The plasminogen activator/plasmin system

The plasminogen activator/plasmin system.

J D Vassalli, … , A P Sappino, D Belin

J Clin Invest. 1991;88(4):1067-1072. https://doi.org/10.1172/JCI115405.

Research Article

Perspectives

The Plasminogen Activator/Plasmin System

Jean-Dominique Vassalli, Andr6-Pascal Sappino,* and DominiqueBein*

Departments ofMorphology, *Medicine, and $Pathology, UniversityofGenevaMedicalSchool, 1211 Geneva4,Switzerland

Overthe last 10 years, our knowledge of extracellular

proteoly-sis has progressed dramatically. Differentenzymatic cascades cooperate toachieveextracellular matrix(ECM)Idegradation, and a numberof participant proteins have been characterized

and cloned. Physiological inhibitorshave been identified for mostofthese enzymes. Also, the concept of focused

proteoly-sis, through binding ofenzymes andinhibitors to specific

re-gions ofthe extracellular milieu, has received broad

experimen-tal support. Finally, thebiosynthesis of many of the relevant proteases andinhibitorshas been shown to be under the con-trol of hormones andgrowthfactors. Plasminogen activators

(PAs)andtheir inhibitors(PAIs) are thought to be key partici-pantsin the balance ofproteolyticandantiproteolyticactivities

that regulates matrixturnover.This articlesummarizes the

evi-dencethat supportsthis contention, discussesthe role of

PA-specific cell surface binding sites,and also draws attention to a numberof instances in whichthe presence of PAs cannot be

reconciled withanexclusivefunctionin ECM degradation.

The plasminogen activator/plasmin system

ENZYMES

PAs are serine proteasesoftryptic specificity. Two enzymes,

differing mostly in the domainorganization and function of theirnoncatalytic regions,have beenidentified inmammals: urokinase-typePA (uPA) andtissue-type PA(tPA) (1). They

arethe products of distinctgenes, andare secretedas single-chain (sc)proteins;whereassctPAisactive,scuPAis

essen-tially inactive (pro-uPA) (2). Cleavage of pro-uPA byplasmin,

kallikrein,Factor XIIa orcathepsinB(3)yieldsthe disulfide-linked two-chain active enzyme. Thetwo PAs have distinct

targeting determinants in their noncatalytic regions: the

"growth factor domain" of uPA directs thebindingofthe

en-zyme(and that ofpro-uPA)toaplasmamembranereceptor(4,

5),whereas other structuraldomainsin tPA(the"finger"

re-gion and the "kringles") allow itsbindingtofibrin and other

Itisimpossibletoselectasmallsetof references that wouldprovide

appropriate coverage of the entire field discussed. We have thus in-cludedreferencestoreviews andtorecentpapers thatcanbe usedto

tracebackearlier work. Weapologizetoallourcolleagueswhose

con-tributionsare notadequatelyreferenced,aswellas to ourreaders.

Receivedfor publicationIApril1991andinrevisedform28May

1991.

1.Abbreviations used in thispaper: ECM, extracellularmatrix; PA, plasminogen activator; PAI, plasminogen activator inhibitor,sc,single

chain;tc,two-chain; tPA, tissue-type PA; uPA, urokinase-typePA.

components of the ECM (6). These and perhaps additional

interactions, for instance with heparinlike molecules, could have adramatic effecton thefocusing ofPA-controlled

prote-olysis (7).Thedifferent extracellularaddressing ofthe two PAs suggeststhatthey play differentbiologicalroles.

Plasminogen is the prefered substrate forPAs, but other molecules may also be cleaved by one or the other PA; for

instance, avianuPA exerts aplasmin-independent effecton the morphologyof chick fibroblasts (8).Plasminogen ispresent in plasma and extracellularfluidsat a 1-2

_AM

concentration,in

therangeof theKmoftheactivationreaction. Itcanassociate withfibrinand other proteins vialysine-bindingsites located in

thekringles of its noncatalytic portion (9).The PA-catalyzed

cleavageof anArg-Val bond yieldsthe active enzymeplasmin, consisting oftwopolypeptide chains linked byadisulfide bond;

Arg-esterases other than PAs can activate plasminogen, but

only tPA anduPAappear todosounderphysiological condi-tions. Like PAs,plasmin belongstothe serineproteasefamily, and hastrypticspecificity; bycontrast toPAs,however, ithas a very broad spectrum ofsubstrates. Whereas the "classical" plasminsubstrateisfibrin, inthe contextofECMproteolysis it is clear thatmostothermatrixproteinscanalso becleaved by thisenzyme. Inaddition, plasmin isone of theactivators of metalloproteasesprecursors(10), and could thusplayapivotal role in ECMdegradation; whether the PA-plasminsystem

con-trolsmatrixturnoveris, however, debatable, because the

rate-limitingstepin suchenzymaticcascades may varyfrom situa-tiontosituation, andneed notbe themostproximal.Plasmin hasalsobeenfoundtoactivate latentforms of certain growth factors (1 1), and its role should hence not be considered as

limitedtodegradativeevents.PA-catalyzed activation of plas-minogen results inadramaticamplificationoftheproteolytic capacityofPA-producing cells, byrecruitment ofthevast

reser-voir of potential, broad spectrum proteolysisrepresented by

plasminogenin thepericellularenvironment.

