TGF beta promotes proliferation of cultured SMC via both PDGF AA dependent and PDGF AA independent mechanisms

TGF-beta promotes proliferation of cultured

SMC via both AA-dependent and

PDGF-AA-independent mechanisms.

G A Stouffer, G K Owens

J Clin Invest.

1994;

93(5)

:2048-2055.

https://doi.org/10.1172/JCI117199

.

Transforming growth factor-beta 1 (TGF-beta 1) has been implicated in mediating smooth

muscle cell (SMC) growth after vascular injury. Studies examining TGF-beta-induced

growth of cultured SMC have identified only modest mitogenic effects which are largely

dependent on autocrine production of platelet-derived growth factor-AA (PDGF-AA). Recent

studies have suggested, however, that TGF-beta also may have delayed growth effects

independent of PDGF-AA. The aims of the present studies were to examine the effects of

beta on chronic growth responses of cultured SMC. Results demonstrated that

TGF-beta elicited a delayed growth response (24 fold increase in 3H-TdR incorp. from 48-72 h)

and enhanced SMC production of PDGF-AA (eightfold increase at 24 h). Neutralizing

antibodies to PDGF-AA, however, inhibited only 10-40% of delayed TGF-beta-induced

growth. Co-treatment with TGF-beta transiently delayed epidermal growth factor (EGF)-,

basic fibroblast growth factor (bFGF)-, or PDGF-BB-induced entry into S phase but

enhanced the delayed growth responses to these growth factors by 16.0-, 5.8-, or 4.2-fold,

respectively. Neutralizing antibodies to PDGF-AA had no effect on these synergistic

responses and exogenous PDGF-AA did not increase growth responses to EGF, bFGF, or

PDGF-BB. In summary, TGF-beta induces marked delayed growth responses, alone and in

combination with EGF, bFGF or PDGF-BB, that are largely independent of PDGF-AA.

Research Article

Find the latest version:

http://jci.me/117199/pdf

TGF-,B

Promotes Proliferation

of Cultured SMC

via Both

PDGF-AA-dependent

and

PDGF-AA-independent

Mechanisms

GeorgeA.Stouffer* and GaryK.

Owens*"

Departments of*Medicine and

WMolecular

Physiologyand CellularBiophysics, UniversityofVirginiaSchoolofMedicine, Charlottesville, Virginia 22908

Abstract

Transforming growth

_factor-,61

(TGF-,81)

has beenimplicated

inmediating smooth musclecell(SMC)growthaftervascular

injury. Studies examining

TGF-fi-induced

growthofcultured

SMChaveidentified onlymodestmitogenic effects whichare

largely dependent onautocrine productionofplatelet-derived

growthfactor-AA (PDGF-AA).Recentstudies havesuggested,

however, that

_TGF-#t

also may have delayed growth effects independent of PDGF-AA. Theaims ofthe present studies

wereto examine theeffects of

_TGF-#t

on chronic growth re-sponses of cultured SMC. Results demonstrated that

_TGF-ft

elicitedadelayedgrowthresponse(24 fold increasein 3H-TdR

incorp. from 48-72 h) and enhanced SMC production of

PDGF-AA (eightfold increaseat 24 h).Neutralizing antibodies

to PDGF-AA, however, inhibited only 10-40% of delayed

TGF-ft-induced

growth.Co-treatment with

_TGF-ft

transiently delayed epidermal growth factor (EGF)-, basic fibroblast growth factor

(bFGF)-,

or PDGF-BB-induced entry into S phase but enhanced the delayed growth responses to these growth factorsby16.0-,5.8-, or4.2-fold, respectively. Neutral-izing antibodiesto PDGF-AA had noeffecton thesesynergistic

responses and exogenous PDGF-AA did notincrease growth responsestoEGF, bFGF, orPDGF-BB.Insummary,

TGF-ft

induces marked delayed growth responses, aloneand in combi-nationwith EGF,bFGForPDGF-BB, thatarelargely

indepen-dentofPDGF-AA.(J. Clin.Invest. 1994.93:2048-2055.)Key words: smooth muscle cells- transforming growth

factor-a

-epidermal growthfactor *basic fibroblast growth factor* plate-let-derivedgrowthfactor

Introduction

Smooth muscle cell

(SMG)1 proliferation

playsa major rolein

the vascularreparative response after mechanical injury. There

isevidencethat avarietyoffactors regulate thisSMCgrowth including angiotensin-Il (1, 2), basic fibroblast growth factor

Addressallcorrespondenceto Gary K.Owens, Departmentof

Molecu-larPhysiologyandCellular Biophysics,Box449, UniversityofVirginia School of Medicine, Charlottesville,VA22908.

Receivedfor publication23September 1993.

1.Abbreviationsused in this paper: bFGF, basicfibroblastgrowth

fac-tor;CM, conditioned media;MLE, mink lungepithelial;SFM,

serum-free medium; SHR, spontaneously hypertensive rats; SMC,smooth muscle cell.

(bFGF) (3), platelet-derived

growth

factor

(PDGF) (4, 5),

and

transforming growth factor-(il

(TGF-,B1) (6).

TGF-,B1 isa 25-kD

_growth

factor released

during

acute

phases

of

injury by

degranulating platelets (7)

and secreted

during

chronic

phases

oflesion

_{development by}

SMCs

(6)

andactivated

macrophages

(8). Majesky

etal.

(6)

demonstratedthat

TGF-f3l

mRNA levels

wereelevatedwithin 6 h ofballoon

embolectomy

catheter-in-duced

injury

ofratcarotidartery,andremained elevatedfor at

least 2 wk.

Immunohistochemical

studiesof vessels2wkafter

injury

revealed thata

large majority

ofintimalcells

expressed

TGF-#3I

protein

inanintracellular pattern consistentwith

ac-tive

_synthesis.

_Furthermore,

infusionofrecombinant

_TGF-031

into rats for 48h, 2 wk after carotidartery

injury,

increased

[3H]thymidine

labeling

indices ofintimal cells

by

fourfold.

These data

_strongly

suggest thatTGF-,Bis present, both

acutely

and

chronically,

and mayinfluence

growth

of SMC at sites of vascular

injury.

Our

understanding

ofthe role of

_TGF-#3

in

myointimal

le-sion formation after

injury

hasbeen confounded

by

in vitro observationsthatTGF-,Bcanboth inhibitaswellasstimulate

growth

ofcultured SMC. Several studies

(7, 9, 10),

including

one from this

laboratory (11),

have shown that TGF-f is a

potentinhibitorof

growth

factor-orserum-induced

prolifera-tion. In contrast, other

investigators

have found that TGF-,B stimulated SMC

proliferation. Majack

etal.

(12)

showed that

TGF-f3

stimulated

growth

ofpostconfluentratSMCand inhib-ited

growth

ofsubconfluentcells.

