stimulation of tissue plasminogen activator.pdf

7/29/2019 stimulation of tissue plasminogen activator.pdf

1/8

1992 80: 981-987

RD Medh, L Santell and EG Levinacid: synergistic effect on protein kinase C-mediated activationStimulation of tissue plasminogen activator production by retinoic

http://bloodjournal.hematologylibrary.org/site/misc/rights.xhtml#repub_requestsInformation about reproducing this article in parts or in its entirety may be found online at:

http://bloodjournal.hematologylibrary.org/site/misc/rights.xhtml#reprintsInformation about ordering reprints may be found online at:

http://bloodjournal.hematologylibrary.org/site/subscriptions/index.xhtmlInformation about subscriptions and ASH membership may be found online at:

reserved.Copyright 2011 by The American Society of Hematology; all rights900, Washington DC 20036.weekly by the American Society of Hematology, 2021 L St, NW, SuiteBlood (print ISSN 0006-4971, online ISSN 1528-0020), is published

For personal use only.by guest on April 3, 2012.bloodjournal.hematologylibrary.orgFrom
http://bloodjournal.hematologylibrary.org/site/misc/rights.xhtml#repub_requestshttp://bloodjournal.hematologylibrary.org/site/misc/rights.xhtml#repub_requestshttp://bloodjournal.hematologylibrary.org/site/misc/rights.xhtml#reprintshttp://bloodjournal.hematologylibrary.org/site/misc/rights.xhtml#reprintshttp://bloodjournal.hematologylibrary.org/site/misc/rights.xhtml#reprintshttp://bloodjournal.hematologylibrary.org/site/subscriptions/index.xhtmlhttp://bloodjournal.hematologylibrary.org/site/subscriptions/index.xhtmlhttp://bloodjournal.hematologylibrary.org/site/subscriptions/index.xhtmlhttp://bloodjournal.hematologylibrary.org/subscriptions/ToS.dtlhttp://bloodjournal.hematologylibrary.org/subscriptions/ToS.dtlhttp://bloodjournal.hematologylibrary.org/subscriptions/ToS.dtlhttp://bloodjournal.hematologylibrary.org/subscriptions/ToS.dtlhttp://bloodjournal.hematologylibrary.org/subscriptions/ToS.dtlhttp://bloodjournal.hematologylibrary.org/subscriptions/ToS.dtlhttp://bloodjournal.hematologylibrary.org/http://bloodjournal.hematologylibrary.org/subscriptions/ToS.dtlhttp://bloodjournal.hematologylibrary.org/http://bloodjournal.hematologylibrary.org/subscriptions/ToS.dtlhttp://bloodjournal.hematologylibrary.org/site/subscriptions/index.xhtmlhttp://bloodjournal.hematologylibrary.org/site/misc/rights.xhtml#reprintshttp://bloodjournal.hematologylibrary.org/site/misc/rights.xhtml#repub_requests


2/8

Stimulation of Tissue Plasminogen Activator Production by Retinoic Acid:Synergistic Effect on Protein K inase C-Mediated ActivationBy RheemD. Medh, Lydia Santell, and EugeneG. Levin

Trans retinoic acid (t-RA) stimulated the productionoftissueplasminogen activator (tPA) in HeLa-S3and human umbilicalvein endothelial cells (huvecs) in a dose-dependent mannerwith maximal release (four to five times control) at 40nmol/L and40 pmol/L, respectively. In endothelial cells, thestimulation of tPA production by phorbol 12-myristate 13-acetate (PMA) was potentiated 1.9-fold by 10 pmol/L t-RA,or 1.8 times the additive effect. In HeLa cells, total tPAsecretionwith 10 nmol/L PMA was increased from 43 ng/mLto 96 ng/mL by 40 nmol/L t-RA, which was two times theadditive effect. Higher concentrations of t-RA (400 nmol/L)depressed tPA secretion by itself and also suppressed PMA-induced tPA production by 50%. Histamine and thrombinalso synergized with t-RA. t-RA (40 nmol/L) and 10 pg/mLhistamine or 10 U/mL thrombin combined to induce tPAproduction 3.4 and 1.3 times the additive effect in HeLa cells.

IBRI NOL Y SIS is a complex process that results in theF issolution of a fibrin clot by the proteolytic enzymeplasmin. Plasmin is formed from its inactive proenzyme,plasminogen, after specif ic cleavage of a single peptidebond by plasminogen activators. The two physiologicallyrelevant formsof plasminogen activators are tissue plasmino-gen activator (tPA ) and urinary plasminogen activator(uPA ). Because of its affinity for and activation by fibrin,tPA is considered to be the plasminogen activator primarilyresponsible for blood clot lysis.1,2The production of tPA invitro is regulated by several factors including thrombin,histamine, tumor-promoting phorbol esters, hormones, andgrowth factors>-5 In human endothelial cells and HeLacells, the elevation of tPA mRNA precedes the increase intPA antigen secretion6z7and is associated with the activa-tion of protein kinase C (PKC).6,7 horbol ester-stimulatedtPA production can be further regulated by cyclic adeno-sine monophosphate (CAMP), which, while ineffective byitself, potentiates the phorbol ester-dependent stimulationanother fivefold.8 Thus, an interaction between the CA MPand PK C pathways leads to a synergistic effect on tPAproduction. Similar interactions between PK C and othersignaling molecules may also be important for the regula-tion of tPA production under various physiologic or patho-logic states.Vitamin A and its derivatives are known to affect cellularmorphology, development, and differentiation by mecha-nisms that are not clearly defined?-12 Several reportssuggest that vitamin A and other retinoids also stimulatetPA production.13-16However, unlike other known agonistsof tPA production, retinoids act via intracellular receptorsthat belong to the familyof nuclear receptors that includesthe receptors for steroid and thyroid hormone ^ ^ Incontrast to cell surface receptors, which initiate specificsignaling pathways leading to the activation of gene expres-sion, these receptors bind directly to specific nucleotidesequences on target Therefore, regulation of tPAproduction by retinoids may involve a pathway distinct fromthose used by the other known tPA agonists.An analysis ofthe mechanisms that mediate retinoid response and how

