody (93; Santa Cruz Biotechnology,Santa Cruz, CA) that cross-reacts with human ERKI and ERK2. Subsequently,the blot was incubated with horseradish peroxidase-conjugated antirabbit IgGand immunoreactive bands were visualized with enhanced chemiluminescemceas described by the manufacturer (Amersham, Arlington Heights, IL).

5369

American Association for Cancer Research Copyright © 1996 on March 12, 2013cancerres.aacrjournals.orgDownloaded from

REDUCED INVASION AND UROKINASE EXPRESSION BY PD 098059

Zymography. Aliquots of conditioned medium normalized for proliferative differences were diluted 100-fold, denatured in the absence of reducing

agent, and electrophoresed in a 10.5% SDS-polyacrylamide gel containing0.1% (w/v) casein and 10 jxg/ml plasminogen. The gel was incubated at 37°C

for 2 h in the presence of 2.5% Triton X-lOOand subsequently for 10 h in abuffer containing 0.15 MNaC1and 100 mr@iTris (pH 7.5). The gel was thenstained for protein with 0.25% Coomassie blue. Plasmimogen-depemdentproteolysis was detected as a white zone in a dark field.

In-Gel Kinase Assays. Kimaseassays were performedas describedelsewhere (19) but with minor modifications. Briefly, cells were extracted withbuffer A [1% NP4O, 25 mM Tns-HCI (pH 7.4), 25 ntM NaCI, 1 mM sodiumvanadate, 10 mMNaF, 10 mMsodium PP, 10 ass okadaic acid, 0.5 mi@iEGTA,and 1 mMphenylmethylsulfomylfluoride], insoluble material was removed bycentrifugation, and extracted protein (750 @xg)was incubated simultaneouslywith 1 ,.tg of the anti-ERK antibody (93; Santa Cruz Biotechnology) andprotein A-agarose beads. The beads were washed with buffer A and resuspended in 2x sample buffer, and the immune complexes were electrophoresedin a polyacrylamide gel containing myelin basic protein. The gel was treated

sequentially with buffers containing 20% 2-propanol, 5 mrsi2-mercaptoethamol, 6 M guanidime HC1, and 0.04% Tweem 20/5 mr@i2-mercaptoethanol.Afterwards, the gel was incubated at 25°Cfor 1h with 40 @xMAlP and 25 @Ciof [@y-32P]ATPin a buffer containing 2 mt@iDTT/0.l me@iEGTA/5 ms@iMgCl2.The gel was themwashed in a solution containing 5% trichloroacetic acid and1% sodium PP@,dried, and autoradiographed.

In Vitro Invasion Assays. Invasionassays were carriedout essentially asdescribed elsewhere (20). UM-SCC-l cells (100,000) were plated out onMatrigel-coated filters (Becton Dickinson, Bedford, MA) and cultured for

72 h. After this time the filter was incubated with MU. Crystals on the lower

and upper aspects of the filter were dissolved in DMSO and color intensitydetermined at 570 mm. The amount of invasion was determined on the basis of

the MU activity on the lower side of the filter as a percentage of the totalactivity in the chamber.

CAT assays. These were as described previously using hexadimethrinebromide (14). UM-SCC-l cells were cotransfected at 70% confluency with

CAT reporter constructs and 5 @xgof an expression vector bearing the @3-galactosidase gene. After 5 h the cells were shocked with 38% DMSO for 8 mmand cultured for an additional 48 h. The cells were then harvested, lysed, andassayed for /3-galactosidase activity. CAT activity was subsequently measured

by incubating cell lysates (normalized for transfectiomefficiency) at 37°Cwith4 @LM[‘4C]chloramphenicol and 1 mg/mI acetyl-CoA. The acetylated products

were extracted with ethyl acetate and subjected to TLC using chloroform!methanol (95:5) as the mobile phase.

Enzyme Assays. Urokinase activity was determined using a urokinasespecific chromogemicsubstrate [Glu-Gly-Arg-pNA-(S2444)] as described bythe manufacturer (KabiVitrum, Stockholm, Sweden).