INHIBITORS

Fourarginine-specific serine-proteaseinhibitors(Arg-serpins)

are ofparticular relevance to the PA/plasmin system (12).

Theirpreferredtargetenzyme(s),asdetermined bythekinetic

constantsof complexformation,aredifferent.

Plasminogen

activator inhibitor1(PAI-1)isthe

major

PAI

inplasma.IthashighaffinityfortPA,bothscand two-chain

(tc),aswellasforuPA. PAI-I is secretedas anactive

antipro-tease, butitrapidlyconverts toaninactive latentform,which

canbe reactivatedbyexposureto

chaotropic

agentsor

phospho-lipids; this latency isnotsharedbyother

serpins,

and its molecu-larbasis remainsunexplained. PAI-1isoften found associated

with plasma and ECM vitronectin, which stabilizesit in the

active conformation(13).Inaddition,vitronectin and

_heparin

affect thespecificity ofPAI-1, by

enhancing

its

reactivity

to-wardthrombin (14).

J. Clin. Invest.

©TheAmericanSocietyfor ClinicalInvestigation,Inc.

0021-9738/91/10/1067/06 $2.00

Plasminogenactivator inhibitor 2 (PAI-2) inhibits uPA and

also, albeit lessefficiently, tPA (tcbut not sc). The rate con-stantsofuPAand tPAinhibitionare not asfavorableas with

PAl-1,andthis raisesthepossibilitythat PAI-2 may control the

activity ofother, yetunidentified,enzymes. Inthiscontextit is intriguingthatPAI-2,in additiontobeinga secreted protein, is

also abundant in the cytosol of cells that synthesize it (15),

whereas PAs areexclusively secretoryproteins.

Protease-nexin I(PN-I)inhibitsuPA, plasmin, and throm-bin; its reactivity towards thrombin (butnottowards the other twoenzymes) isdramatically enhanced inthe presenceof

hepa-rin.The dual roleof PN-Ias aninhibitor of fibrindeposition andremovalsuggeststhat itmaybedesignedtodampen multi-pleenzymatic pathways which, if active, wouldaffect the

me-tabolism oftheextracellular milieu.

a2-antiplasministhe primary plasmin inhibitor in plasma.

Interestingly, ifthe lysine-binding sites of plasmin are

occu-pied, inhibitionby a2-antiplasmin is markedly reduced (16);

this is thoughtto localizethe activity of plasmin to specific

regions oftheextracellular environment, such as on fibrin or in

the immediate vicinity ofthe cellsurface.

CELL SURFACEBINDING SITES

Asharper image ofthebiological role of the PA/plasmin sys-temhascomefromobservations indicatingthatPAsand plas-minogencanbindtocellsurfaces.

uPAreceptor ispresent on the plasma membraneofa

num-berofcelltypes(4, 5).Itbinds with similar kinetics uPAand pro-uPA, formerly considered as secreted proteins acting in solution inextracellularfluids.Thisreceptorcomprisesat least onepolypeptidechain, which is inserted in the plasma

mem-branethroughaglycerophospholipidanchor(17). Binding

in-volves the"growth factor-like" domainofuPA, and

receptor-bound uPA remainsfunctionallyactiveatthe cell surface with

a half-life of 4-5 h(18).Recent studies(18, 19)demonstrate thatreceptor-bounduPAisnotprotected frominhibitionby

PAI- I orPAI-2.

Different cellular binding sitesfortPA have beendescribed. Someoftheseareclearly involved in the clearance(in

particu-larby the liver) of theenzyme, orof tPA-PAI complexes (20), whereasotherslocalize tPAtotheplasmamembrane, for

in-stance onendothelial cells andneurons(21, 22);theyare

differ-ent from the characterized uPA receptor. Because tPA also binds tocomponents of the ECM, positive identification of these cellular binding site(s) will be required to ascertain whetherthey are plasmamembrane

constituents.

Plasminogenand plasmin bind to different cell types,

in-cludingcells that alsoexpressuPAreceptors (9).

Binding

ap-pearstobetoproteinswithanexposed

carboxy-terminal lysine

residue,and recentresultsidentifyaplasma membrane form of

enolase as onemajor plasminogenreceptorsite(23).Thisand

otherobservationssuggestthat thelysine-bindingdomains of plasminogenareinvolved.