Cyclic adenosine monophosphate (CAMP) evels were notsignificantly affected by 10 nmol/L to 10 pmol/L t-RA. Nordid 10 nmol/L PMA and 40 nmol/L t-RA together affect cAMPlevels, suggesting that t-RA-mediated potentiation of PMA-induced tPA production occurred via a mechanism that wasindependent of cAMP levels. Downregulation of proteinkinase C (PKC) by pretreatment of huvecs with 100 nmol/LPMA completely blockeda secondary response to PMA, butdid not have a significant effect on t-RA induction. P retreat-ment with 10 pmol/L BRA, on the other hand, did notsignificantly affect a secondary stimulus by 100 nmol/L PMA,but completely suppressed a secondary stimulation by 10pmol/L t-RA alone. These studies suggest that the mecha-nism mediating t-RA stimulation of tPA production interactswith the PKC pathway, resulting in synergism.o 1992byThe American Society of Hematology.they influence response to other agonists can contribute toour understanding of the complex series of events thatprecede stimulated tPA production. A complete understand-ingof the events leading to tPA production and the factorsaffecting it can enable the developmentof a system that canpotentially be used to manipulate the levels of fibrinolyticactivity in vivo.Using HeLa and endothelial cells as model systems, wehave evaluated various aspectsof the effect of retinoids ontPA production. We report that (1) the stimulation of tPAsecretion by trans retinoic acid (t-RA) in both cell types isdose-dependent with kinetics of release similar to that forother agonists; (2) t-R A potentiates phorbol 12-myristate13-acetate (PMA)-, histamine-, and thrombin-induced tPAproduction; and (3) t-R A pretreatment results in desensiti-zation to the homologous agonist, but does not affectheterologous agonist action significantly. These studiessuggest a distinct mechanism for t-RA-mediated stimula-tion of tPA secretion, with possible interaction with thePK C pathway.

From the Department of Molecular and Experimental Medicine,Submitted September 27,1991; accepted April 27, 1992.Supported by the National I nstitutesofHealth Grant No. HL40435andby the US Council or Tobacco Research. E .G.L . s a recipient ofan American Heart Association E stablished Investigator Award.R.D.M. i s a recipient of an American Heart Association, Cali fomiaAfi liate, postdoctoral fellowship.This i s publication no. 7012-MEM fr om The Scripps ResearchInstitute, La J olla, CA.Address reprint requests to Eugene G. Levin, PhD, The ScnppsResearch I nstitute, Department of M olecular and Experimental Medi-cine, 10666 N Torrey Pines Rd, L a J olla, CA 92037.The publicahon costsof this article were defrayed in part by pagecharge payment. ms article must therefore be hereby markedadvertisement in accordance with 18 U.S.C . section 1734 solely toindicate this fact.

Scnpps Research Institute, L a J olla, CA.

0 1992byThe American Society of H ematology.0006-4971 92/8004-OOI7$3.OO/O

Blood, Vol80, No4 (August15), 1992: pp981-987 981

http://bloodjournal.hematologylibrary.org/subscriptions/ToS.dtlhttp://bloodjournal.hematologylibrary.org/subscriptions/ToS.dtlhttp://bloodjournal.hematologylibrary.org/http://bloodjournal.hematologylibrary.org/subscriptions/ToS.dtlhttp://bloodjournal.hematologylibrary.org/


3/8

982 MEDH, SANTELL, AND LEVIN

MATERIALS AN D METHODS regression analysis of logit percent band counts versus log tPACefl culture. HeLa-S3 cells obtained from ATCC (Rockville,M D) were grown in RPM I 1640 containing 10% newborn calfserum, 200 U/mL penicil lin, 200 p,g/mL streptomycin, 500 ng/mLfungizone, and 2 mmol/L glutamine (all from Whittaker B ioprod-ucts, Walkersvil le, MD). F or treatment with various agonists, cells