Results

Effect of PD 098059 on Urokinase Secretion and ERK1 Activity. We first asked whether the secretion of urokinase by UM-SCC-lcells was sensitive to PD 098059. Cells were treated with the MEK1inhibitor for 4 days, and the conditioned medium was assayed forurokinase protein by Western blotting. These cells are avid secretors(14) of this plasminogen activator (Fig. lA), this being similar to theproduction of this enzyme by squamous cell carcinoma in vivo asshown by in situ hybridization and immunohistochemical studies (21).Treatment of UM-SCC-l cells with increasing concentrations of PD098059, however, resulted in a dose-dependent inhibition in theamount of urokinase secreted by UM-SCC-l cells. Densitometricanalysis indicated that 10 @LMof the MEK1 inhibitor reduced urokinase secretion by @°80%with 50 @.LMbeing slightly more effective(@85% inhibition). PD 098059 treatment also decreased the amountof urokinase secreted by another head and neck squamous cell carcinoma cell line (MDA-TU-l38), indicating that the effect ofthe MEK1inhibitor was not limited to UM-SCC- 1 cells (Fig. lB). It is unlikelythat PD 098059 causes a general shut-down in protein synthesis, since

the level of the 72-kDa type IV collagenase was unchanged in UMSCC-i cells treated with the inhibitor.4

To confirm that the concentrations of PD 098059 that reducedurokinase secretion were inhibitory for MEK1 activity, UM-SCC-1cells were treated with the agent and cell extracts were immunoprecipitated with an anti-ERK antibody and assayed for ERK activityusing an in-gel kinase assay (Fig. 1C). A kinase activity, indistinguishable in size from ERK1 (44 kDa), was detected in UM-SCC-lcells treated with the carrier only. However, 1 p.M PD 098059 substantially reduced the activity of ERK1. Higher concentrations of thisMEK1 inhibitor had an even more dramatic effect, reducing theactivity of this kinase below the detection limits of this assay. Thedecreased activity did not reflect a reduced amount of ERK1 proteinas evident in Western blots (Fig. 1D). Thus, concentrations of PD098059 that reduced ERK1 activity also diminished the amount ofurokinase protein in the conditioned medium.

Effect of PD 098059 on Urokinase Synthesis Is Not Due to aToxic Effect of a Small Molecular Weight Compound. To rule outthe possibility that the diminished secretion of urokinase broughtabout by PD 098059 was a consequence of a nonspecific toxic effectof a small molecular weight compound, the following experimentswere carried out. First, the ability of daidzein (4', 5-dihydroxyisoflavone; Calbiochem, La Jolla, CA), which like PD 098059 has a triplering, to attenuate urokinase secretion was determined. Although daidacmhadonlya marginaleffectontheamountof secretedurokinase,an identical concentration range of PD 098059 caused a dramaticreduction in the secretion of this protease (Fig. 14). Second, wedetermined whether the effect of PD 098059 on urokinase activity wasreversible. UM-SCC-l cells were pretreated with or without theMEK1 inhibitor for 72 h. The cells were then washed and replenishedwith serum-free medium supplemented with or without PD 098059.Conditioned medium was collected 24 h later and assayed for urokinase activity by zymography (Fig. 2B). A proteolytic activity, whichcomigrated with authentic two-chain urokinase, was detected in theconditioned medium of untreated UM-SCC-l cells (Control). Expectedly, PD 098059 treatment reduced protease activity in the conditioned medium (PD 098059). However, higher amounts of urokinaseactivity were observed when the PD 098059 was omitted from theserum-free medium (Remove PD 098059), indicating reversibility ofthe PD 098059 effect on urokinase production. Taken together, thesedata suggest that the inhibitory effect of PD 098059 on urokinasesecretion is neither due to an unspecific effect of a small molecularweight compound nor a consequence of cell toxicity.

Effect of PD 098059 on the Invasion ofan Extracellular Matrixcoated Porous Filter by UM-SCC-1 Cells. We previously demonstrated that UM-SCC-l cells were highly invasive in vitro, this beingurokinase dependent (22). Consequently, UM-SCC-l cells weretreated with concentrations of PD 098059 that reduced urokinasesecretion and assayed for in vitro invasion (Fig. 3A). A concentrationof S ,.LMPD 098059 caused an approximate 55% reduction in invasion, whereas 10 @.LMagent diminished this parameter by 65%. Similarly, an antiurokinase antibody (394; American Diagnostica) directed at the catalytic moiety reduced invasion by more than 70%.Since these assays were carried out over a 72-h period, it is possiblethat the reduced invasion is secondary to a slower proliferation ofUM-SCC-l cells. To address this possibility, UM-SCC- 1 cells weretreated with varying amounts of PD 098059 and assayed for cellgrowth (Fig. 3B). We found that concentrations of 1—10 @M,whichdiminished urokinase secretion and in vitro invasion, had minimal

4 R. Gum, H. Wang, E. Lemgxel, J. Juarez, and D. Boyd, Regulation of 92 kDa type IVcollagenase expression by the JNK and ERK-dependent signaling cascades, submitted forpublication.