Cell

biology ofplasminogen

activation

REGULATION OF SYNTHESIS

The rangeofcell types that canproduceone(or more) ofthe

constituants of thePA/plasminsystemisextremely broad,and includes derivativesfrom the threegermlayers, cells from the male and female germ lines, aswell ascells from

extraem-bryonic tissues.Infact,itmaybethat,giventhecircumstances,

all cell types canproducePAs and/or PAIs. PAs and PAI-1 are

presentin blood plasma,andsoisPAI-2duringlatepregnancy.

PAI-l isalso amajorconstituant of certain ECM.However,

theseproteinsarebynomeansconstitutivecomponentsofthe

extracellular milieu:

expression

of the

PA/plasmin

systemisa highlydynamicaspectof the cellularphenotype.

Alargebodyof workhasexploredtheregulationof

expres-sionofPAs(1)andPAIs

(12),

inmanydifferentcell

lineages.

Mostoften, transcriptional mechanisms have been shownto

play adeterminingrole (24-26), althoughin atleastone

in-stance

regulation

isclearly

posttranscriptional (27).

The

impor-tantconcepttoemergefromthesestudiesis that

probably

all

signaltransduction pathwayscan affectthe

PA/plasmin

sys-tem.Somepathwayshavesimilar effectsin manydifferent cell

types:for instance

glucocorticoids

seem to

uniformly

decrease

uPAtranscription and mRNA levels, andoften increase

ex-pressionof thePAI-I gene. Bycontrast,agiven pathwaycan

haveopposite effects depending onthe celltype: cAMP and

relatedagents canincrease (26)ordecrease(28)uPA gene

tran-scription.Inthiscontext,the observation that thesamenuclear

factorcanactivateuPAandcollagenase

transcription

provides

a newexampleof

cooperation

betweenthe

plasmin

and

metal-loprotease cascades(24). The intricate regulationofPAand

PAIsynthesisallowsexpressionofthePA/plasminsystemtobe

preciselycontrolled intime,inamannerthatdependsupon the

endocrineorgrowthfactor balance. Onestrikingillustration of thisisprovided bytheeffectofbFGF and

TGF(B

onendothelial cells: bothagentsenhancePAI-I anduPA mRNA andprotein synthesis, but the kinetics and amplitude ofthe changes in

expression ofthetwogenes areremarkably different, sothat

the overall balance is tilted towards enhanced

proteolysis

in

responsetobFGF,and towardsantiproteolysisinresponseto

TGF,3

(29). Thisshift in theproteolyticbalanceof endothelial cellsprobably playsan

important

partinthe

_{TGFfl-mediated}

antiangiogenic

properties

of

cartilage

(30, 31). SECRETION

Itis ingeneralconsidered thatPAsand PAIs followthe consti-tutivesecretory pathway, i.e.,thatsecretionrapidlyfollows

syn-thesis

(with

the

exception

of the

cytosolic

form of PAI-2

[15]).

However,immunohistochemicalanalysisofuPAin themouse vasdeferenssuggeststhat theantigenispresentinapical

gran-ules

(32),

andcantherefore follow thepathwayfor

regulated

secretion.Also,in humanPMNs

expression

of uPA

activity

at

the cellsurfacecanbe

triggered

byexposuretosecretagogues

(33). Thus,

regulation

ofenzyme orinhibitor secretion may representyetanother controlpointfor

expression

ofthe

PA/

plasmin system. One aspect that hasnot yetreceived

appro-priateattentionis the

possible polarization

ofPAandPAI

se-cretionby

epithelial

cells(34).A

systematic study

of this issue

mayhelpelucidatethe

physiological

role(s)ofthe

PA/plasmin

systemin

epithelial

tissues.In

addition,

a

perturbation

of the

polar release ofPAs or PAIs could be detrimental to tissue architectureor

function,

and

pathogenic;

for instance the

baso-lateral release ofa PAnormally destined for

apical

secretion

may result in

proteolytic

damageto the

underlying

basement

membrane and henceplayapartin the

acquisition

of invasive

properties by

malignant epithelial

cells. CELL SURFACE EXPRESSION: uPARECEPTOR

The

biological

consequences of

plasminogen

activation may

differas afunction of the type ofPA

produced:

for

instance,

catalyti-callyequivalentamountsoftPA areineffective (35). This could

resultfromthebinding ofeach PA todifferentcellsurfaceor ECMsites. The uPAreceptor thusprobablyplays animportant

partin thespatial control of PA-catalyzed proteolysis. Cell types expressing uPA receptors include monocytes/

macrophages, PMNs, fibroblasts, endothelial cells,

keratino-cytes.Expression ofuPAreceptors canbemodulated,sothat their number and affinity will vary under the influence of

growthfactors(5,36).Immunochemical studies have

demon-strated the presence of uPA, presumably receptorbound, at

sites ofattachment offibroblaststothesubstratumandits colo-calization with the cytoskeletalcomponentvinculin (37). Fur-thermore, there isarapidredistributionoftheuPA receptorto

theleading edge ofmonocytesinducedtomigrateinresponse to achemotacticgradient(18). Thus, expression of the uPA

receptor can beahighly dynamicpropertyofat leastcertain cell types.