Battegay

etal.(13) found that

TGF-f

stimulated

growth

ofcultured human SMC in a

bi-modal,

concentration dependent manner.Both Majacketal. and

Battegay

et al. found that TGF-fl stimulation of

growth

wasassociated withincreased productionofPDGF-AA, mim-icked

by

treatment with exogenous PDGF-AA and

partially

inhibited

by

neutralizing antibodiestoPDGF-AA. The effects of

TGF-,3

onSMC

growth

weremodest in both ofthese

_studies;

maximal[3H]thymidine incorporation induced by TGF-,Bwas

200% of control in the studies ofMajacket al. and 400% of controlin the studies of

Battegay

etal.Basedonthe results of these

studies, TGF-3

iscurrently thoughttohavemodest ef-fects on SMC

proliferation

which are mediated

primarily

through

autocrineproductionofPDGF-AA.

Results ofrecentstudies havesuggestedthatTGF-p may also have

delayed

effectsonSMC growththatare notmediated

solely by

PDGF-AA

production.

Forexample,we foundthat

angiotensin-TI (Ang-II)

enhanced growthresponsesofSMC

de-rivedfrom spontaneously hypertensiverats(SHR)to

epider-mal

growth

factor

(EGF),

bFGF,and PDGF-BB and thatthese

effectsweredelayedanddependentonautocrine production of

TGF-f

(14). The effect of Ang-TI on growth responses to

PDGF-BBareunlikelytobe mediated viaautocrine

produc-tion of PDGF-AA since PDGFa receptors, the onlyknown

effectors ofPDGF-AA actions, should have been fully acti-vatedbythe saturating concentrations of PDGF-BB used in these experiments. Using cultured rabbit SMC, Janat and Liau

2048 Stouffer and Owens J.Clin. Invest.

©TheAmerican SocietyforClinical Investigation,Inc.

0021-9738/94/05/2048/08 $2.00

(15) showed that TGF-i3 alone had no effect on

[3H]thymidine

incorporation. However, similar to our observations, cotreat-ment with TGF-,B and PDGF-BB resulted in a growth effect which was 10-fold greater than that elicited by PDGF-BB alone. Thiseffectwasdelayed,with maximal effects observed 30 h after treatment with TGF-fl + PDGF-BB as opposed to 16 hafter treatment with PDGF-BB. Taken together, results of these studies suggest that

_{TGF-#l-induced}

increases in PDGF-AAproductioncannotexplain

_{TGF-f3-induced}

growth under allconditions, and that TGF-(3 haseffects on SMC growth that aredistinct from those previously described.

The aims ofthe present studies were: (a) to examine the effectsofexogenous

TGF-f3,

alone as well asin combination with EGF, bFGF, or PDGF-BB, on acute and chronic growth responsesof cultured SMC, and (b) to examine the contribu-tion of PDGF-AA inthisprocessbydetermining whether exog-enousPDGF-AA, alone or in combination with EGF, bFGF, or PDGF-BB, stimulated growth of SHR-derived SMC, and whetherneutralizing antibodiestoPDGF-AA inhibited

TGF-#-induced

growth

effects.

Methods

Smooth muscle cell culture. SMCs were isolated from aorta ofspontane-ouslyhypertensive or Sprague-Dawley rats by enzymatic digestion as previously described(16).The cells were cultured inmedium contain-ing a1:1 formulation of Dulbecco's modified Eagle's Medium (DME; GIBCO BRL, Gaithersburg, MD) and Ham's F12 medium (GIBCO BRL) supplementedwith10%fetal bovine serum (FBS, Hyclone Labo-ratories, Logan, UT), L-glutamine (0.68 mM; Sigma Chemical Co., St. Louis, MO), penicillin (100 U/ml), and streptomycin (100 sg/ml).

Cellswere harvested for passagingat subconfluence witha trypsin-EDTA(0.05%trypsin, 0.02% EDTA, GIBCO BRL) solution. Cell

cul-tures wereincubatedat370Cinahumidified atmosphere of5%COJ 95%air. SMCidentityandthepurityof cultureswasverified by immu-nocytochemicalanalysis using antibodies specific for smooth muscle a-actin(17).

SMC between passages 6 and 19wereplatedat3X I03

cells/cm2

in

serum containingmedia. Theyweregrown toconfluence and then growth arrested in a definedserumfree medium (SFM)containinga 1:1 formulation ofDME and Ham's F12 supplemented with

trans-ferrin (5 ,ug/ml, Sigma ChemicalCo.), insulin (5 x 10-7M, Sigma

ChemicalCo.),ascorbate(0.2mM,Sigma Chemical Co.),selenium(38

nM, Sigma Chemical Co.), glutamine, andpenicillin/streptomycin.

This SFM has been showntomaintain SMC inaquiescent,

non-cata-bolicstateandtopromoteexpressionofSMC-specificcontractile pro-teins(18).

Fibroblast cellculture. 3T3 Swiss albino mouse fibroblastswere

obtained from American Type Culture Collection (Rockville, MD)

andmaintained in DMEsupplementedwithL-glutamine,penicillin/

streptomycin,and10% FBS.Passagingwasatsubconfluenceusing simi-larmethodstothosedescribed above. The cellswereplatedat3X 103

cells/cm2 and grown to subconfluencein DME supplemented with

L-glutamine, penicillin/streptomycin and 10% FBS. The cells were

thengrowth arrested for 8-24hinDMEsupplementedwith L-gluta-mine, penicillin/streptomycinand 1%FBS.Conditioned media

experi-ments wereperformedasdescribed infigure legends.

RadioreceptorassayofPDGF concentration. This assay is

essen-tiallythesame asthatdescribedbyDiCorleto andBowen-Pope(19).

Briefly, subconfluent human foreskin fibroblasts wereplated in 1% humanplasma-derivedserum.Between 2 and 10 dlater,the cellswere

placedon anice tray and washed with 0.5mlof ice-coldbinding

me-dium (Dulbecco-Vogt medium without bicarbonate containing 25

mM Hepesbuffer, pH7.2, and bovineserumalbumin [BSA]).The mediumwasremoved andaliquotsofconditioned mediumfrom

TGF-3orvehicle-treated SMCwereadded in

_binding

medium. The cultures

wereincubatedat40Cwith gentlemixingfor 2 h. The test substance

was removed and the cellswererinsed with binding medium.

125[1

PDGF(0.2 ng)wasthen added and theincubationwascontinued for

two morehours. Themediawasremoved and thecells washed three times with 1mlcoldphosphate-bufferedsalinecontainingBSA. Bound

radioactivitywasdeterminedbysolubilizingthe cellswith 1% Triton

X-100inwatercontainingBSA.Nonspecific binding,measured in the presence ofatleasta 100-foldexcessof unlabeled crudegrowthfactor,

was <5% ofspecific binding.Standardcurveswerederivedusingwells

containingvariousamountsof unlabeled PDGF.

[3Hlthymidineincorporation.RelativeratesofDNAsynthesiswere

assessedbydeterminationof

[3H]thymidine

incorporation

into trichlo-roacetic acid(TCA)precipitablematerial. Cellswere_pulsedwith

[3H]-thymidine (1

gCi/ml,

6.7Ci/mmol; NewEngland Nuclear, Boston,

MA)andthen washedwithacalcium- and

magnesium-free

phosphate

buffered saline(PBS) (NaCl137mM,Na2HPO48.1mM,KCI 2.7_mM,

KH2PO41.5 mM, pH7.4).Thiswasfollowedby 10-minwasheswith 10%TCA,firstat40Candthenatroomtemperature. Cellswerethen dissolved in 1 NNaOH andplacedin

_Ready-Safe

scintillation fluid

(Beckman Instruments,Inc.,PaloAlto, CA).