5370

American Association for Cancer Research Copyright © 1996 on March 12, 2013cancerres.aacrjournals.orgDownloaded from

REDUCED INVASION AND UROKINASE EXPRESSION BY PD 098059

rsc-uPA(mg)

c@@ ,i,@ @0 — C'@@0C)

Fig. I. PD 098059 reduces urokinase secretion and ERKIactivity in UM-SCC-l cells. A and B. UM-SCC-l cells (A) orMDA-TU-l38 (B) cells in 10% FBS were treated for 2 dayswithout (Control) or with varying concentrations of PD098059 or DMSO. In A, the amount of DMSO is equivalent to50 @LMPD 098059 whereas in B, 0.001% DMSO is equivalentto I @.sMPD 098059. The cells were them replenished withserum-free medium containing PD 098059 or the carrier. After48 h, the conditioned medium was harvested and proliferationrates were assessed with MiT. Aliquots of conditioned medium, nonnalized for proliferation differences, along withstandard amounts of authentic recombinant single-chain urokinase (rsc-uPA) were subjected to Western blotting using apolyclonal antiurokinase antibody. C and D, UM-SCC-l cellswere treated as described for A, with the exception that eachconcentration of PD 098059 was paired with an equivalentamount of carrier (1 ELM,0.001%; 5 @M,0.005%; 10 @sM,0.01% for PD 098059 and DMSO. respectively). Cells wereextracted with buffer A or radioimmunoprecipitation assaybuffer and analyzed for ERK activity (C) and ERK protein (D)using in-gel kinase assay and Westem blotting, respectively.The data are typical of duplicate experiments.

D

¼S'

-i

25 50

..O..m. @*55@a

I ••- -.@ ______ @-.-—— 44 kDa

w W

PD 098059 (pM) CARRIER (%)

effect on proliferative rates. A higher concentration of 50 LMdid,however, reduce the growth of UM-SCC-l cells. PD 098059 did nothave a direct effect on urokinase catalytic activity, since the activationof a specific chromogenic substrate (52444; KabiVitrum) by exogenous two-chain urokinase (Calbiochem) was unaffected (Fig. 3C) bya concentration of the MEK1 inhibitor (10 tiM) which reduced urokinase protein levels and in vitro invasion.

Effect of PD 098059 on Urokinase Promoter Activity. To determine whether the reduced urokinase secretion observed in PD098059-treated UM-SCC-l cells was a consequence of an attenuatedpromoter activity, the following experiments were carried out. First,UM-SCC-l cells were pretreated with 10 p.MPD 098059 for varyingperiods and subsequently transfected with a CAT reporter driven bythe upstream regulatory sequence of the urokinase gene. UntreatedUM-SCC-l cells strongly activated the urokinase promoter-drivenCAT reporter construct (Fig. 4A), consistent with the high rate ofurokinase transcription as reported previously (14). Treatment ofUM-SCC-l cells with PD 098059 immediately following transfection(0 days pretreated) caused a 65% reduction in CAT activity. Thisreduction was further enhanced by pretreating the cells with theMEK1 inhibitor for longer periods, with a 2-day lead time resulting inan 88% reduction in CAT activity. The lower CAT activity evident

with PD 098059 pretreatment was not a consequence of lower transfection efficiencies, since all experiments were corrected for alteredtransfection efficiencies as assessed with a (3-galactosidase expressionvector. UM-SCC-l cells were then pretreated for I day with varyingconcentrations of PD 098059 and assayed for urokinase promoteractivity (Fig. 4B). Under these conditions, 1 p@ MEKI inhibitorreduced urokinase promoter activity by 75% with higher concentratioms of the agent reducing CAT activity below the detection limits ofthe assay. In contrast, the activity of a CAT reporter construct drivenby the Rous sarcoma virus long terminal repeat was not decreased bythe MEK1 inhibitor, indicating that the PD 098059 was not causing ageneral reduction in gene transcription.