were cultured i nto 24-well tissue culture dishes (Corning, NY ) untilconfluent, washed twice with R PM I 1640, and incubated at 37C in500 p,L of 5% NuSerum (Collaborative Research, Waltham, MA,final concentration, 1.25% newborn calf serum) in RPM I 1640containing t-RA (Sigma Chemical, St Louis, MO), PMA (Calbio-chem, L a Jolla, CA), histamine (Calbiochem), forskolin (Calbio-chem), or thrombin (a gift from Dr J .W. Fenton, 11, Department ofHealth, Albany, NY). Stock solutions of retinoic acids wereprepared in DM SO at mol/L and stored at -20C. Beforeuse, the retinoic acid was diluted into culture medium. Controlcultures were exposed to an identical final concentration of DM SOthat never exceeded 0.1%.Endothelial cells were isolated from human umbilical cord veinsas described previously4 and cultured into 20 mg/mL gelatin-(Eastman K odak, Rochester, NY) coated tissue culture flasks(Corning) in RPM I 1640 containing 10% fetal calf serum, 200U /mL penicill in, 200 pg/mL streptomycin, 500 ng/mL fungizone,10 p,g/mL endothelial cell growth factor (ECGF) (BiomedicalTechnologies, Stoughton, MA), 90pg/mL heparin (Sigma), and 2mmol/L glutamine. Studies were performed onsecondary culturesgrown to confluence in 12-well dishes (Coming) under the sameconditions as primary cultures, except that 50 p,g/mL E CGF wasused. Cells were washed twice with RPM I 1640 and incubated at37C with 500 p,L of t-RA , PMA, histamine, forskolin, or thrombinin RPM I 1640 containing 5% NuSerum, 50 pg/mL E CGF , and90p,g/mL heparin.The culture supernatants were collected after the appropriateincubation period, centrifuged at 15,OOQg to remove cell debris,made to 0.01% Tween-80 (Sigma), and frozen at -70C until used.For experiments involving downregulation, cells were treated withthe primary agonist for 16 hours, after which they were washedthree times with RPM I 1640. Cells were then incubated with thesecondary agonist for 24 hours and supernatants collected for tPAmeasurements by enzyme-linked immunoadsorbent assay (EL ISA ).Ninety-six-well microtiter plates (Dynatech Laborato-ries, Chantil ly, V A ) were coated with polyclonal rabbit anti-humantPA IgG4 by incubating 150 pL/well of a solution of 10 pg/mL IgGin 50 mmol/L sodium borate, pH 9.0, overnight at 4C. T he coatedplates were washed three times with wash buffer (phosphate-buffered saline [PBS], 0.05% Tween-20 [Mallinkrodt, Paris, KY],0.1% albumin [Calbiochem], 0.01% thimerosal [Sigma]), andstored at -20C until used. For assays, standards (25 ng to 0.2ng/mL), as well as samples, in a final volume of 100 pL of 5%NuSerum in RPM I 1640 were added to antibody-coated wells.Twenty microliters of horseradish peroxidase-conjugated mono-clonal antibody to human tPA (Corvas Pharmaceuticals, La J olla,CA) diluted 1:50 into PBS-1% albumin was added to each well, andplates were incubated for 1.5 hours on a tilting table at roomtemperature. T he samples were removed and the plates werewashed three times with distil led water to remove excess antibody.One hundred microliters of 0.12 mg/mL 3, 3 , 5, 5 - tetramethylbenzi -dine (Sigma) and 0.012% H202 (Mallinkrodt) in 0.1 mol/L sodiumacetate, pH 4.5, was added to each well for the chromogenicreaction to occur. Plates were covered with aluminum foil for 10minutes. T he reaction was terminated by the addition of 100 pL of1N H2S04 per well. The absorbance was measured on an ELISAplate reader (Molecular Devices, Menlo Park, CA) at 450 nm andstandard curves were analyzed by logit transformation and linear

ELZSA.

concentration.The extent of stimulation of tPA secretion by various agonistswas calculated either in terms of fold stimulation or net increase intPA secretion. T o determine the fold increase, tPA concentrationsmeasured after agonist treatment were divided by the valueobtained from control samples (5% NuSerum treatments). Tocalculate net increase in tPA secretion, tPA levels in controlcultures were deducted from those obtained after agonist treat-ment under identical conditions. The expected additive effect oftwo dif ferent agonists was determined by adding the net amount oftPA secreted in cultures treated with each agonist individually.Synergism s defined as occurring when tPA values,oncotreatmentwith two agonists, exceed the additive value.In desensitization experiments, where sequential agonist treat-ment was required, the residual effect of a primary agonist on tPAsecretion was determined using 5% NuSerum for the secondarytreatment. The residual response was used as a correction factorwhen interpreting the effect of the second agonist during sequen-tial treatments. Again, the actual ng tPA secreted in response toeach agonist was determined by deducting the ng tPA secreted inresponse to 5% NuSerum alone. Statistical analysis was performedusing Wilcoxons two-tailed rank test.Confluent cultures in 75-cmZ flasks were incu-bated with the appropriate agonist for 10minutes, the medium wasaspirated, and the cells washed once with ice-cold PBS. The cellswere extracted for 2hours with 3mL of chilled n-propyl alcohol at4C. The extracts were transferred to tubes, centrifuged at 12,000gto remove cell debris, evaporated under a streamof NZat 5Y C, andstored at -70C until used. The extract was reconstituted in 200 p,Lof 50 mmol/L Tris-HC1, pH 7.4, containing 4 mmol/L EDTA,centrifuged to remove particulate matter, and supernatant used forradiometric assays using a CAMP assay kit from AmershamInternational (A rlington Heights, IL ). Results were calculated aspicomoles CA MP per flask of cells, and are represented as foldincrease over control flasks, which were treated with 5% NuSerumalone.

CAMP assays.