Theexistenceofplasma membranebinding sitesforuPA,

plasminogen,andplasmin pointstothe cell surface as the site

ofassemblage ofapowerful proteolytic system. Atleasttwo stepsinthePA/plasmin cascadearedramatically enhanced by thisprocess: (a) activation ofpro-uPAis markedlyincreased

whenit isreceptor-bound and whenplasminogen is simulta-neouslypresent onthe cellsurface, andthis inturnaccelerates plasminformation (38);(b) plasminonthecellsurface is

pro-tectedfrom a2-antiplasmin in plasma and extracellularfluids

(9). Taken together, these observations suggesthow plasmin generation andactivitycanbefocusedtorestricted domainsof

the cellularenvironment, suchastheleadingedge ofmigrating

cells.Currentviewsontheinvolvement ofuPAin cell

migra-tion and invasionareillustrated inFig. 1.

ThesourceofuPAthatbindstothereceptor maydiffer for different celltypes. IncertaincasesuPA issynthesizedby

re-ceptor-bearing cells,andbinds inanautocrinewayafter

secre-tion (5). However, pro-uPA is presentin blood plasma and could bind to receptor-bearing circulating cells. Similarly,

mouse spermatozoabinduPAsecreted byepithelial cells ofthe

vasdeferens(39). Inhumancolonadenocarcinomas, increased

levelsofuPA areproduced by stromalcells, whereasuPA re-ceptor mRNAispresentin themalignant cells (40). The possi-bility ofparacrine loading of uPAon receptor-bearing cells increases theversatility of thissystem, and couldprovidean

example of cooperation between different cells in tissues.

The uPA receptor may also serve otherfunctions.

Mem-brane-bound uPA/PAI-l anduPA/PAI-2complexes are

rap-idly cleared from the cell surface by endocytosis, suggesting that thereceptor promotestheirintracellular degradation(5, 18). Mostinterestingly,recentresultsindicate thatuPA bind-ingtoitsreceptor canstimulatethedifferentiationor prolifera-tionof myeloid and other cells (41). These observationssuggest

that theuPA receptor canalsofunctioninasignaltransduction mode,andthusthat uPAshould perhaps beviewedas a bifunc-tional molecule with both growth factor and enzymatic activi-ties.

Plasminogen activators in physiology and pathology

Inadditiontotheirestablished role infibrinolysis, PAs might catalyze localizedpericellular proteolysis inaverywide spec-trum of physiological and pathological situations. Astriking

illustration of this isprovided by ahistologicalassaythat re-veals PAactivity insections ofmouseembryos(Fig.2). A

num-beroftissues, in particularkidney,skin,intracranialstructures, andlung,express PAactivity, inapatternwhich changes with the developmentalstage.Whereasrecentinvivoexperiments

haveconfirmedthat PAsaresynthesized and secretedduring

cellmigration and tissueremodeling, other observations have demonstrated that PAsare alsoexpressedby cells thatare not

involved in suchprocesses. PAs maythus fulfillalarger spec-trumof relatedfunctionsthanhadpreviouslybeensuspected. PAs AND ECM TURNOVER DURING CELL MIGRATION AND TISSUE REMODELING

Embryonic development. The expression of PAs by invasive embryoniccells invivohasbeen best documented in

tropho-C

Pro-MMPs

Fi-prob r uPA

_PgnR

i asiv~~~~~~ _MMPs

PAls Fibrin

Fibronectin Solublen

Laminin

~Peptides

Clae

Figure1.uPAin cellmigration. (A)Insituhybridizationwitha3H-labeledcRNAprobereveals uPA mRNA ininvasivetrophoblastsmigrating froma6.5-d-old mouseembryo(top)into thesurroundinguterinestroma.Seereference 42 forexperimentaldetails. Silvergrainsappearas white spots in this dark-fieldmicrograph(magnification, 800). (B)Autoradiographicdetectionofreceptor-bound '251-uPAatthesurfaceofa human monocyte inachemotacticgradient. Thesourceof thegradientisatthe bottomof thephotograph.Seereference 18 forexperimental

details. Silvergrainsappearasblack spots in thislight-fieldmicrograph,and reveal that the uPA receptor is concentratedatthe_{leading edge}of the cell(magnification, 1,400). (C) Schematicrepresentationof theuPA-plasminsystematthesurface ofreceptor-bearingcells.MMPs,matrix

19 d

Figure 2. Visualization ofPAenzymatic activities

ontissuesectionsof murineembryos.Histological

zymographieswereperformedasdescribed(49).