Counting

wasdonewitha WallacLKBscintillationcounter.

Growth curves. Cultures werewashed with

PBS,

harvested with

trypsin-EDTA, dilutedwith 0.9% NaCl

_{(Columbia Diagnostics}

Inc., Springfield, VA)and counted

using

anElectrozone

Celloscope

(Parti-cle Data, Inc.,Elmhurst, IL) withorifice size of 95 _gmand

sample

volumeof 500Ml.

Reagents.TGF-13derived from human

platelets

wasobtained from R&DSystems(Minneapolis, MN).Recombinant human

EGF,

bFGF, PDGF-BB,and PDGF-AAwereobtainedfromUpstate

Biotechnology

Inc.(LakePlacid,NY).PDGF-AA

neutralizing

antibodies

_(pfa4)

were akindgiftof Dr. Michael Pech(F.Hoffmann-LaRoche, Basel,

Swit-zerland).Control antibodies for PDGF-AAneutralizing antibody

ex-perimentswerea-actin monoclonalantibodies(SigmaChemicalCo.)

PDGF-AApolyclonalneutralizing antibodywas_purchasedfrom Gen-zymeCorp.

Statistical

analysis.

Resultsare

presented

as mean ±standard de-viation unlessotherwise stated. One wayanalysisof variance followed

bytheNewman-Keuls'

multiple

rangetest wasusedtoanalyzedata.A

P valueof less thanorequalto0.05wasconsideredstatistically

signifi-cant.Triplicatewellswereanalyzedfor each

experiment

andeach

ex-periment was

performed independently

a minimum of three times.

Datashownarefromrepresentative experiments.

Results

TGF-3 induced

a

delayed

increase in

[3H]thymidine

incorpora-tion in

cultured

aorticSMC. Since

_TGF-#

levelsareincreased

in vascular lesions foratleast 2 wk after

_{injury (6),}

the initial

aim ofthesestudieswastodeterminethe effects of exogenous

TGF-i3

on

mitogenesis

of SHR-derived

SMC,

maintained ina

defined serum-free

medium,

at_acute,aswellas

_chronic,

time

points.

Previous studies

(14)

demonstrated that

_TGF-,3

wasnot

mitogenic

for SHR-derived SMC within theinitial 24 h after

treatment. In contrast, results of the

present

studies

demon-stratedthatTGF-13 elicitedaconcentration

dependent

increase

in

[3H]thymidine

incorporation

atboth 24-48 h

_(2-4-fold)

and 48-72h

_{(Fig. 1).}

SMC

_growth

wasobservedatconcentrations

of10

pM

andgreater withan

_ED50

for this effect of 20-30

pM.

Maximal effectswereobservedatconcentrationsof 40

_pM

or

greater. The

magnitude

of

_{TGF-,B-induced}

increases in

[3H]-thymidine

incorporation (48-72

hafter

treatment)

increased

withpassage number. The response

averaged

threefoldincells

at_{passages 5-9}versus49-foldincellsatpassages 15-19 times.

TGF-pl-stimulated production of

PDGF-AA

by

SHR-de-rived

SMC.Todetermine if

delayed

mitogenic

effectsof

120000r o

Ca ₀₃

00

o t a.

o st c am

)

c mi

a) 0 o

_L

c m

I1-r _'Z

N

400

300

200

100

a

0

0

0 a-o

0o

a -DD

(t) 0 C sC

>a "

Ic

I

_

veh -13 -12 -11 -10

TGF-ft Concentration [log M]

Figure 1.Concentration dependence of growth responsestoTGF-lI. SMCderived from SHRweregrowntoconfluence and growth arrested in SFM.4dlater the SFM waschangedandTGF-llI(various concentrations)orvehicle added. The SMCwerepulsed with

[3H]-thymidine(2,uCi/ml)48 haftertreatmentand[3H]thymidine incor-porationwasassayed 24 h laterasdescribed in Methods. The control group was treatedsimilarlybutwithout the addition of

TGF-fl1

or

vehicle.

aspreviouslyreported inSMC derived from Sprague-Dawley

rats(12) orhumans (13), PDGF-AA levelsweremeasuredin conditioned media from

_{TGF-fl-treated}

and vehicle-treated

SMCusing aradioreceptor competitionassay.Results demon-strated an eightfold increaseinPDGFconcentrationin condi-tioned media(CM)24 hafterTGF-,3ascomparedwithvehicle treatment (Fig. 2). PDGF-AA levelsremainedelevatedforat least 48 h(Fig. 2). Despite increasedlevelsofPDGF-AAinCM fromTGF-i3-treatedSMC (TGFCM),there was noincreasein

mitogenic activity for SMC in TGF CMascomparedwith CM

from vehicle treated SMC (CntCM) (datanotshown).Human

recombinant PDGF-AAalsostimulatedonly modestincreases

in

[3H]thymidine

incorporationandrequired concentrations

far in excessof that in TGF CM (Fig. 3). Results indicatethat

although TGF-3 stimulated production of PDGF-AA, these cells responded weakly to either endogenous or exogenous

1.2r C

0

0

c

a)

Q) c

0

0

UL-(.2

0-)

1.0k

I

0.8 F

0.6 L

0.4 F

0.2

-0.0

-24 48

Hours after treatment

Figure 2. PDGF concentration in conditioned media from

TGF-#lI

treatedSMC. SMC derived from SHR were grown to confluence and growth arrested in SFM. 4 d later the SFM was changed and

_TGF-#lI

(100 pM) or vehicle added. 24 or 48 h later, BSA(1 mg/ml)was added and CM removed and frozen. PDGF levels were measured using

ra-dioreceptorcompetition assays as described in Methods.(mControl;

n

TGF-,ll

treated).

80000k

veh 0.2 2 20 200

PDGF-AA Concentration [ng/ml]

Figure3. Concentrationdependenceof growth responses to

PDGF-AA.SMC derived from SHRweregrowntoconfluence andgrowth

arrestedin SFM. 4 d later the SFM was changed and PDGF-AA (various concentrations) or vehicle added. The SMC were pulsed with

[3H]thymidine(2 MCi/ml)attime of treatment and[3H]thymidine incorporationwasassayed 24 h later. The control group was treated similarly but without the addition of PDGF-AA or vehicle.

PDGF-AA. In contrast to effects on SHR-derived

SMC,

PDGF-AA isa potentmitogen for 3T3 fibroblasts which ex-press abundant PDGF-a receptors (20). TGF CM had in-creasedmitogenic activityfor 3T3 cells comparedtoCnt CM thatwaspartially inhibited byamonoclonalantibody

(desig-natedpfa4) that has previously been showntoneutralize rat

PDGF-AA (21) (Fig. 4).