Discussion

There is now a large body of evidence implicating the u-PA in theinvasiveness of many types of cancers (2). An emerging picture is thatthe synthesis of this protease in the tumor cells and/or in the stromalcompartment is modulated, at least in part, by growth factors bindingto and activating cell surface receptors (9, 23, 24). Certainly, a varietyof mitogens, including transforming growth factor a (25) and hepatocyte growth factor (6), can induce urokinase expression in cultured

5371

A@ ___@ 1 5 10 50

0―B

PD 098059(jiM) - 1 : 5 ;; 10 ; 50DMSO(%) - -@ -@ -@ -

C

I

—@ u@ 55

:@!@ :::::@ “.—.@i@@ :@: .@____@ kDa

1 5 10 0.001 0.005 0.010

American Association for Cancer Research Copyright © 1996 on March 12, 2013cancerres.aacrjournals.orgDownloaded from

511

REDUCED INVASION AND UROKINASE EXPRESSION BY PD 098059

tumor cells (26) and are present in invasive malignancies such as headand neck cancers (27). Considering the likelihood that protease cxpression is regulated by the concerted action of multiple growth

@ 4— 55 kDa factors present in invasive cancers, the targeting of any one particular

A

,p,@ @.

@ 1 5 10 1 5 10@ ____ ___ a@ —.-----—-— 8

DAIDZEIN (iiM) PD 098059 (jiM) @.

mitogen may prove inefficient in repressing protease synthesis. Amore practical approach might be to target a signaling cascade utilizedby some of the growth factors that regulate urokinase expression.Since the c-Raf-1-MEK1-ERK cascade represents one such commonsignaling module (1 1—13),we have investigated whether interferingwith this pathway diminishes urokinase synthesis. This study showsfor the first time the repressive effect of a MEK1 inhibitor (PD098059) on urokinase secretion by two cell lines characterized as avidproducers of this protease. Moreover, treatment of UM-SCC-l cellswith PD 098059 reduced their in vitro invasive capacity. Although thediminished urokinase synthesis brought about by PD 098059 is mostlikely a contributing factor to the attenuated invasiveness of UMSCC-l cells, it should be emphasized that it is unlikely to fullyaccount for the reduced invasion. More likely is that this agent affectsthe expression of multiple genes involved in the invasive process (26,28—31).In fact, in a separate study4 we have shown that PD 098059also reduces the expression of the 92-kDa type IV collagenase inUM-SCC-1 cells.

Although the PD 098059 was an effective inhibitor of urokinasesynthesis, a small amount of the plasminogen activator could still bedetected even using the highest concentrations of the MEK1 inhibitor.There could be several possible reasons for this. First, a portion of theendogenous MEK1 may have escaped the inhibitory effect of the PD098059. Although the in-gel kinase assay would argue against thispossibility, we cannot rule out the prospect that trace amounts ofMEK1, which are below the detection limit of our assay, are presentin PD 098059-treated UM-SCC-l cells. An alternative explanation isthat urokinase synthesis in these cells is driven by both MEK1-dependent and -independent signaling pathways, the latter being insensitive to the PD 098059. Indeed, urokinase expression can beup-regulated by transforming growth factor @3(32), IFN-'y (33), andtumor necrosis factor a (34, 35) in vitro, and there is growingevidence that these cytokines utilize MEK1-independent signaling