RESULTSThe effectof increasing

concentrations of t-FL4 on the abilityof HeLa and humanendothelial cells to secrete tPA was evaluated. Picomolarconcentrations did not significantly affect tPA secretion ineither cell type. At higher concentrations (240 nmol/L),there was a steady dose-dependent increase in tPA secre-tion in endothelial cells (Fig l), with an average maximumincrease of 3.9 times (range, twofold to fivefold) at 40p,mol/L. HeLa cells showed a dose-dependent increase intPA secretion up to4.7 times control cultures at40nmol/Lt-RA (range, 2- to 10-fold). At higher concentrations, therewas a rapid decline in the amount of tPA antigen secreted.The viability of both cell types, as evident by trypan blueexclusion, was not significantly affected at anyof the t-RAconcentrations used, suggesting that the decrease in tPAsecretion in HeLa cells at higher concentrations was notdue to cell death. At concentrations greater than 40p,mol/L, there was significant cell detachment, restrictingthe feasibility of using higher concentrations. Thus, bothHeLa cells and endothelial cells respond to various concen-trations of retinoic acids by increasing tPA production.However, while the effective concentrations observed in theHeLa cell experiments were physiologic, the doses needed

Effectof t-RAontPA secretion.



4/8

RET lNOlC ACID AND TP A PRODUCTION 983^ ^bUU=

Cg 5008c 4000

t-retinoic acid ( M)Fig 1. Dose-dependent stimulation of tPA secretion by t-RA.

Confluent cultures of once-passaged human umbilical vein endothe-lial cells (0)or HeLa cells ( 0 )were incubated in 5% NuSerumcontaining the indicated concentration of t-RA for 20 hours. Controlcells were treated with 5% NuSerum alone. tPA antigen levels insupernatants were determined by ELISA. Each data point representsthe mean f SDoftriplicate wells assayed induplicate.

to promote tPA production in endothelial cells exceededthat range. We havepreviously demonstrated that the elevation of CAMPpoten-tiates the phorbol ester-stimulated production of tPA8 andproposed that tPA production may be regulated by theinteraction of different signaling pathways. To determinewhether the mechanism of t-RA induction may also interactwith other agonists of tPA production, endothelial andHeLa cells were treated with 10 Fg/mL histamine or 100nmol/L PM A in the presence and absence of increasingconcentrations of t-RA. I n endothelial cells (Fig 2), theaddition of 10 pmol/L t-RA with 100 nmol/L PM Aincreased the level of tPA secreted from 70.1 to 135.4ng/mL , a concentration 1.8 times the expected additive

Effectof t-RAonagonist-induced tPA secretion.

- 00 10~10' 10-1 10~ 10-1Retinoic Acid (M)

3

2%GGGGRetinoic Acid (M)

Fig 2. Synergistic effect between retinoic acid and PMA/hista-mine. Endothelial cells were treated with t-RA (4 nmol/L to 40pmol/L) inthe absence or presence of (A) 100nmol/L PMA or (B ) 10ng/mL histamine (0). xpected additive tPA secretion (A ) s the sumof tPA produced by treatment with PMA/histamine alone and theindicated t-RA concentration. Cells were treatedwiththe appropriateagonistfor20hours and tPA production measured by assaying culturesupernatants by ELISA. Each data point represents the average ofduplicate wells assayedinduplicate.

level. This synergistic effect was dose-dependent and wasobserved with as little as 1x mol/L t-RA (1.4 timesadditive), despite the fact that this concentration itself hadminimal effect on tPA secretion. Thus, it appears thatendothelial cells were more sensitive to lower dosesof t-RAduring activation with phorbol esters than by themselves.This same effect was also observed when PM A was replacedwith 10 Fg/mL histamine. Combined treatment with 1 xmol/L t-R A and histamine induced secretion of tPAwhich was 1.5 times the expected additive effect, and whichincreased to 1.7 times at 1xThe effect of t-RA along the entire PMA dose-responsecurve was examined in HeLa cells. PM A induces tPAsecretion in HeLa cells in a dose-dependent manner, withmaximal secretion of 48 ng/mL at 100 nmol/L PM A (Fig3).Addition of t-RA (40 nmol/L ) with each PM A concentra-tion enhanced tPA secretion in a synergistic manner, butonly within the effective PMA dose range. No effect wasobserved at substimulatory levels. The lowest PM A concen-tration that was significantly affected by t-RA was 1nmol/L . A t 100 nmol/L PMA, addition of 40 nmol/L t-RAcaused a 1.7-fold increase in PMA-induced tPA secretion,which was 1.5 times the expected additive values. However,t-RA had the greatest effect at a PM A concentration of 10nmol/L (96ng/mL , 2.1 times the expected additive effect).Thus, peak tPA secretion occurs at a lower concentration ofPMA (10 times less) in the presence of retinoic acid. In thepresence of 400 nmol/L t-RA , there was a suppression ofPM A-induced tPA secretion to approximately 20% to 50%of that caused by PM A alone (Fig 3). This is the t-R Aconcentration at which tPA secretion starts declining in thet-R A dose-response curve (Fig 1).

The effect of histamine and thrombin on the t-RAdose-response curve was also examined in HeLa cells (Fig4). T he synergistic effect was observed at all concentrationsof t-RA (400 pmol/L to 4 pmol/L ), although the degreeofsynergism increased with histamine from 1.7 times theexpected additive effect at 400 pmol/L to 3.4 times the

mol/L t-RA.

=100E -.U= 7 5 -

" 0" 10-10 i o - 8 10-7 i o - 6PMA (M)

Fig 3. Effect of t-RA on PMA-induced tPA secretion. Confluentcultures of HeLa cells were incubated in 5% NuSerum containing theindicated concentrations of PMA inthe absence( 0)r presence of 40nmol/L ( 0 ) r 400 nmol/L ( A) -RA for 20 hours. Supernatants werecollected and tPA antigen levels were determined by ELISA. Each datapoint represents the mean f SD of duplicate assaysoftriplicate wells.