Delayedmitogenic

effect

stimulated by TGF-3waspartially mediated by autocrine production of PDGF-AA. Treatment

withpfa4 antibodyinhibitedonly 10-40% of

TGF-f3-induced

mitogenesismeasured48-72 h aftertreatment(Fig. 5).Similar

resultswereobtained with apolyclonal antibody preparation

250000

0 0 A.-_

0

0Q

Is-a)

C-.-_

'-3

>I

Io

c

E

a0

0) 0 c

N

0

CL 200000

150000

100000

50000

Figure4. Effect of PDGF-AAneutralizing antibodyonmitogenic activity in conditioned media fromTGF-#l-treatedSMC. SMC de-rived from SHR were grown toconfluenceandgrowth arrested in SFM. 4 d later the SFM waschangedand

_TGF-P

I(100pM)orvehicle added. 48 h later, CM was harvested byaddingBSA (1mg/ml)and removing CM from SMC. CM was added to 3T3 cells which had been grown toconfluence and growth arrested for 24 h. PDGF-AA

neu-tralizing antibody (pfa4; 10,g/ml)orcontrolantibody (Cnt Ab)were added. SMC were pulsed with[3Hjthymidine(2

_IACi/ml)

8 h after addition of CM and[3H]thymidineincorporation assayed16 hour later.(veh CM, conditioned media from vehicle treatedSMC; TGF-(3

CM,conditioned media from

TGF-Itl

treatedSMC;Ab, antibody).

(X) conditionedmedia;(i)CM+PDGF- AAantibody;(8)CM + controlantibody.

2050 Stoufferand Owens

2000 r

0 0 -.

0

0

0) o

-c ._0

>1

I :-,

c

Q)

E

a

-C N fl

5000F

2500

vehicle _{TGF-# TGF-fl TGF-fl}

+ +

PDGF Ab Cnt Ab

Figure 5. Effect of PDGF-AA neutralizing antibody on

TGF-fl-in-ducedSMC mitogenesis. SMC derived from SHR were grown to confluence and growth arrested in SFM. 4 d later the SFM was changed and

_TGF-flI

(100 pM) or vehicle plus or minus PDGF-AA neutralizing antibody (pfa4; 10

_gg/ml)

orcontrol antibody were added. TheSMC were pulsed with[3H]thymidine(2,Ci/ml)48 h after treatment and [3H]thymidine incorporation assayed 24 h later.

raisedin goats which has been shown to neutralize rat

PDGF-AA(22) (data not shown). To ensure that sufficient

neutraliz-ing antibodywas present throughout the experiment, the pfa4

antibodywas used at concentrations 33 fold greater than the IC50 whichblocked binding of PDGF-AA (2 ng/ml) to NIH 3T3 cells (22a). Additionally, control studies revealed that: (a)

dailyaddition ofneutralizing antibody had similar effects on

TGF-f3-induced

mitogenesisasdidonetime addition (data not

shown),and

(b)

sufficient antibodywaspresent in CM at 48 h

to neutralize 1 ng/ml ofratPDGF-AA (obtained from SMC

CM)or 5ng/mlofhuman recombinant PDGF-AA (based on assays of mitogenic activity for 3T3 cells). Taken together,

thesestudies demonstratethatTGF-f3enhancedSHR-derived SMC production ofPDGF-AA but that thiseffect was only

partially

responsible for delayedgrowth effects stimulatedby

TGF-j3.

TGF-/t

delayedEGF-, bFGF-,orPDGF-BB-inducedentry intoS

phase

inconfluent, growth-arrested SMC. Saltis et al.

(23)havepreviouslyshown that low concentrations ofTGF-,3 (40 pM) modestly enhanced growthresponses ofSHR-derived SMCtoEGF, bFGF, PDGF-AB,orPDGF-BBwithin the

ini-tial48 h oftreatment. The aimofthe present studieswas to

determinetheeffects ofhigher concentrations of

TGF-#

(100 pM)ongrowthresponsestoEGF, bFGF,orPDGF-BB,atboth

acute and delayed time points. Results demonstrated that

TGF-13 (100 pM) inhibited EGF-, bFGF-, or PDGF-BB-in-duced

[3H]thymidine

incorporation by 71, 61,and 77%,

respec-tively

when measured between 15 and 16 hafter treatment

(Fig. 6). This effectwastransient, however, in that TGF-,3 had

no effect on EGF-, bFGF-, or PDGF-BB-induced

[3H]-thymidine

incorporation measuredat23-24 h(Fig. 6)or0-24

h(Fig. 7). Taken together, thesedatashow that

_TGF-,#

(100 pM) delayedEGF-,bFGF-,orPDGF-BB-inducedentry into S

phasebutdidnothaveanoverallinhibitory effectongrowth

responsesofSMCduringthe initial 24 haftertreatment.

TGF-f3

induced a delayed, synergistic increase ingrowth

responses

of

SMCtoEGF,

bFGF,

orPDGF-BB. Subsequent studies examinedtheeffects ofTGF-#ongrowthresponsesto

EGF, bFGF,orPDGF-BB at latertimepoints.Results

demon-C

o

0

O

0

0 0

0

L.

-I -

.-I

1500F

1000 F

500F

15-16

0

__gTGF-a

EmlEGF

_ EGF+TGF-f

_M_ bFGF

M bFGF+TGF-#

qPDGF-BB

E PDGF-BB+TGF-$

23-24

Time of H-thymidine pulse (h)

Figure 6. Effects of

_TGF-#I

on acute growth responses of SMCto

EGF, bFGF, or PDGF-BB. SMC derived from SHR were grown to confluence and growth arrested in SFM.4d later the SFMwas

changed andTGF-fll (100 pM) or vehicle plusorminus EGF (1.7 nM),bFGF (1.25 nM) or PDGF-BB (0.7 nM) added. The SMCwere

pulsed with[3H]thymidine(2MCi/ml)ateither15 or23hafter treat-ment and[3H]thymidineincorporation assayed 1 hour later. The control groupwastreatedsimilarly but without the addition of growth factors.

strated that co-treatmentwithTGF-j (100 pM) markedly in-creasedEGF-, bFGF-,orPDGF-BB-induced growthresponses measuredfrom48 to 72 haftertreatment(Fig.7). For example,

TGF-(3 increased the growth responses to EGF, bFGF, or

PDGF-BB by 16.0-, 5.8-, or4.2-fold, respectively,ascompared withtreatmentwithEGF, bFGF, or PDGF-BB, alone.