B

55 kDa —@

Fig. 2. PD 098059, but not daidzein, reduces urokimase secretion in a reversiblemanner. A, UM-SCC-l cells in 10% FBS were treated for 3 days with the indicatedconcentration of PD 098059, daidzein, or an amount of carrier (DMSO) corresponding tothe highest level of the compounds. The cells were them replenished with serum-freemedium supplemented with the agents or the carrier and incubated for another 48 h.Conditioned medium was harvested and aliquots, adjusted for varying cell proliferation,were analyzed for urokinase using Westem blotting. Recombinant single-chain urokinase(rsc-uPA) was included in a separate lane of the immunoblot. B, UM-SCC-l cells in 10%FBS were cultured for 72 h without (Control) or with 10 @sMPD 098059 (PD 098059 andRemove PD 098059) or with an equivalent amount of carrier (DMSO). After this time, alldishes were replenished with serum-free medium supplemented as follows: cells treatedwith the MEK1 inhibitor were either supplemented with PD 098059 or changed toserum-free medium without the agent (Remove PD 098059); cells initially cultured in theabsence of the inhibitor were replenished without PD 098059 (Control) or with DMSO.After an additional 24 h, conditioned medium was harvested and aliquots, corrected forproliferation differences, were subjected to zymography in a gel containing plasminogenand casein. As a control, two-chain urokinase (tc-uPA) was run in parallel lanes. Theexperiment was repeated twice.

c:@0.160.14

In 0.12

@ 0.1“1-,@0.080.060.04

0.02

0

Fig. 3. Inhibition of in vitro invasion by PD098059. A, UM-SCC-l cells were pretreated for 1day with the indicated concentration of PD 098059or an equivalent amount of DMSO. After incubation, the cells were harvested and cells (l0@) wereplated onto a Matrigel-coated porous filter in theabsence or presence of the MEK1 inhibitor. Thecells were cultured for 3 days and assayed for invitro invasion. Invasion is expressed as a percentagedecrease with respect to the carrier alone. B, UMSCC-l cells (0.5 X l0@)were plated in 10% FBS.After 12 h, the cells were treated with PD 098059(day 0). At the indicated times thereafter, the cellswere assayed for proliferation using MiT. Proliferatiom is expressed as a fold increase in A570 valuerelative to that determined at day 0. C, a solutioncontaining varying amounts of two-chain urokinase(tc-uPA) and S2444 was incubated with (A) orwithout (@) 10 @LMPD 098059 at 37°Cfor 5 mm.Absorbance at 405 mm was then determined with amicroplate reader. The data are typical of duplicateexperiments.

50 100 150 200

tc-uPA (Ploug units)

—4-1@iAPD-@--5uM PD

lOuMPD-@@-00aM

—•—0.05%DM50

5 10 10PD 098059 (j.tM)

0 1 2 3 4Days treated with PD 098059

5372

A75c.I_@ 70

@0@ 60.@0

.—.— 55,.o Cl)

:@@

@@ 45— -—40

B @ioo

@ 102C.)

0tz@

American Association for Cancer Research Copyright © 1996 on March 12, 2013cancerres.aacrjournals.orgDownloaded from

REDUCED INVASION AND UROKINASE EXPRESSION BY PD 098059

A

- +++++++u-PACAT

0 1 2 (days pretreated)

0.01% DMSO

.fffffffff- +++++++- u-PACAT

— — 1 5 10 - - - 0 10 PD 098059 (riM)

- - - - -@@@ - - DMSO(%)

- - - - - - - - + + RSV-CAT

Fig. 4. PD 098059 reduces urokinase promoter activity. A, UM-SCC-l cells weretreated with PD 098059 or DMSO for the indicated times, after which they weretransfected with 8 ,xg of a CAT reporter construct driven by the urokinase promoter (u-PACAT). After transfectiom, the cells were returned to culture medium containing PD 098059or DMSO. The cells were harvested 48 h later and assayed for CAT activity. B,UM-SCC-l cells were pretreated for 2 days with the indicated concentrations of PD098059 or DMSO. After this time, the cells were transfected with a CAT reporterconstruct driven by the urokinase promoter (u-PA CAT, 8 @sg)or a CAT construct with theRoes sarcoma virus long terminal repeat (RSV-CAT, 6 @g)and the cells were cultured withPD 098059 or DMSO. The cells were harvested after 48 h and assayed for CAT activity.The amount of acetylated chloramphenicol was determined with a 603 Betascope. Theexperiment was carried out on two separate occasions.

012

l0@iMPD098059

B<1 8 2 <1 <1 6 8 8 6 8

•0

Chloramphemcolacetylated (%)

5373

References

I. Ossowski, L., Russo-Payne, H., and Wilson, E. Inhibition of urokinase-type plasminogen activator by antibodies: the effect on dissemination of a human tumor in thenude mouse. Cancer Res., 5!: 274—281,1991.