5/8

984 MEDH, SANTELL, AND LEVIN

= 15-E T Ip 12.I- 9vma 6v)a 3% 0

t -Retinoic Acid (M)Fig 4. Dose-dependent pot entiation of agonist-induced PA secre-

tion by t-RA. HeLa cells were incubated in 5% NuSerum containingthe i ndicated concentrations of t-RA in the absence( 0)r presence of10 pg/mL histamine( 0 ) r 10 U/mL thrombin (A).Culture superna-tants were estimated or th e level of tPA antigen by ELISA. Each datapoin t is the mean f SD of tripl icate sets of wells assayed in duplicate.

additive values at the peak of the t-RA dose-titration curve(40 nmol/L). Synergism between t-RA and thrombin wasconsistent with t-RA plus thrombin, producing a 1.3-foldincrease over additive values along the entire curve.Previous studies haveshown that cAMP acts synergistically with PM A in theproduction of tPA.8 To determine if the effect of t-RA onPM A-induced tPA production was mediated by the eleva-tion of CA MP, HeLa cells were incubated with variousconcentrations of t-RA or forskolin,ort-RA +forskolin for10 minutes, and the CAMP levels determined. T reatmentwith forskolin, a known CAMP-elevating compound, re-sulted in a dose-dependent increase in CAMP (Fig 5A),with 100 pmol/L stimulating a 10-fold increase (4.32 to45.36 pmol). Treatment with 10 nmol/L to 100 pmol/Lt-RA, on the other hand, did not result in CAMP levelssignificantly different from control levels (F ig 5A). Increas-ing the duration of incubation in the presenceof t-RA up to6 hours also did not significantly affect CAMP evels (datanot presented), which ruled out the possibility of a delayedresponse. In fact, when cells were incubated with 10 or 100pmol/L forskolin in the presence of 100 nmol/L t-RA,there was a 40% suppression of the forskolin-mediatedelevation of CAMP (Fig 5A). A s expected, 10 nmoI/L PM Aalone did not affect CAMP evels; however, its coincubationwith 100 pmol/L forskolin potentiated the forskolin-mediated elevation of CAMP 1.8-fold (Fig 5B). These arethe concentrations of PM A and forskolin that synergisti-cally stimulate maximum tPA secretion. When 10nmol/LPM A and 40 nmol/L t-RA, concentrations that causesynergistic tPA stimulation, were added simultaneously,there was no effect on CAMP evels (Fig 5B). These resultssuggest that t-RA-mediated increase in tPA production isindependentof CA MP evels.Timecourseof tPA secretion. The kineticsof tPA releasein response to t-RA, PMA, t-RA + PMA, and PMA +forskolin were analyzed (Fig 6). The time courses of tPAsecretion in treated cultures were similar and showed a8-hour lag period, followed by increased secretion that

Effect of t-RA on CAMP levels.

persisted throughout the 24-hour monitoring period. A t 24hours after incubation, 40 nmol/L t-RA , 10 nmol/L PMA,40 nmol/L t-RA + 10 nmol/L PMA, and 10 nmol/LPMA + 100 pmol/L forskolin-treated cells secreted 2.7-,78.6-, 143.9-, and 144.4-fold higher tPA, respectively, ascompared with 5% Nuserum-treated cells.PK C anti-gen and activity can be downregulated after prolongedexposure to tumor-promoting phorbol esters." Previousstudies have determined that PMA, histamine, and throm-bin pretreatment of endothelial cells results in desensitiza-tion to a homologous secondary stimulus." This propertywas used to compare the mechanisms of PM A-, histamine-,and thrombin-induced tPA secretion versus t-RA-inducedtPA secretion. Endothelial cells were treated with primaryagonist for 16 hours, the medium removed, and a secondagonist added for an additional 16 hours (agonist >agonist;Table 1). tPA accumulation during the second phase wasmeasured. Pretreatment with NuSerum followed by second-

ary treatment with t-RA orPM A induced net tPA secretionof 14.7 and 45.6 ng/mL, ie, 2.2 and 4.7 times control,respectively. After pretreatment with t-RA or PMA fol-lowed by NuSerum, residual secretion from the primaryresponse was 5.7 and 18.53 ng/mL or 1.5 and 2.5 times

Homologous downregulationof tPA secretion.

1400 ( A1200 I-840560

= 280E4-c10-8 10-7 10-6 10-5 10-4s o-s M

" Cont t-RA PMA 1-RA+PMA For s PMA+ForsFig 5. cAMP levels in cells treated wi th various agonists. Cultures

of HeLa cells in 75-cmZ flasks were incubated with 5% NuSerumcontaining he indicated concentrations of (A) forskolin [ml, t-RA (0 ) .or 100nmol /L t-RA +forskolin (EA); or (B) 10nmol /L PMA, 10 nmol/LPMA +10 nmol /L t-RA, 100 pmo l/L forskolin, or 10 pmol/L PMA +100 pm ol /L forskolin for 10 minutes at 37C. Th e cells were washedtwice w ith ice-cold PBS and extracted with n-propanol for 1 hour at4C. The extract s were dried and the residue assayed for cAMP asdescribed n Materials and Methods. All values represent he mean oftwo determinations.