TGF-f3-inducedincreasesin[3H]thymidine incorporationwere

accom-panied by corresponding increases in cell number (Fig. 8). Treatmentwith

TGF-f3,

EGF, bFGF, or PDGF-BB alone

re-sulted in28, 50, 36,or89% increasesincell number,

respec-tively,

7dafterasingletreatment.Co-treatment with TGF-f3 andEGF, bFGF, orPDGF-BBincreasedcell numberby 212,

117,or188%,

respectively,

compared withcontrols. Consistent

c

0

O 0

Q) c

0)

C-I

6000

3000

E TGF-P

EM EGF _ EGF+TGF-P

M bFGF

M bFGF+TGF-#

PDGF-BB

__g PDGF-BB+TGF-P

0-24 48-72 Time of _H-thymidine pulse (h)

Figure 7.Delayed effects of

TGF-#3I

onSMCgrowthresponsesto

EGF, bFGForPDGF-BB.SMCderived from SHRweregrownto

confluence andgrowtharrested in SFM.4dlaterthe SFMwas

changed andTGF-,B1 (100 pM)orvehicleplusorminus EGF(1.7

nM), bFGF (1.25 nM)orPDGF-BB (0.7nM) added.TheSMCwere

pulsed with[3H]thymidine(2,Ci/ml)ateither 0or48hafter

treat-mentand

[3H]thymidine

incorporation assayed24 h later.The

con-trol groupwastreatedsimilarlybutwithout the addition ofgrowth

factors.

7500 F

0

C:

U

'Z

350 _11c

aE

F

fl-c

0s lo

TGF+EGF

TGF+PDGF-BB

v TGF+bFGF

...-.--.-.. PDGF-BB

.EGF

bFGF

...TGF

c-o

0

0 Cl

a1)

._0

>E

I-

AC-150

3 5

Days after growth factor pulse

7

Figure8.Delayed effects of

TGF-#

I onEGF-,bFGF-,or

PDGF-BB-induced SMC proliferation. SMC derived fromSHRweregrown toconfluenceandgrowtharrested in SFM. 4 d later the SFMwas

changed andTGF-,l (100pM)orvehicleplusorminusEGF (1.7

nM),bFGF(1.25 nM)orPDGF-BB (0.7 nM) added. The SMCwere

trypsinized and cellcountsdoneat3,5,or7d. The controlgroup wastreatedsimilarlybutwithoutthe addition of growth factors.

with[3H]thymidine incorporation data, the largest increase in cellnumberoccurred between three and five days after treat-ment.

TGF-,3 induced increases in growth responses to EGF, bFGF, or PDGF-BB were concentration dependent with

ef-fectsdependentonTGF-13 concentrationsgreaterthan 10pM (Fig. 9). Dailyaddition of TGF-f (100 pM)didnotaltergrowth

responsestoTGF-fi orTGF-j3 +EGF,asopposedtoonetime

addition, suggesting that delayed growthresponsestoTGF-13

werenotdependentonchanging TGF-,3 concentrations (data

notshown).

TGF-f$-induced increases in growth responses to EGF, bFGF, orPDGF-BB werenot mediated byautocrine produc-tionofPDGF-AA. To determine whether autocrine production ofPDGF-AAwasrequired for the synergistic growthresponses

250000r

° _

O _)

o 0

0o P

L..

-°i

Vx

c a

I-V

Q) 3

I

r-200000k

150000F

100000

50000

TGF-#+ PDGF-BB

// TGF-fl+ bFGF

/ TGF-P+ EGF

f v - I_

. /

../

0TGF-30

0 -13 -12 -11 -10

TGF-g1

Concentration [log M]

Figure9.Concentrationdependence oftheeffectsof

_TGF-f31

on

growthresponsestoEGF, bFGF,orPDGF-BB. SMC derived from

SHRweregrowntoconfluenceandgrowth arrested in SFM.4 d later

the SFMwaschanged and TGF-13I (variousconcentrations)orvehicle

plusorminus EGF(1.7 nM), bFGF (1.25nMOorPDGF-BB 0.7

nM) added. The SMCwerepulsed with [3H]thymidine(2gCi/ml)48

haftertreatmentand[3H]thymidineincorporationwasassayed 24

hlaterasdescribedin Methods. The controlgroup wastreated

simi-larlybutwithout the addition ofgrowth factors.

EGF bFGF PDGF-BB

Figure10.Effectsofneutralizing antibodyto PDGF-AAon

TGF-#31

synergismwithEGF,bFGForPDGF-BB. SMC derived from SHR

weregrowntoconfluenceandgrowtharrestedin SFM. 4 d later the

SFMwaschangedandTGF-,Bl (100 pM)orvehicleplusorminus

EGF(1.7 nM),bFGF(1.25 nM),orPDGF-BB(0.7 nM)added.

PDGF-AAneutralizing antibody (pfa4; 10gg/ml)orcontrolantibody

wereaddedtothe indicatedgroups.TheSMCwerepulsedwith

[3H]-thymidine (2 GCi/ml)48haftertreatment and[3H]thymidine

incor-porationassayed24hlater.The controlgroupwastreatedsimilarly

but without theaddition ofgrowthfactors.

seenwithTGF-,3 +EGF, TGF-,3 +bFGF,orTGF-(3 +

PDGF-BB, the effects of PDGF-AA neutralizingantibodiesonthese

growth responses wereexamined. Results demonstrated that

neutralizing antibodiestoPDGF-AA didnotinhibitsynergistic growthresponsestoTGF-lB+EGF, TGF-,3 +bFGF,orTGF-(3

+PDGF-BB(Fig. 10). Conditioned media from TGF-,3 treated SMC did not increase growth responses to EGF (data not

shown) andtreatmentwithexogenousPDGF-AA didnot

en-hancegrowthresponsestoEGF, bFGForPDGF-BB(Fig. 11).

TGF-f3-enhanced

growth responses toEGFin SMC

pri-marylineswhichareunresponsivetoexogenousPDGF-AA. We

have identifiedanumberof SMCprimarylinesderived from

Sprague-DawleyratsthatdonotexpressPDGFa_receptorsand

in which PDGF-AA elicits no measurable growth response

(22). These cellswereusedtofurtherinvestigate whether

TGF-C

0 ,

4

O °

0 0

cn

(a) 0

,1

>-'8

= atm

2000r

M +PDGFveh

m +PDGF-AA _T

1500F

1000F

500k

0

EGF bFGF PDGF-BB

Figure11.Effectsof PDGF-AAongrowthresponsesof SMCtoEGF,

bFGForPDGF-BB. SMCderived fromSHRweregrownto

con-fluence andgrowtharrested inSFM. 4 d latertheSFMwaschanged

and PDGF-AA(0.7 nM)orvehicleplusorminus EGF(1.7 nM),

bFGF(1.25 nM),orPDGF-BB(0.7 nM)added. The SMCwere

pulsedwith[3H]thymidine (2,Ci/ml)20 haftertreatmentand [3HJ-thymidine incorporationwasassayed20 hlater. Thecontrolgroup wastreatedsimilarlybutwithout theaddition of growthfactors.

300~

)-), J 0

_ 0

C)

200~

3000

.

-0

(- '6

C- V)

a1) (N

.0

-2

I- 0

2000

k

1000F

GF-//-//

TGF-fl

TG+

0

EGF

Figure 12. Delayed effects of

TGF-3l

on SMC growth responsesto

EGF in a SMC line which was unresponsive to PDGF-AA. SMC de-rived from Sprague-Dawley rats were grownto confluence and growth

arrested in SFM. 4 d later the SFM was changed and

_TGF-fl1

(100

pM) or vehicle plus or minus EGF (1.7 nM) added. TheSMC were

pulsed with

[3H]thymidine

(2

gCi/ml)

48 h after treatment and [3H1-thymidine incorporation assayed 24 h later. Thecontrol group was

treated similarly butwithout the addition of growth factors.