2. Tryggvason, K., Hoyhtya, M., and Salo, T. Proteolytic degradation of extracellularmatrix in tumor invasion. Biochim. Biophys. Acta, 907: 191—217, 1987.

3. Markus, G. The relevance of plasminogen activators to neoplastic growth. Enzyme,40:158—172,1988.

4. Presta, M., Maier, A., Rusmati, M., Moscatelli, D., and Ragnoui, 0. Modulation ofplasminogen activator activity in human endometrial adenocarcinoma cells by basicfibroblast growth factor and transforming growth factor b. Cancer Res., 48: 6384—6389,1988.

5. Rorth, P., Nerlov, C., Blasi, F., and Johmsen, M. Transcription factor PEA3 participates in the induction of urokimase plasminogen activator transcription in murimekeratimocytes stimulated with epidermal growth factor or phorbol-ester. NucleicAcids Res., 18: 5009—5017, 1990.

6.Pepper,M.S.,Matsumoto,K.,Nakamura,T.,Orei,L.,andMontesano,R.Hepatocytegrowth factor increases urokinase-type plasminogen activator (u-PA) and u-PA receptor expression in Madin-Darby canine kidney epithelial cells. J. Biol. Chem., 267:20493—20496,1992.

7. Korczak, B., Kerbel, R. S., and Dennis, J. W. Autocrine and paracrine regulation oftissue inhibitor of metalloproteinases, transin, and urokinase gene expression inmetastatic and non-metastatic mammary carcinoma cells. Cell Growth & Differ., 2:335—341,1991.

8. Borchers, A. H., Powell, M. B., Fusenig, N., and Bowden, G. T. Paracrine factorand cell-cell contact-mediated induction of protease and c-ets gene expression inmalignant keratinocyte/dermal fibroblast cocultures. Exp. Cell Res., 213: 143—147,1994.

9. Pyke. C.. Kristensen. P., Ralfakiaer, E., Eriksen, J., and Dano, K. The plasminogenactivation system in human colon cancer: messenger RNA for the inhibitor PAl- 1 islocated in endothelial cells in the tumor stroma. Cancer Res., 51: 4067—4071,1991.

10. Stacey, D. W., Roudebush, M., Day, R., Mosser, S. D., Gibbs, J. B., and Feig, L. A.Dominant inhibitory ras mutants demonstrate the requirement for ras activity in theaction of tyrosine kimase oncogenes. Oncogene, 6: 2297—2304, 1991.

I 1. Mindem,A., Lin, A., McMahom,M., Lange-Carter, C., Derijard, B., Davis, R. J.,Johnson, G. L., and Karin, M. Differential activation of ERK and iNK mitogenactivated protein kinases by raf-l and MEKK. Science (Washington DC), 266:1719—1722,1994.

12. Bruder, J. T., Heidecker, G., and Rapp, U. R. Serum-, TPA-, and Ras-imducedexpression from Ap-llEts-drivem promoters requires Raf-l kimase. Genes Dev., 6:545—556,1992.

13. Crews, C. M., Alessandrini, A., and Erikson, R. L. The primary structure of MEK, aprotein kinase that phosphorylates the ERK gene product. Science (Washington DC).258: 478—480, 1992.

14. Lemgyel. E.. Gum, R., Stepp, E., Juarez, J., Wang, H., and Boyd, D. Regulation ofurokinase-type plasmimogen activator expression by an ERK-l-dependent signalingpathway in a squamous cell carcinoma cell line. J. Cell. Biochem.. 6!: 430—443,1996.

15. Dudley, D. T., Pang, L., Decker, S. J., Bridges, A., and Saltiel, A. R. A syntheticinhibitor of the mitogen-activated protein kinase cascade. Proc. NatI. Acad. Sci. USA,92: 7686—7689, 1995.

16. Alessi, D. R., Cuenda. A., Cohen, P., Dudley, D. T.. and Saltiel, A. R. PD 098059 isa specific inhibitor of the activation of mitogem-activated protein kimase kinase invitro and in vivo. J. Biol. Chem., 270: 27489—27494, 1995.

17. Servant, M. J., Giassom, E., and Meloche, S. Inhibition of growth factor-inducedprotein synthesis by a selective MEK inhibitor in aortic smooth muscle cells. J. Biol.Chem., 271: 16047—16052, 1996.