Sagittal(top) and frontal (bottom)cryostattissue sectionsof16-, 18-, and 19-d-old murine embryos

wereoverlayedwithasolution containingagar,

plasminogen, and casein. Slideswereincubatedat 370Cfor 3h. PAsdiffusefrom thetissues into the overlay,wheretheyactivateplasminogento plas-min, which inturnlysestheinsoluble casein.No lysis is observedintheabsence of plasminogen in theoverlay (not shown). Photographsweretaken using dark-ground illumination,andcaseinolytic

areasappearblack. Note PAenzymatic activityin kidneys,skin, lungs,andintracranialstructures (magnification, 2.2).

blastcells(42), which produce uPA inatemporalpattern com-patible with its involvement in theendometrialdisruption ac-companying both implantation and early growth of the em-bryo. Similarly, other migrating embryonic cells, such as hemopoietic cells colonizing the bursa ofFabricius(43);neural

crestcells, and cerebellar granularneurons,havebeen foundto

produce PAs (44).

Ovulation. Theproductionof PAsbyrodent ovaries

corre-lateswithovulation. Theparticipationof theenzymesin

follic-ularrupture has receivedstrong experimental support: both anti-tPAantibodies and a2-antiplasmin canblock ovulation (45). The regulated expression of tPAinmaturing rodent oo-cyteshas ledtothesuggestion that it might participatein oo-cytematuration and fertilization (46).

Inflammation, wound healing, and angiogenesis.

Mono-cytes/macrophagesand PMNsproduce uPA, suggesting that it playsarole inthe migration ofinflammatory cellstothe sites of inflammation and in the extensive extracellular remodeling

thatfollowstissueinjury.Inaddition, other celltypesinvolved ininflammatoryprocesses,suchaskeratinocytes in wound ree-pithelialization and endothelial cellsinangiogenesis, also pro-duceuPA. Theangiogenic activityofendothelial cells reflects

thebalanceofPAs andPAIs,asillustratedbythebehavior of

hemangiomas, in whichashiftofthe balancetowards proteoly-sis resultsinabnormalvessel formation(47).

Neoplasia.Alarge body of evidence implicates uPA in the invasive properties of malignant cells (1). The most direct

comesfromstudiesshowing that inhibitory anti-uPA

antibod-ies interfere with metastasis oftumorcells in chickembryos.In

thismodel, cell surface receptor-bound uPA is requiredforone

specificstage inmetastasis, i.e.,stromal invasion(35).Invivo, animal and humantumors often containhigh levels of uPA and,inanumber ofcases,metastaticpropensitycorrelates with the levels ofenzymeproduction:forinstance,uPA is abundant

insquamousbutnotinbasal cellcarcinomas,the former

hav-ingahigher invasive and metastatic potential (48).

PAs AND THE MAINTENANCE OFFLUIDITY IN TUBULAR

STRUCTURES

Endothelialcellsproduce tPA;theenzyme,whichmayremain

boundtothecellsand/orbe released into theblood,isthought

to prevent inappropriate clot formation. Recently, uPA and

tPAsynthesishas been demonstratedinvariousepithelialcells

lining tubularstructures, asituationreminiscent of the

func-tionoftPA in the vascular bed.Inthe murinekidney,both PAs

are secreted into urine by epithelial cells lining distinct

seg-16 d

18 d

PI.g

"

V..i.1.

i i

., 4

.lt

mentsof urinarytubules (49). Similarly, epithelialcellsin the mammary gland, the vas deferens and the seminalvesicle

pro-duceand secrete uPA (32, 39). These cumulated observations

supportarole forPAsinthemaintenanceof patencyand

fluid-ityintubular structures.

PAs AND THE BIOLOGY OF GROWTH FACTORS, HORMONES, ANDNEUROPEPTIDES

PAs have been detected in the adult central nervous system

(44),as well asinamajority of endocrineglands. Because ECM turnoveris probablynot a veryactiveprocess in such tissues,

other functions for PA-catalyzed extracellular proteolysis

shouldbeenvisioned. Thishas led to the suggestion that PAs

might participate in prohormone processing; recent data,

whichhavetentatively identifiedintracellular processing pro-teases,donotsupportthis hypothesis.However, the concept of

proteolytic activation ofhormones and growth factors should berevisited inthelight ofrecentstudies:plasmin can activate

the latentform of TGF,3 and release bFGF-glycosaminoglycan complexes from the matrixandthe cell surface (11), whereas PAs might cause posttranslational processing of hepatocyte growth factor, which has extensive homology with plasmino-gen. In addition, proteolytic inactivation of growth factors

couldinvolvethePA/plasmin system. Modulation of

expres-sion ofreceptorsfor growth factors,hormones, and

neurotrans-mitters may also beachieved by PA-catalyzed proteolysis, as

initiallysuggested for the EGF receptor (50). Finally, it is

con-ceivable that PAs andPAIs could play a partincontrolling the

plasticity ofsynaptic structures.