6/8

RETlNOlC ACID AND TPA PRODUCTION

lo8 -

98 5

'A

0Q)) 2 0 0 t B 1d 160 -

120 -80 -40 -O 8 12 16 20 24Time (hours )

Fig 6. Time course of tPA secretion in response to various ago-nists . Cultu res of HeLa cells were incubated at 37C in 5% NuSerumcontaining the appropr iate agonist for the ind icated period of time.Supernatants were collected and assayed for tPA antigen l evels byELISA. (A) Cont rol (0),40 nmol/Lt-RA(O); (B ) 10nmol/LPMA(O), 10nmol/L PMA +40 nmol /L t-RA IO,10 nmol/L PMA +100 pmo l/Lforskolin (A). All values represent the mean SD of quadruplicatesets assayed in dup licate.

control. When both primary and secondary treatments werewith the same agonist (PM A >PMA or t-RA >t-RA ), thelevel of tPA secreted was only 90%of the level secreted ifno secondary addition was performed (agonist >Nuserum),

Table 1. Homologous Downregulationof tPA Secretion inEndothelial Cells

First Treatmen t(1 6 h)

1. 5% NuSerum2. 5% NuSerum3. 5% NuSerum4. 10-5 mol/L t-RA5. mol/L PMA6. 10-5 mol/L t-RA7. mol/L PMA8. mol/L t-RA9. mol/L PMA

Second Treatment(24 hl

5% NuSerummol/L t-RAmol/L PMA

5% NuSerum5% NuSerum

mol/L t-RAmol/L PMAmol/L PMA

10-5 mol/L t-RA

12.32 f 1.1627.06 2 5.1857.98 2 10.9617.99 f 5.0530.85 f 4.2116.90 2 1.0627.59 2 3.1979.69 f 2.9745.11 f 8.18

Confluent cultures of endothelial cells were pretreated with theindicated agonist in 5% NuSerum for 16 hours. Culture supernatantswere collected and cells were washed three times with RPMl 1640 toremove residual agonist. This was followed by incubation with theindicated secondary agonist for 24 hours. Culture supernatants ollow-ing secondary treatment were collected and assayed for tPA antigenlevels by ELISA. Values represent data from three experiments per-formed in duplicate. The actual ng/mL tPA levels from individualexperiments were normalized to values from one experiment to correctfor variability in the efficiency of basal tPA secretion from differentendothelial cell preparations.

indicating that the cells were not responsive to secondarytreatment of the same agonist. When t-RA pretreatmentwas fol lowed by secondary treatment with PM A, net tPAsecretion was67.4ng/mL or 6.5 times control. This value is1.3 times that expected from an additive effect of residualtPA release after pretreatment with t-R A (t-RA >5%NuSerum, net tPA = 5.7 ng/mL) plus secondary PM Atreatment (5% NuSerum >PMA, net tPA =45.7), ie,51.4.This difference was statistically significant (P =.002), ndi-cating that t-RA pretreatment does not desensitize cells toa secondary stimulus with PM A and may continue to have asynergistic effect. If the cells were first treated with PM Aand then t-RA, net tPA secretion was32.8ng/mL . This wascomparable to the expected additive value (33.3ng/mL) ofPM A > 5% NuSerum (net tPA = 18.5 ng/mL ) andt-RA >5% NuSerum (net tPA =14.7ng/mL). T he smalldifference was not significant ( P> .5) and it appears thatPMA does not downregulate endothelial cells to t-RA-induced tPA production.

DISCUSSIONThis report demonstrates that t-R A promotes enhancedtPA production and potentiates the effects of PMA, hista-mine and thrombin in a synergistic manner. This synergismdoes not appear to be mediated by the activation of theCAMP-dependent pathway, which has been previouslyshown to potentiate PM A-induced tPA production. Pre-treatment with t-RA does not affect response to a second-ary stimulus by PM A, and vice versa. These results suggestthat induction of tPA production by t-RA occurs via aseparate mechanism that interacts with the PK C-depen-dent activation of tPA production following PMA, hista-

mine, or thrombin treatment.Several reports have suggested that retinoids affectfibrinolytic activity in vitro.14J5,25-27ome of these studieshave concentrated on correlating this effect to other knownchanges induced by retinoids, such as cartilagetissue remodeling,14 and di f ferenti ati ~n.~~or example,Meats et a1 have shown that increased plasminogen activa-tor activity in response to retinoid treatment may mediatecartilage resorption in chondrocytes,26 while Hamilton14reported that retinoids, especially t-RA , induce plasmino-gen activator activity in synovial fibroblasts, which maycontribute toward basement membrane degradation, inflam-matory lesions, and cartilage damage. Differentiation of F9teratocarcinoma stem cells into primitive endoderm-l ikecells and.parietal endoderm cells occurs in response tot-R A and dibutyryl CAMP. These changes are also associ-ated with a two-step elevation of plasminogen activatoractivity" and tPA antigen and mRNA levels.15 K ooistra eta1 have studied the effects of various retinoids on tPAantigen levels in endothelial cells and have shown thatanalogues with a terminal carboxyl group were most potentin stimulating tPA production. Retinols or retinol esterswere less effective.13 They also reported that plasminogenactivator inhibitor (PA I-1) activity was not significantlyaffected by any of the retinoids tested.Our studies have also demonstrated that retinoids pro-mote enhanced tPA production in a dose- and time-