#

enhanced growth was dependent on PDGF-AA. Results dem-onstrated that exogenous PDGF-AA had no effect on

[3H]-thymidine incorporation from 0 to 72 h after treatment in these cells (data not shown). In contrast, TGF-f3 or EGF stimulated two- or sevenfold increases in[3H]thymidineincorporation re-spectively, between 48 and 72 h and co-treatment with

_TGF-0

and EGF stimulated a 24-fold increase in [3H]thymidine incor-poration (Fig. 12) as well as a significant increase in cell num-ber (data not shown).

Discussion

TGF-#

is a multifunctional growth factor which is present, both acutely and chronically, at sites of vascular injury. Effects ofendogenous

_TGF-#

on SMC in injured vessels are unknown, although infusion of TGF-13 2 wk after vascular injury has been shown to promote SMC mitogenesis (6). Studies to date have shown that TGF-(3 stimulates modest increases in [3H]-thymidine incorporation of cultured SMC which is largely de-pendent on autocrine production ofPDGF-AA (12, 13). Re-sults of the present studies extend these earlier findings by dem-onstrating that higher concentrations of TGF-(3 also have pronounced, delayed effects on SMC growth. TGF-(3 (100 pM)-induced an average 24- (range 3-49) fold increase in [3H]thymidine incorporation when measured 48-72 h after treatment. This effect on SMC growth is much larger than ob-served in prior studies (12, 13). Additionally, TGF-(3 stimu-lated a marked delayed growth response to EGF, bFGF, or PDGF-BB in SHR-derived SMC and to EGF in Sprague-Daw-ley-derived SMC. These data areconsistent with earlier studies demonstrating that

_TGF-#

enhanced growth responses to EGF in bovine SMC (7), to bFGF or PDGF in rat SMC (12) and to PDGF-BB in rabbit SMC (15). Similarly, in SHR-derived SMC, Saltis et al. (23) showed that, while

_TGF-#

(40pM) alone did not stimulate an increase in cell number, treatment with

TGF-#

increased the proliferative response to EGF or bFGF by

- _{160% and the proliferative response to AB or}

PDGF-BB by - 60%. The concentrations of

TGF-#

used in those

studies was less, and the onset of action earlier, than in the

present studies. Taken together, these studies suggest that

TGF-f effects on growthof culturedratSMC are concentra-tion dependentwith arelativelymodesteffectseenwithin48 h of

_TGF-#

treatment and alarger effect seen within 72 h of

treatment atTGF-(3 concentrations of 100pMorgreater. Resultsof the presentstudiesdemonstrate that

TGF-,3

pro-motes SMC growth via both PDGF-AA-dependent and

PDGF-AA-independentmechanisms.Consistent with earlier

studiesin rat (12) or human SMC(13),wefound thatTGF-/3

increased production of PDGF-AA inSHR-derived SMC,that exogenous PDGF-AA wasmitogenic for these cells and that neutralizingantibodiestoPDGF-AA

partially

inhibited TGF-,B-induced growth. Novelfindings of thepresentstudieswere

thatTGF-(3haddelayed growth

promoting

activity,

both alone and incombination with EGF, bFGF,orPDGF-BB and that autocrine productionofPDGF-AAplayedaminorrole in

me-diating theseeffects. Evidence

supporting

thelatterconclusion

includes: (a) neutralizing antibodies (either monoclonal or

polyclonal) toPDGF-AAinhibited,atmost,

only

40%of

TGF-,3-induced

growth and had no effect on

TGF-j

enhanced

growth responses to EGF,bFGF, orPDGF-BB measured be-tween 48 and 72 h;(b)exogenousPDGF-AAstimulated

only

a

modest increase in [3H]thymidine incorporation anddid not

enhance growth responses to EGF, bFGF or

PDGF-BB;

(c)

conditioned media from

TGF-f3-treated

SMC did not have increasedmitogenicactivityforculturedSMC,aloneorin

com-binationwith EGF,bFGF,orPDGF-BB,when

compared

with

conditioned mediafrom vehicle treated

SMC;

and

(d)

TGF-f enhancedgrowth responsestoEGFina

primary

SMC culture

linewhich did not respondmitogenically toPDGF-AA. There are several potential mechanisms by which

TGF-#

maystimulate SMC growthindependent of PDGF-AA

produc-tion. Onepossibility is that

TGF-(3

maystimulate SMC

produc-tionof secondaryfactors,inadditionto

PDGF-AA,

whichhave growth promoting activity. Results demonstrated that there was increased mitogenicactivity for 3T3,butnot

SMC,

in TGF CMcomparedwithCntCMwhichwas

only

partially

inhibited

byneutralizing antibodies toPDGF-AA. This suggests that if secondaryfactorsare produced which mediate TGF-(3 effects

on SMC, they were nottransferrable in conditioned media. A

second possibility is that TGF-( may increase SMC

growth

factor receptor expression. Janat and Liau

(15)

showed that

TGF-#

increased PDGF

(-receptor

mRNA inrabbit

SMC,

con-current with TGF-,B enhancement of

growth

responses to

PDGF-BB. However, to explain the current results

by

this

mechanism wouldrequire TGF-,Btreatment to

simultaneously

increaseexpressionof EGF,bFGF,and PDGF receptors.

Addi-tionally, co-treatmentwith TGF-f (100

pM)

has

delayed,

syn-ergistic effects on a-thrombin- or

Ang-TI-induced

growth

(23a)implyingthatTGF-,B wouldalsohavetoincrease expres-sionofa-thrombinandAng-IIreceptors. Athird

possibility

is

that TGF-,B enhancement ofprotooncogene

expression

may explain

TGF-f3

actions. Recent studies

(15)

have shown that co-treatmentwith TGF-,B and EGF hasa

synergistic

effecton

c-mycexpression.Theseeffects have, however, been observed within 8 hof treatment and there isnoevidencetosuggest that asimilareffect occurs at

delayed

time

points.

Alternative explanations for the

synergistic

growth

growthfactors. TGF-,3 has been shown to increase SMC

pro-duction of extracellular matrix and this mayaccountfor the delayed growth promoting activity through alterations in

cell-matrixinteractions. Inparticular, TGF-,3 has beenshownto increaseexpression of fibronectin, collagen, and elastic fiber proteins.IgnotzandMassague(24)demonstratedthat

TGF-f3-inducedanchorage-independent growthandenhanced

expres-sionof fibronectin in cultured chick embryo fibroblasts. Fur-thermore, inhibitors of fibronectin binding blocked, and exogenous fibronectinmimicked,

_TGF-#

induction of

anchor-age-independent growth.InSMC,Majacketal.(12)and Janat and Liau (15) demonstrated that

TGF-#

enhanced SMC

ex-pression of thrombospondin. Majack et al. (12) also showed

that co-treatmentwiththrombospondinand EGF resulted in a

synergistic proliferative

response. TGF-f3 hasbeen shownto

enhanceSMCexpressionof as and

f3

classes ofintegrin (25,

26), but these effects have not been directly linked to SMC

growth.TGF-3has alsobeenshown toregulate levels of

extra-cellular matrix degradation by reducing the synthesis of en-zymesinvolved inmatrix proteolysis andincreasing levels of proteolytic inhibitors(e.g.,plasminogen activatorinhibitor-l) (27).