18. Mossman, T. Rapid colorimetric assay for cellular growth and survival: application toproliferation and cytoxicity assays. J. Immunol. Methods, 65: 55—63,1983.

19. Monno, N., Mimura, T., Hamasaki, K., Tobe, K., Ueki, K., Kikuchi, K., Takehara, K.,Kadowaki, T., Yazaki, Y., and Nojima, Y. Matrix/integrin interaction activates themitogen-activated protein kimase, p44°°@1and p42°―2.J. Biol. Chem., 270: 269—273, 1995.

20. Schlechte, W., Murano, 0., and Boyd, D. D. Examination of the role of the urokinasereceptor in human colon cancer mediated laminin degradation. Cancer Res., 49:6064—6069,1989.

21. Sappino, A., Belin, D., Huarte, J., Hirschel-Scholz, S., Saurat, J., and Vassalli, J.Differential protease expression by cutaneous squamous and basal cell carcinomas.J. Clim.Invest., 88: 1073—1079,1991.

22. Clayman, G., Wang, S. W., Nicolson, G. L., El-Naggar, A., Mazar, A., Henkin, J.,Blasi, F., Goepfert, H., and Boyd, D. Regulation of urokinase-type plasminogenactivator expression in squamous cell carcinoma of the oral cavity. Int. 3. Cancer, 54:73—80.1993.

23. Boxman, I. L. A., Quax, P. H. A., Clemens, W. G. M., Papapoulos, S. E., Verheijen,J., and Ponec, M. Differential regulation of plasminogen activation in normal keratinocytes and SCC-4 cells by fibroblasts. 1. Invest. Dermatol., 104: 374—378, 1995.

24. Jeffers, M., Rong, S., and Vande Woude, G. F. Enhanced tumorigenicity and invasion-metastasis by hepatocyte growth factor/scatter factor-Met signalling in humancells concomitant with induction of the urokinase proteolysis network. Mol. Cell.Biol.,16:1115—1125,1996.

25. Jensen, P. J., and Rodeck, U. Autocrine/paracrine regulation of keratinocyte urokinaseplasminogen activator through the TGF-aIEGF receptor. J. Cell. Physiol., 155:333—339,1993.

26. Matsumoto, K., Nakamura, T., and Kramer, R. H. Hepatocyte growth factor/scatterfactor induces tyrosine phosphorylation of focal adhesion kinase (pl25'@―) and

<1 13 6 2 2 14 11 10acetylated (%)

..@ 0••.0

pathways. For example, recent studies have shown that the effect oftransforming growth factor (3 is propagated through a TAK1-dependent signaling module leading, via MKK6, to p38 activation (36, 37).Conversely, the biological effects of the IFNs are mediated throughthe JAK-STAT pathway (38), whereastumor necrosis factor a signalsto the nucleus through ceramide activation of the c-Jun aminoterminalkinases (39, 40).

We have demonstrated the effectiveness of the MEK1 inhibitor PD098059 in repressing urokinase synthesis and the in vitro invasivenessof cultured head and neck cancer cells. These findings raise theexciting possibility that PD 098059, or related compounds, may be ofclinical value in reducing the invasiveness of tumors in which thissignaling cascade is constitutively activated (41).

Acknowledgments

We are grateful to Dr. Alan Saltiel (Parke Davis, Ann Arbor, MI) for thegenerous gift of the PD 098059. We express our appreciation to AmericanDiagnostica for the gift of the antiurokimaseantibodies (389 and 394). Theurokinase promoter CAT reporter was kindly provided by Dr. Francesco Blasi(University of Milan, Milan, Italy). We are also indebted to Dr. Jack Henkin

(Abbott Laboratories, Chicago, IL) for kindly providing both the single- andtwo-chain urokimases.

American Association for Cancer Research Copyright © 1996 on March 12, 2013cancerres.aacrjournals.orgDownloaded from

REDUCED INVASION AND UROKINASE EXPRESSION BY PD 098059

promotes migration and invasion by oral squamous cell carcinoma cells. J. Biol.Chem.,269:31807—31813,1994.

27. Grandis, J., and Tweardy, D. J. Elevated levels of transforming growth factor a andepidermal growth factor receptor messenger RNA are early markers of carcimogenesisin head and neck cancer. Cancer Res., 53: 3579—3584, 1993.