7/8

986 MEDH, SANTELL, AN D LEVlN

dependent manner. Enhancement of tPA production byt-RA occurs in both endothelial and HeLa cells, althoughmaximum stimulation occurs at different concentrations (40pmol/L v 40 nmol/L , respectively). While optimum t-RAlevels inducing tPA production in HeLa cells are withinphysiologic range, the concentrations of t-R A needed toinduce a twofold induction in endothelial cells are quitehigh, reaching supraphysiologic concentrations. These dataare consistent with other reports of t-RA -induced tPAproduction.13These differences may be due to the availabil-ity of the required intermediates in the tPA responsepathway or the efficiency with which each cell type iscapable of targeting t-RA toward induction of tPA secre-tion (eg, through the number or type of retinoic acidreceptors), that endothelial cells in culture are less sensitiveto t-R A than in vivo, or that retinoic acid is not thephysiologic retinoid responsible for this activity.However, much lower concentrations of t-RA are moreeffective at synergizing with PM A and histamine (to1X than inducing tPA production by themselves, andit is also possible that retinoic acid may play a different rolein tPA regulation than that of a primary stimulator, ie, asecondary modulator of other pathways. The synergy be-tween t-RA and the other agonists employed suggestsmultiple mechanisms of tPA stimulation. Previous studiesof histamine- and thrombin-induced tPA production sug-gest that both act through a single pathway, since coincuba-tion stimulates tPA secretion to levels no greater than thatfor each agonist added separately." In these same studies,we proposed that pathway was related to or the same as thatactivated by phorbol esters, ie, the PK C pathway. PMA isknown to directly activate PK C, while histamine andthrombin act via cell surface receptors which trigger signaltransduction events leading to the activation of PK C. T hefact that the effect of t-RA and any oneof these agonists ontPA secretion is not merely additive, but also synergistic,suggests an interaction between the two pathways. Earlierstudies have indicated an interaction between the PK C andCAMP-dependent pathways in the production of tPA anti-gen and mRNA, despite the fact that cAMP elevation hadno positive effect on tPA itself. T o determine whethert-RA- and PM A-induced synergism was also a result of theactivation of the cAMP pathway by t-RA , we determinedCA MP evels after t-R A treatment. t-RA, in the concentra-tion range of 10 nmol/L to 100 pmol/L , did not affectCAMP evels. Instead, 100nmol/L t-RA suppressed forsko-lin-induced CAMPelevation. Coincubation of cells with100

pmol/L forskolin and various concentrations of t-RA didnot alter the t-R A dose-response for tPA production (datanot shown). This is in contrast to the observation in F9teratocarcinoma cells, where CAMP potentiates the effectof t-RA on tPA mRNA and antigen levels.15Previous studies with endothelial cells have demon-strated the onset of a desensitized state on pretreatmentwith PK C-activating agonists such as PMA, thrombin, andhi~tamine.~~uring this state, secondary treatment with thesame agonist after tPA production has returned to normaldoes not further stimulate tPA production. However, pre-treatment with histamine or thrombin desensitizes the cellsto a secondary treatment with PMA; fully 100%of the PMAeffect is observed after 16 hours of treatment with eitheragonist. On the other hand, prolonged treatment withPM A, which downregulates PK C, reduces both histamineand thrombin effects by 75% suggesting that both of theseagonists stimulate tPA production via the PK C pathway.With respect to t-RA, we found that this agonist alsodesensitizes the cells to itself, but not to PM A. However,PM A pretreatment does not have the same effect on t-RAas thrombin and histamine; the t-RA response is notaffected by PMA pretreatment. We thus conclude thatt-RA induction of tPA production is independent of PK Cactivation and that the observed synergistic mechanismoccurs downstream of PKC activation.It has been reported that t-RA acts via retinoic acidreceptor~,l~-~~hich bind to specif ic sequences called retin-oic acid response elements (RA RE) on genes that areresponsive to it.21-23 ARE have been identified on theretinoic acid receptor f3,22 alcohol dehydrogenase 3,23andlaminin B1 and constitute a 6-nucleotide directrepeat flanked by five nucleotides (GTTCAC NNNNNGTTCA C). tPA production by t-RA is probably stimulatedby a similar mechanism, although there are no data as yet tosubstantiate this hypothesis. In mouse F9cells, Rickles etalZ9 have mapped the tPA gene promoter for regionsresponsible for conferring t-R A and CAMP responses, andboth are within 190 bp of the 5' flanking region. While aCAMP-responsive element exists within 100 bp of the 5'flanking region of the human tPA the regionconferring t-RA response has not been mapped. Thehuman and mouse tPA genes, especially the 5' flankingregions, have significant differences, with approximately62% h~mo l o gy . ~~. ~~his might explain the differentialresponses to t-RA and CA MP by the mouse and humangenes.

REFERENCES1. Camiolo SM , Thorsen S, Astrup T : F ibrinogenolysis andfibr inolysis with tissue plasminogen activator, urokinase, streptoki-nase-activated human globulin, and plasmin. Proc SOCExp BiolM ed 138:277,19712. Levin EG, Loskutoff DJ : Cultured bovine endothelial cellsproduce both urokinase and tissue-type plasminogen activators. JCell Bi ol94:631,19823. Gerard RD, M eidell RS: Regulation of tissue plasminogenactivator expression. Annu Rev Physiol51:245,19894. Levin EG, Marzec U, Anderson J , Harker LA: Thrombin

stimulates tissue plasminogen activator release from culturedhuman endothelial cells. J Clin Invest 74:1988, 19845. LeeLS, Weinstein IB: Epidermal growth factor, like phorbolester, induces plasminogen activator in HeL a cells. Nature 274:696,19786. Levin EG, Marotti K R, Santell L Protein kinase C and thestimulation of tissue plasminogen activator release from humanendothelial cells. Dependence on the elevation of messenger RNA.J Biol Chem 264:16030,19897. Waller E K , Schleuning W-D: Induction of fibrinolytic activity