Resultsof thepresentstudies addtothe

growing

evidence thatTGF-f3 stimulates SMC growth in vitro

(12, 13),

aswellas

invivo (6). Thereis, however, extensive evidencethatTGF-fl

canalsoinhibit growth ofculturedSMC (7,9, 10). The mecha-nismsresponsible for the bifunctional growth effects of TGF-13

on SMC are not known. However, results ofrecent studies suggestthat themultiplicity of

TGF-,3

functionsmaybedueto

differential expression of TGF-3receptors(28,29, 30). Chenet

al.(30) demonstrated thatoverexpression ofatruncatedtype II receptor mutantactedinadominant negative fashion in mink

lungepithelial(MLE) cellsresulting in completelossoftype II receptor binding and

TGF-f3-induced

growth inhibition. In contrast,therewas nochange intype Ibindingor in

TGF-fl-inducedincreases in

fibronectin, plasminogen

activator

inhibi-tor 1 or

jun

B mRNA

expression.

Massague and coworkers (29),usingmutant MLEcells,have shown that type I receptors arenecessaryfor

_{TGF-f3-induced}

growth inhibition. Taken

to-gether,thesestudiessuggestthat bothtype Iandtype II recep-tors are

required

to mediate the growth

inhibitory

effect of

TGF-13

inMLEcells,whereas type I receptorsalonecan

medi-ate other effects of

TGF-,3

including

matrix

production.

In

SMC, Goodman and Majack (31) demonstrated that

con-fluent,

growtharrested SMCexpressedlowerlevelsof 65 kD

(now referredto astype Ireceptors),and higher levels of 280

kD(typeIII receptors) binding proteins relativeto proliferat-ing, subconfluent SMC. Levels of 85-kD (type II receptors)

binding protein expression

weresimilar in bothgroups.TGF-,B inhibitedgrowthofratSMC atsubconfluent densitiesbut

po-tentiated growth ofSMC atconfluent densities suggestingthat theseeffectsweremediatedbydifferential

TGF-f3-binding

pro-tein expression. Taken together, these studies suggest that

TGF-,B receptorexpression could play a role in determining growth responses of cultured cells to

TGF-3

and that this will be animportantarea of research for furtherunderstanding of

TGF-# effectson SMC growth.

In the present studies, TGF-,B delayed EGF-, bFGF-, or

PDGF-BB-induced

entry of SHR-derivedSMC into S phase.

Co-treatment with TGF-,B markedly inhibited EGF-, bFGF-,

or

PDGF-BB-induced

[3H]thymidine

incorporation when

measured 15-16 h after treatment. Thiseffect was transient,

however,

and

TGF-,3

hadnoeffecton

EGF-,

bFGF-,

or

PDGF-BB-inducedgrowthover the initial 24 h oftreatment.

Revers-ible growthinhibitory effects ofTGF-j3 have beenseen inSMC, epithelial, endothelial and fibroblast cultures. These effects have been shown to be due to TGF-1 interference with the

abilityof cells to traverse a stage in lateG. phase. Usingbovine orhumanSMC,Reddyand Haure(32)found thatTGF-fl inhi-bition of serum stimulated growth resulted from reversible arrest at apointinthe cellcyclelocated 1-2 h from Sphase.A moreprolongedgrowthinhibitoryactivityof

TGF-#

has also been observed in serum stimulatedSMC(1 1). Potential insight into cellular events associated with

_{TGF-f3-induced}

growth arrest wasprovided by recentstudiesbyKoffetal.(33) They

showed that TGF-(i treated MLE cells failed tostablyassemble

cyclin E-Cdk2 complexes or accumulate cyclin E-associated

kinase activity.

In summary, thesestudiesprovide further data regarding

the effects ofTGF-,3onSMCgrowth.TGF-3(100 pM)elicited

marked,delayed growthresponses,both alone and in combina-tion with EGF, bFGFor PDGF-BB. TGF-fl enhanced

SHR-derived SMC production of PDGF-AA which partially

me-diated TGF-(3-induced growth. However, TGF-(3 markedly

en-hancedgrowthresponses toEGF,bFGF,orPDGF-BB through mechanisms thatwerelargely

independent

of PDGF-AA. Fur-ther studies will be required to

identify

the mechanisms

whereby prolonged treatment with

TGF-,3

enhances SMC growth resposiveness in cultured SMC and to determine

whether similar effects occurin vivo.

Acknowledgements

This workwassupportedin partbyPublic HealthServicegrants RO

I-HL-38854 and POI-HL-19242 from the National Institutes of Health and by National Research Service Award Training Grant T32

HL07355.

References

1.Powell,J.S.,J. P.Clozel,R. K.Muller,H.Kuhn,F._Hefti,M.Hosang,and H. R.Baumgartner. 1989. Inhibitorsofangiotensin-convertingenzymeprevent

myointimal proliferation aftervascularinjury.Science(Wash. DC). 245:186-188.

2.Daemen,M. J. A.P.,D.M.Lombardi,F. T.Bosman,andS.M.Schwartz. 1991.AngiotensinIIinduces smooth muscle cellproliferationin the normal and

injuredratarterial wall. Circ.Res.68:450-456.

3.Lindner,V.,D. A.Lappi,A.Baird,R.A.Majack,andM. A.Reidy.1991. Roleofbasic fibroblastgrowthfactor in vascular lesion formation. Circ. Res. 68:106-113.

4.Majesky,M.W., M.A.Reidy,D. F.Bowen-Pope,C. E.Hart,J. N.Wilcox,

and S.M.Schwartz. 1990. PDGFligandand receptorgeneexpression during

repairof arterialinjury.J.CellBiol.111:2149-2158.

5.Ferns, G.A.A., E.W.Raines,K. H.Sprugel,A.S.Motani,M. A.Reidy,

and R. Ross.1991.Inhibition of neointimal smooth muscleaccumulationafter

angioplasty byanantibodytoPDGF. Science(Wash. DC).253:1129-1132. 6.Majesky,M.W.,V.Lindner,D. R.Twardzik,S.M.Schwartz,andM. A.

Reidy. 1991. Production oftransforminggrowth factor#Iduringrepairof

arte-rialinjury.J.Clin.Invest.88:904-910.

7.Assoian, R.,andM. B.Sporn.1986.Type#transforming growthfactor in humanplatelets: releaseduring platelet degranulationandactiononvascular smooth muscle cells. J.CellBiol. 102:1217-1223.