Inconclusion,thePA/plasmin system is part of a cohort of

enzymatic pathwaysthat catalyze extracellular proteolysis. We nowknowthatitis expressedina remarkably broad spectrum

of biological situations, indicatingthedynamic nature ofthe

interaction ofmost cellswith their environment. To elucidate

thefunctionsofPA-catalyzedproteolysis, weneedto develop newtools: thesemay comefromtheidentification of selective

and pharmacologically useful inhibitors, or from the genetic

approach of transgenic animals. Meeting this challenge will

teachusmuchabout theecology ofcellsintheorganism.

Acknowledgments

WethankourcolleaguesDrs. A.Wohlwend, J. Huarte, and M. Pepper for theiressential contributionstomany aspects ofourwork, and Mr. J.-P.Gerberfor photographic work. The seminal and provocative influ-enceof Dr. Ed Reichonallfacets ofplasminogenactivationis acknowl-edged.

The work from ourlaboratories has been funded by the Fonds National Suisse de la Recherche Scientifique, the Ligue Genevoise Contre leCancer, the Sir Jules Thorn Charitable Overseas Trust,and theCommissionF&Ieraledes Maladies Rhumatismales.

References

1.Dano, K.,P. A.Andreasen,J.Grondahl-Hansen, P. Kristensen,L.S.

Niel-sen,andL.Skriver. 1985.Plasminogenactivators,tissuedegradation,and cancer. Adv.Cancer Res. 44:139-266.

2.Petersen, L. C., L.R.Lund,L.S. Nielsen,K. Dano, and L.Skriver. 1988. One-chainurokinase-typeplasminogen activator from human sarcoma cells isa

proenzyme with littleor nointrinsic activity.J.Biol.Chem. 263:11189-11195. 3. Kobayashi,H., M.Schmitt,L.Goretzki,N.Chucholowski,J.Calvete, M.

Kramer, W.A.Gunzler,F.JRnicke,and H. Graeff. 1991. CathepsinBefficiently

activatesthesoluble and thetumorcellreceptor-bound form of theproenzyme urokinase-type plasminogenactivator(pro-uPA).J.BioL.Chem. 266:5147-5152.

4. Vassalli, J.-D., D. Baccino, and D. Belin. 1985. A cellular binding site for the Mr55,000form ofthe humanplasminogen activator, urokinase. J. Cell Biol.

100:86-92.

5. Blasi, F., N. Behrendt, M. V. Cubellis, V. Ellis, L. R. Lund, M. T.Masucci,

L. B.Moller, D. P. Olson, N. Pedersen, M. Ploug, E. Ronne, and K. Dano. 1990. Theurokinase receptor and regulation of cell surface plasminogen activation.

CellDiffer.Dev.32:247-254.

6.Gething,M.J., B. Adler, J. A. Boose, R. D. Gerard, E. L. Madison, D. McGookey, R. S. Meidell, L. M. Roman, and J. Sambrook. 1988. Variants of humantissue-type plasminogen activator that lack specific structural domains of theheavychain. EMBO (Eur. Mol. Biol. Organ.) J. 7:2731-2740.

7. Andrade-Gordon, P., and S. Strickland. 1986. Interaction of heparin with plasminogen activators and plasminogen: effects on the activation of plasmino-gen.Biochemistry.25:4033-4040.

8.Quigley,J.P., M. B. Berkenpas, R. T. Aimes, and J. M. Chen. 1990.Serine

protease and metalloprotease cascade systems involved in pericellular proteolysis.

CellDiffer.Dev.32:263-276.

9. Plow, E. F., and L. A. Miles. 1990. Plasminogen receptors in the mediation

ofpericellular proteolysis. CellDiffer.Dev.32:293-298.

10.He, C.S., S. M. Wilhelm, A. P. Pentland, B. L. Marmer, G. A. Grant, A. Z. Eisen, and G.I.Goldberg. 1989. Tissue cooperation in a proteolytic cascade activating human interstitial collagenase. Proc.Natd.Acad. Sci. USA. 86:2632-2636.

11. Rifkin, D. B., D. Moscatelli, J. Bizik, N. Quarto, F. Blei, P. Dennis, R.

Flaumenhaft,and P. Mignatti. 1990. Growthfactorcontrol of extracellular

pro-teolysis. CellDiffer.Dev.32:313-318.

12. Andreasen, P. A., B. Georg, L. R. Lund, A. Riccio, and S. N. Stacey. 1990. Plasminogen activator inhibitors: hormonally regulated serpins.Mol.Cell.

Endo-crinol.68:1-19.

13. Seiffert, D., J. Mimuro, R. R. Schleef, and D. J. Loskutoff. 1990. Interac-tions between type Iplasminogen activator inhibitor, extracellular matrix and vitronectin. CellDiffer.Dev.32:287-292.