8/8

RETlNOlC ACID A ND TPA PRODUCTION 987

in HeLa cells by phorbol myristate acetate. J Biol Chem 2606354,19858. Santell L , Levin EG : Cyclic A M P potentiates phorbol esterstimulation of tissue plasminogen activator release and inhibitssecretion of plasminogen activator inhibitor-1 from human endothe-lial cells. J Biol Chem 263:16802,19889. Chiocca EA , Davies PJ A , Stein JP: Regulation of tissuetransglutaminase gene expression as a molecular model for reti-noid effects on proliferation and differentiation. J Cell Biochem39:293,198910. Floyd EE, Jetten AM: Regulation of type 1 (epidermal)transglutaminase mRNA levels during squamous differentiation:Down regulation by retinoids. M ol Cell Biol9:4846,198911. Nara K, Nakanishi K, Hagiwara H, Wakita K -i, K ojimaS,Hirose S: Retinol-induced morphological changes of culturedbovine endothelial cells are accompanied by a marked increase intransglutaminase. J Biol Chem 264: 19308,198912. Scheibe RJ , Ginty DD, Wagner J A Retinoic acid stimulatesthe differentiation of PC12 cells that are deficient in CA MP-dependent protein kinase. J Cell Biol 113:1173, 199113. K ooistra T, Opdenberg J P, T oetK, Hendriks HFJ , van denHoogen RM , Emeis J J : Stimulation of tissue-type plasminogenactivator synthesis by retinoids in cultured human endothelial cellsand rat tissues in vivo. Thromb Haemost 65565,199114. Hamil ton J A : Stimulation of the plasminogen activatoractivityof human synovial ibroblasts by retinoids. A rthritis Rheum25:432,198215. Pecorino LT, RicklesRJ , StricklandS:Anti-sense inhibitionof tissue plasminogen activator production in differentiated F9teratocarcinoma cells. Dev Biol129:408,198816. Campbell IK, Piccoli DS, Butler DM , Singleton DK , Hamil-ton J A Recombinant human interleukin-1 stimulates humanarticular cartilage to undergo resorption and human chondrocytesto produce both tissue- and urokinase-type plasminogen activator.Biochim B iophys Acta 967:183,198817. Giguere V, Ong ES, Segui P, Evans RM: Identification of areceptor for the morphogen retinoic acid. Nature 330:624,198718. Petkovich M, Brand NJ , Krust A, Chambon P: A humanretinoic acid receptor which belongs to the family of nuclearreceptors. Nature 330:444,198719. KrustA, K astner PH , Petkovich M, Zelent A , Chambon P: Athird human retinoic acid receptor, hRA R-y. Proc Natl A cad SciUSA 865310,1989

20. B rand N, Petkovich M, Krust A, Chambon P, deThe H,MarchioA,Tiollais P, DejeanA Identificationof a second humanretinoic acid receptor. Nature 323:850, 198821. Umesono K, Giguere V, Glass CK , Rosenfeld MG, EvansRM: Retinoic acid and thyroid hormone induce gene expressionthrough a common responsive element. Nature 336:262,198822. Sucov HM , M urakami KK, Evans RM : Characterization ofan autoregulated response element in the mouse retinoic acidreceptor type pgene. Proc Natl A cad Sci USA 875392,199023. Duester G, Shean ML , McBride MS, Stewart M J Retinoicacid response element in the human alcohol dehydrogenase geneADH3: Implications for regulation of retinoic acid synthesis. M olCell Biol11:1638,199124. L evin EG, Santell L Stimulation and desensitization oftissue plasminogen activator release from human endothelial cells.J Biol Chem 263:9360,198825. Ghezzo F, Pegoraro L Effects of retinoic acid on thefibrinolytic activi ty of H L 60 human promyelocytic leukemia cells.Experientia 37:425,198126. M eats J E , Elford PR, Bunning RA D, Russell RGG : Retin-oids and synovial factor(s) stimulate the production of plasmino-gen activator by cultured human chondrocytes. A possible role forplasminogen activator in the resorption of cartilage in vitro.Biochim B iophys Acta 838:161,198527. Goldstein B, Rogelj S, Siege1 S, Farmer SR , Niles RM :Cyclic adenosine monophosphate-mediated induction of F9 terato-carcinoma differentiation in the absence of retinoic acid. J CellPhysiol143:205,199028. Vasios GW, Gola J D, Petkovich M, Chambon P, GudasLJ:A retinoic acid-responsive element is present in the 5 flankingregion of the laminin B1gene. Proc N atl Acad Sci USA 86:9099,198929. Rickles RJ , Darrow AL, Strickland S: Differentiation-responsive elements in the 5 region of the mouse tissue plasmino-gen activator gene confer two-stage regulation by retinoic acid andcyclic A M P in teratocarcinoma cells. Mol Cell B iol91691,198930. M edcalf RL , Ruegg M , Schleuning W-D: A DNA motifrelated to the CAMP-responsive element and an exon-locatedactivator protein-2 binding site in the human tissue-type plasmino-gen activator gene promoter cooperate in basal expression andconvey activation by phorbol ester and CA MP. J Biol Chem265:14618,199031. Friezner Degen SJ , Rajput B, Reich E: T he human tissueplasminogen activator gene. J Biol Chem 261:6972,1986


Documents

stimulation of tissue plasminogen activator.pdf