28. Bernhard, E. J., Muschel, R. J., and Hughes, E. N. M, 92,000 gelatinase releasecorrelates with the metastatic phenotype in transformed rat embryo cells. Cancer Res.,50: 3872—3877,1990.

29. Garbisa, S., Scagliotti, G., Di Francesco, C., Caenazzo, C., Onisto, M., Micela, M.,Stetler-Stevenson, W. G.. and Lions, L. Correlation of serum metalloproteinase levelswith lung cancer metastasis and response to therapy. Cancer Rca., 52: 4548—4549,1992.

30. Crawford, H. C., and Matrisian, L. M. Tumor and stromal expression of matrixmetalloproteinases and their role in tumor progression. Invasion & Metastasis, 14:234—245, 1994.

31 . Xuan, J. W., Hots, C., Shigeyama, Y., D'Errico, J. A., Somerman, M. J., andChambers, A. F. Site-directed mutagenesis of the arginime-glycime-aspartic acidsequence in osteopontin destroys cell adhesion and migration functions. J. Cell.Biochem., 57: 680—690, 1995.

32. Keski-Oja, J., Blasi, F., Leof, E. B., and Moses, H. L. Regulation of the synthesis andactivity of urokinase plasminogen activator in A549 human lung carcinoma cells bytransforming growth factor-b. J. Cell Biol., 106: 45 1—459,1988.

33. Collart, M., Belin, D., Vassalli, J., Kossodo, S., and Vassalli, P. Gamma interferonenhances macrophage transcription of the tumor necrosis factor/cachetin, imterleukin1. and urokinase genes, which are controlled by short-lived repressors. J. Exp. Med.,164:2113—2118,1986.

34. Georg, B., Helseth, E., Lund, L. R., Skandsem, T., Riccio, A., Dano, K., Umsgaard, G.,and Andreasen, P. A. Tumor necrosis factor-alpha regulates mRNA for urokinasetype plasminogen activator and type-l plasmimogen activator inhibitor in humanneoplastic cell limes. Mol. Cell. Endocrinol., 61: 87—96,1989.

35. Niedbala, M. J., and Picarella, M. S. Tumor necrosis factor induction of endothelialcell urokimase-type plasminogen activator mediated proteolysis of extracellular matrixand its antagonism by gamma-interferon. Blood, 79: 678—687,1992.

36. Shibuya, H., Yamaguchi, K., Shirakabe, K., Tonegawa, A., Gotoh, Y., Uemo, N., Inc.K., Nishida, E., and Matsumoto, K. TAB1: an activator of the TAK1 MAPKKK inTGF-b signal transduction. Science (Washington DC), 272: 1179—1182, 1996.

37. Moriguchi, T., Kuroyanagi, N., Yamaguchi, K., Gotoh, Y., Inc. K., Kano, T.,Shirakabe, K., Mum, Y., Shibuya, H., Matsumoto, K., Nishida, E., and Hagiwara, M.A novel kinase cascade mediated by mitogen-activated protein kinase kimase6 andMKK3. J. Biol. Chem., 271: 13675—13679,1996.

38. Karmn, M. Signal transduction from the cell surface to the nucleus through thephosphorylation of transcription factors. Curr. Opin. Cell Biol., 6: 415—424, 1994.

39. Westwick, J. K., Bielawska, A. E., Dbaibo, G., Hannun, Y. A., and Brenner, D. A.Ceramide activates the stress-activated protein kinases. J. Biol. Chem., 270: 22689—22692,1995.

40. Hirano, M., Osada, S., Aoki, T., Hirai, S., Hosaka, M., Inoue, J., and Ohmo, S. MEKkinase is involved in tumor necrosis factor a-induced NF-.cB activation and degradation of 1KB-a. J. Biol. Chem., 271: 13234—13238, 1996.

41. Oka. H., Chatani, Y., Hoshino, R., Ogawa, 0., Kakehi, Y., Terachi, T., Okdada, Y.,Kawaichi, M., Kohmo, M., and Yoshida, 0. Constitutive activation of mitogenactivated protein (MAP) kinases in human renal cell carcinoma. Cancer Res., 55:4182—4187,1995.

5374

American Association for Cancer Research Copyright © 1996 on March 12, 2013cancerres.aacrjournals.orgDownloaded from