8. Assoian,R. K.,B. E. Fleurdelys,H.C.Stevenson,P.J. Miller, D. K. Madtes, E. W.Raines,R.Ross, andM. B.Sporn.1987.Expressionand secretion

oftypebetatransforminggrowth factorbyactivatedhumanmacrophages.Proc.

Natl. Acad.Sci. USA. 84:6020-6024.

9.Majack,R.A.1987.Beta-typetransforminggrowthfactorspecifies

10. Ouchi, Y., J.Hirosumi, M. Watanabe, A. Hattori, T. Nakamura, and H. Orimo. 1988. Inhibitory effect of transforming growthfactor-a on epidermal growthfactor-induced proliferation of cultured rat aortic smooth muscle cells. Biochem. Biophys. Res. Commun. 157:301-307.

11. Owens, G. K., A. A. Geisterfer, Y. W. Yang, and A. Komoriya. 1988. Transforming growth factor-beta-induced growth inhibition and cellular hyper-trophyin cultured vascular smooth muscle cells.J. Cell Biol. 107:771-780.

12. Majack, R. A., M. W. Majesky, and L. V. Goodman. 1990. Role of PDGF-A expression in the control of vascular smooth muscle cell growth by transforming growth factor-beta. J. Cell Biol. I1 1:239-247.

13. Battegay, E. J., E. W. Raines, R. A. Seifert, D. F. Bowen-Pope, and R. Ross. 1990._TGF-#induces bimodal proliferation of connective tissue cells via complex controlof an autocrine PDGF loop. Cell. 63:515-524.

14.Stouffer, G. A., and G. K. Owens. 1992. Angiotensin II induced mitogene-sis of SHR derived cultured smooth muscle cells is dependent on transforming growthfactor-,8. Circ. Res. 70:820-828.

15. Janat, M.F., andG. Liau. 1992. Transforming growth factor ,1 is a powerful modulator of platelet-derived growth factor action in vascular smooth muscle cells.J. Cell. Physiol. 150:232-242.

16.Geisterfer, A. A., M. J. Peach, and G. K. Owens. 1988. Angiotensin II induces hypertrophy, not hyperplasia, of cultured rat aortic smooth muscle cells. Circ. Res. 62:749-756.

17.Skalli, O., P. Ropraz, A. Trzeciak, G. Benzonana, D. Gillessen, and G. Gabbiani. 1986.Amonoclonalantibody against a-smooth muscle actin: a new probefor smooth muscle differentiation.J.CellBiol.103:2787-2796.

18.Libby,P., and K. V.OBrien. 1983. Culture of quiescent arterial smooth musclecells in adefinedserum-freemedium.J. Cell.Physiol.115:217-223.

19.DiCoreleto, P., andD.Bowen-Pope. 1983. Cultured endothelial cells pro-duce aplatelet-derivedgrowth factor-likeprotein.Proc.Nadl.Acad.Sci. USA. 80:1919.

20.Seifert, R. A., C. E. Hart, P. E.Phillips,J. W.Forstrom, R.Ross, M. J. Murray, and D. F.Bowen-Pope.1989.Twodifferent subunits associateto create

isoform-specific platelet derived growth factor receptors. J. Biol. Chem.

264:8771-8778.

21.Stouffer,G. A., I. J.Sarembock, C.A.McNamara,L.W.Gimple,and G. K. Owens.1993.Thrombin-inducedmitogenesisof vascular SMC ispartially

mediated by autocrineproduction of PDGF-AA.Am. J. Physiol. 265:C806-811.

22.Stouffer,G. A.,R. T.Shimizu,M. B.Turla, and G.K.Owens. 1993.

Angiotensin-Il-orAVP-induced increases inprotein synthesisare notdependent

onautocrinesecretion ofPDGF-AA.Am.J.Physiol.264:C390-C395. 22a. Herren, B., K.A.Weyer,M.Rouge, P.Lotscher,and M. Pech. 1993.

Conservation in sequence and affinity of human and rodent PDGF ligands and receptors. Biochim.Biophys. Acta. 1173:294-302.

23.Saltis, J., A. Agrotis, and A. Bobik. 1992.TGF-,9lpotentiatesgroth factor-stimulatedproliferationof vascular smooth muscle cells in genetic hypertension. Am. J.Physiol.263(Cell. Physiol. 32):C420-C428.

23a.Stouffer, G. A.,I. J. Sarembock, L. W. Gimple, and G. K. Owens. 1994. Cotreatment with TGF-beta andangiotensin-II or alpha-thrombin induces a

syn-ergistic growth response in cultured smooth muscle cells. J. Cell. Biochem. 18:288a. (Abstr.)

24. Ignotz, R. A., and J. Massague. 1986.Transforminggrowthfactor-beta stimulates theexpression of fibronectin and collagen and their incorporation into theextracellular matrix. JBiol Chem. 261:4337-4345.

25. Janat, M. F., W. S. Argraves, and G. Liau. 1992. Regulationof vascular smooth muscle cellintegrin expressionbytransforminggrowth factor betaIand byplatelet-derivedgrowth factor-BB. J. Cell.Physiol. 151:588-595.

26. Basson, C. T., 0. Kocher, M. D. Basson, A.Asis,and J. A.Madri. 1992. Differential modulation of vascular cellintegrinand extracellular matrix expres-sion invitro by TGF-betaIcorrelates withreciprocal effectsoncellmigration.J.

Cell.Physiol. 153:118-128.

27.Reilly,C. F., and R. C. McFall. 1991. Platelet-derivedgrowth factor and

transforminggrowth factor-beta regulateplasminogenactivator inhibitor-I syn-thesis in vascular smooth muscle cells. J. Biol.Chem.266:9419-9427.

28. Ebner, R., R. H.Chen, L. Shum, S. Lawler,T. F.Zioncheck,A.Lee,A.R. Lopez,and R. Derynck.1993. CloningofatypeITGF-betareceptorand itseffect

onTGF-betabindingtothe type II receptor. Science(Wash. DC). 260:1344-1348.

29.Laiho, M., Weis,F.M.,Boyd, F. T., Ignotz, R. A., and J.Massague. 1991.

Responsivenesstotransforminggrowth facotr-beta(TGF-beta)restoredby ge-neticcomplementationbetween cells defective in TGF-beta receptorsIand II. J.

Bid. Chem.266:9108-9112.

30.Chen, R. H., R. Ebner, and R. Derynck. 1993. Inactivation of the typeII receptor reveals two receptorpathways for the diverse TGF-beta activities. Science(Wash. DC).260:1335-1338.

31.Goodman, L. V., andR. A.Majack.1989. Vascular smooth muscle cells express distincttransforming growthfactor-beta receptorphenotypesasa func-tionof celldensityin culture. JBiolChem. 264:5241-5244.

32.Reddy, K. B., and P. H. Howe. 1993.Transforming growthfactor

,B1-me-diated inhibition of smooth muscle cellproliferation is associatedwithalate G 1 cellcyclearrest.J.Cell.Physiol. 156:48-55.

33.Koff, A., M.Ohtsuki,K.Polyak, J. M. Roberts, and J. Massague. 1993. Negative regulation of GI in mammalian cells: inhibition ofcyclinE-dependent