14.Ehrlich, H. J., J.Keijer,T.Preissner,R. K.Gebbink, and H. Pannekoek. 1991. Functionalinteraction of plasminogen activator inhibitor type I(PAI-l)

andheparin. Biochemistry. 30:1021-1028.

15. Belin, D., A. Wohlwend, W.-D. Schleuning, E. K0.Kruithof, and J.-D. Vassalli. 1989. Facultative polypeptide translocation allows a single mRNA to encode the secreted and cytosolic formsofplasminogenactivators inhibitor 2. EMBO (Eur. Mol.Biol. Organ.)J.8:3287-3294.

16.Longstaff, C., and P. J. Gaffney. 1991. Serpin-serine protease binding

kinetics:alpha 2-antiplasminasamodelinhibitor.Biochemistry.30:979-986. 17.Ploug, M., E. Ronne,N.Behrendt, A. L. Jensen, F.Blasi,and K. Dano. 1991. Cellular receptor for urokinase plasminogen activator. Carboxyl-terminal

processingand membraneanchoring byglycosyl-phosphatidylinositol.J. Biol.

Chem.266:1926-1933.

18.Estreicher,A.,J.Mithlhauser,J.-L.Carpentier,L.Orci,and J.-D. Vassalli. 1990. The receptor for urokinase typeplasminogenactivatorpolarizes expression ofthe proteasetotheleading edgeofmigratingmonocytes and promotes

degrada-tion of enzyme inhibitorcomplexes.J.CellBiol. 111:783-792.

19.Ellis, V., T.-C. Wun,N.Behrendt,E.Ronne, and K.Dano.1990.

Inhibi-tionof receptor-boundurokinasebyplasminogen-activatorinhibitors. J.Biol. Chem.265:9904-9908.

20.Rijken,D.C., M.Otter,J.Kuiper,and T. J.vanBerkel. 1990.

Receptor-mediatedendocytosisof tissue-typeplasminogenactivator (t-PA) by liver cells.

Thromb.Res. 10:63-71.

21.Hajjar,K.A., andN. M.Hamel.1990.Identificationandcharacterization of humanendothelialcell membranebindingsites for tissueplasminogen

activa-torand urokinase. J.Biol. Chem.265:2908-2916.

22.Verrall,S., and N. W. Seeds. 1989. Characterization of '25l-tissue

plasmin-ogen activator bindingtocerebellargranuleneurons.J.CellBiol.109:265-271. 23.Miles, L. A., C. M. Dahlberg, J.Plescia,J.Felez,K.Kato, and F. Plow. 1991.Roleofcell-surfacelysinesinplasminogen bindingtocells:identification of

alpha-enolaseas acandidateplasminogenreceptor.Biochemistry. 30:1682-1691.

24.Rorth, P.,C. Nerlov, F.Blasi,and M. Johnsen. 1990.Transcriptionfactor PEA3participatesin the induction of urokinaseplasminogenactivator

transcrip-tion in murinekeratinocytesstimulated withepidermal growthfactoror phorbol-ester.Nucleic AcidsRes. 18:5009-5017.

25.Medcalf, R.L., M. Ruegg, and W.-D. Schleuning. 1990.ADNAmotif relatedtothecAMP-responsiveelement andanexon-locatedactivatorprotein-2 bindingsite in the humantissue-typeplasminogenactivator gene promoter

coop-eratein basalexpressionand convey activationbyphorbolesterandcAMP.J. Biol. Chem. 265:14618-14626.

26. Von derAhe, D.,D.Pearson, andY.Nagamine.1990. Macromolecular interaction on acAMPresponsive regionin the urokinase-typeplasminogen

activator gene:aroleofprotein phosphorylation. Nucleic AcidsRes. 18:199

1-1999.

27. Huarte,J.,D.Belin,A.Vassalli,S.Strickland,andJ.-D. Vassalli. 1987. Meiotic maturationofmouseoocytestriggersthetranslation and

28. Collart, M. A., D. Belin, J.-D. Vassalli, and P.Vassalli. 1987. Modulations

offunctionalactivity indifferentiatedmacrophages areaccompanied by early and transient increase or decrease in c-fos gene transcription. J. Immunol.

139:949-955.

29. Pepper, M. S., D. Belin, R. Montesano, L. Orci, and J.-D. Vassalli. 1990.

Transforminggrowthfactor-beta I modulates basic fibroblast growth factor-in-ducedproteolyticandangiogenicpropertiesof endothelial cells invitro.J.Cell

Biol. 111:743-755.

30. Pepper, M. S., R. Montesano, J.-D.Vassalli,andL. Orci. 1991. Chondro-cytesinhibit endothelialsproutformationinvitro:evidenceforinvolvement ofa

transforming growth factor-beta.J.Cell.Physiol.146:170-179.

31. Pepper, M. S., R. Montesano, L. Orci, and J.-D. Vassalli. 1991. Plasmino-genactivator inhibitor-I is induced inmicrovascular endothelial cellsby a

chon-drocyte-derivedtransforming growth factor-beta. Biochem. Biophys.Res. Com-mun.176:633-638.

32.Larsson, L. I., L.Skriver,L.S.Nielsen,J.Grondahl-Hansen,P.

Kristen-sen, and K. Dano. 1984.Distributionofurokinase-typeplasminogenactivator

immunoreactivityin the mouse. J.Cell Biol.98:894-903.

33.Heiple,J. M., and L.Ossowski. 1986. Human neutrophil plasminogen

activatoris localized inspecificgranules and is translocated to the cell surface by

exocytosis.J. Exp.Med.164:826-840.

34. VanHinsbergh, V. W. M., E. A. Van den Berg, W. Fiers, and G. Dooije-waard. 1990. Tumornecrosisfactorinduces the production of urokinase-type

plasminogenactivator by humanendothelialcells.Blood.75:1991-1998. 35.Ossowski,L. 1988.Invivoinvasion of modified chorioallantoic mem-brane by tumor cells: the roleofcellsurface-boundurokinase. J. Cell Biol.

107:2437-2445.

36.Estreicher,A., A.Wohlwend,D.Belin,W.-D.Schleuning, and J.-D.

Vas-salli. 1989.Characterization ofthe cellularbinding sites fortheurokinase-type

plasminogenactivator. J.Biol. Chem.264:1180-1189.

37.Vaheri, A., R. W. Stephens, E.-M. Salonen, J. Pollanen, and H. Tapio-vaara. 1990. Plasminogenactivationat the cellsurface-matrix interface. Cell.

Differ.Dev.32:255-262.

38.Ellis, V.,M. F.Scully,and V. V. Kakkar. 1989.Plasminogen activation

initiated bysingle-chainurokinase-type plasminogen activator. Potentiation by U937 monocytes. J.Biol. Chem. 264:2185-2188.

39. Huarte, J., D. Belin, D. Bosco, A.-P. Sappino, and J.-D. Vassalli. 1987. Plasminogenactivator and mouse spermatozoa: urokinase synthesis in the male genital tractand binding of the enzyme to the sperm cell surface. J. Cell Biol.

104:1281-1289.

40.Pyke,C., P. Kristensen, E. Ralfkiaer, J. Grondahl-Hansen, J. Eriksen, F. Blasi, and K. Dano.1991.Urokinase-typeplasminogen activator is expressed in stromal cells andits receptor in cancer cells atinvasive fociin human colon

adenocarcinomas.Am.J.Pathol. 138:1059-1067.

41.Nusrat, A. R., and H. A. Chapman. 1991. An autocrine role for urokinase in phorbolester-mediated differentiation of myeloid cell lines. J. Clin. Invest. 87:1091-1097.

42.Sappino,A.-P., J. Huarte, D. Belin, and J.-D.Vassalli.1989.Plasminogen

activatorsin tissue remodelingandinvasion:mRNAlocalizationin mouse ova-ries andimplanting embryos. J. Cell Biol. 109:2471-2479.

43.Valinsky, J. E., E. Reich, and N. LeDouarin.1981. Plasminogen activator inthe bursaof Fabricius:correlations withmorphogenetic remodeling and cell

migrations.Cell.25:471-476.

44.Festoff,B.W.,editor.1990. Serine proteases and theirserpininhibitors in the nervous system. InRegulationinDevelopmentand inDegenerativeand

Malignantdisease. PlenumPublishingCorp.,New York.

45.Tsafriri,A., T. A.Bicsak, S. B.Cajander,T.Ny, and A. J. Hsueh. 1989.

Suppressionofovulationratebyantibodiestotissue-type plasminogenactivator andalpha2-antiplasmin. Endocrinology. 124:415-421.

46.Huarte, J., D.Belin,and J.-D.Vassalli. 1985.Plasminogenactivator in mouseand rat oocytes:induction duringmeiotic maturation. Cell.43:551-558. 47.Pepper, M. S., and R. Montesano. 1990.Proteolyticbalance and capillary morphogenesis.CellDiffer.Dev.32:319-328.

48.Sappino,A.-P., D.Belin,J.Huarte,S.Hirschel-Scholz,J. H.Saurat,and J.-D.Vassalli. 1991. Differential protease expressionbycutaneoussquamous and basal cellcarcinomas.J.Clin.Invest.88:1074-1080.

49.Sappino, A.-P.,J.Huarte,J.-D.Vassalli,and D. Belin. 1991. Sites of

synthesisofurokdnaseandtissue-typeplasminogenactivators in the murnne kid-ney. J.Clin.Invest.87:962-970.

50.Gross,J.L.,M.N.Krupp,D. B.Rifiin,and M. D.Lane. 1983. Downreg-ulation ofepidermal growthfactor receptor correlates withplasminogenactivator

activityin humanA431epidermoidcarcinoma cells. Proc.Natl.Acad. Sci. USA.