regulation of enos expression in brain endothelial...

Regulation of eNOS Expression in Brain Endothelial Cellsby Perinuclear EP3 Receptors

Fernand Gobeil, Jr, Isabelle Dumont, Anne Marilise Marrache, Alejandro Vazquez-Tello,Sylvie G. Bernier, Daniel Abran, Xin Hou, Martin H. Beauchamp, Christiane Quiniou,

Asmaa Bouayad, Sanaa Choufani, Mousumi Bhattacharya, Stephane Molotchnikoff,Alfredo Ribeiro-da-Silva, Daya R. Varma, Ghassan Bkaily, Sylvain Chemtob

Abstract—We reported upregulation of endothelial nitric oxide synthase (eNOS) by PGE2 in tissues and presence ofperinuclear PGE2 receptors (EP). We presently studied mechanisms by which PGE2 induces eNOS expression incerebral microvessel endothelial cells (ECs). 16,16-Dimethyl PGE2 and selective EP3 receptor agonist M&B28767increased eNOS expression in ECs and the NO-dependent vasorelaxant responses induced by substance P on cerebralmicrovessels. These effects could be prevented by prostaglandin transporter blocker bromcresol green and actinomycinD. EP3 immunoreactivity was confirmed on plasma and perinuclear membrane of ECs. M&B28767 increased eNOSRNA expression in EC nuclei, and this effect was augmented by overexpression of EP3 receptors. M&B28767 alsoinduced increased phosphorylation of Erk-1/2 and Akt, as well as changes in membrane potential revealed by thepotentiometric fluorescent dye RH421, which were prevented by iberiotoxin; perinuclear KCa channels were detected,and their functionality corroborated by NS1619-induced Ca2� signals and nuclear membrane potential changes.Moreover, pertussis toxin, Ca2� chelator, and channel blockers EGTA, BAPTA, and SK&F96365, as well as KCa channelblocker iberiotoxin, protein-kinase inhibitors wortmannin and PD 98059, and NF-�B inhibitor pyrrolidine dithiocar-bamate prevented M&B28767-induced increase in Ca2� transients and/or eNOS expression in EC nuclei. We describefor the first time that PGE2 through its access into cell by prostaglandin transporters induces eNOS expression byactivating perinuclear EP3 receptors coupled to pertussis toxin–sensitive G proteins, a process that depends on nuclearenvelope KCa channels, protein kinases, and NF-�B; the roles for nuclear EP3 receptors seem different from those onplasma membrane. (Circ Res. 2002;90:682-689.)

Key Words: perinuclear EP3 receptor � prostaglandin E2 � transporter� nitric oxide synthase � potassium channels

Endothelial nitric oxide synthase (eNOS) plays a crucialrole in the maintenance of systemic as well as cerebral

hemodynamics.1 The regulation of the constitutive eNOS andneuronal NOS gene expression is relatively stringent. Werecently reported that some of the biological effects ofprostaglandins, specifically prostaglandin E2 (PGE2), involvethe induction of eNOS in cerebral microvascular endotheliumand are mediated distinctly via EP3 receptors;2 this effect ofPGE2 operates in the mature and the developing subjectwhere it may be instrumental in controlling oxygenation ofneuronal tissues.3

The biological actions of prostaglandins have been attrib-uted to their interaction with cell surface G protein–coupledreceptors. However, circumstantial evidence supports the

idea that prostaglandins, either formed within the cell orcaptured from extracellular space, may also act intracellu-larly. For example, the enzymes involved in the biosynthesisof prostaglandins, namely COX-1, COX-2, and PLA2, havebeen found to be localized at the nuclear envelope of differentcell types.4,5 A specific prostaglandin transporter that facili-tates the influx of prostaglandins has also been identified.6,7

More importantly, in addition to plasma membrane EP3

receptors, presence of functional perinuclear EP3 receptors forPGE2 has been demonstrated in a variety of cells such ascerebral microvascular ECs, Swiss 3T3 cells, and host cellsHEK293.8,9 Genomic effects of PGE2 through its perinuclearreceptor, notably on highly inducible genes such as mitogenictranscription factor c-fos and iNOS, have been recently

Original received May 30, 2001; resubmission received December 13, 2001; revised resubmission received January 24, 2002; accepted February 6,2002.

From the Departments of Pediatrics, Ophthalmology, and Pharmacology (F.G., I.D., A.M.M., A.V.-T., S.G.B., D.A., X.H., M.H.B., C.Q., A.B., M.B.,S.C.), Research Center of Hôpital Sainte-Justine, Montréal; the Faculty of Biological Sciences (I.D., S.M.), Université de Montréal; the Department ofPharmacology and Therapeutics (A.M.M., M.B., A.R.-d.-S., D.R.V., S.C.), McGill University, Montréal, Québec, Canada; and the Department of CellularBiology (S.C., G.B.), Université de Sherbrooke, Sherbrooke, Québec, Canada.

Correspondence to Dr Sylvain Chemtob, MD, PhD, Research Center, Hôpital Sainte-Justine, Depts of Pediatrics, Ophthalmology and Pharmacology,3175, Côte Sainte-Catherine, Montréal, Quebec, Canada, H3T 1C5. E-mail [email protected]

© 2002 American Heart Association, Inc.

Circulation Research is available at http://www.circresaha.org DOI: 10.1161/01.RES.0000013303.17964.7A

682

by guest on July 11, 2018http://circres.ahajournals.org/

Dow

nloaded from

http://circres.ahajournals.org/

reported.8,9 However, the regulation of constitutive genes,such as eNOS by PGE2 acting through perinuclear receptors,has never been shown. Such a conjecture can be furtherextended with regards to the relative contribution of plasmamembrane G protein–coupled receptors versus their perinu-clear counterparts in the regulation of cellular physiologicalevents.

Because COX-2 localizes principally to the nuclear enve-lope4 and this enzyme plays a dominant role in prostaglandinproduction in brain microvasculature in early postnatal de-velopment,10 one could speculate that in the presence offunctional intracellular prostaglandin receptors, PGE2-induced upregulation of eNOS gene might involve EP3

receptors localized intracellularly, specifically at the cellnucleus.2 The present study was undertaken to test thehypothesis that PGE2 receptor EP3 on nuclear envelopeexhibits different functions from those on plasma membrane,such that PGE2-induced eNOS expression is largely mediatedby activation of nuclear EP3 receptors in brain microvascularECs. Our findings support this hypothesis and reveal thatPGE2-induced eNOS expression is mediated by a pertussistoxin–sensitive EP3, which involves activation of Ca2�-dependent K� channels (KCa), phosphatidylinositol 3-kinase(PI-3 kinase), MAP kinase kinase (MEK), and NF-�Bpathways.

Material and MethodsChemicalsM&B28767 was a gift from Rhone-Poulenc Rorer, Dagenham Essex,UK; EP3� antibody was a gift of Dr H. Shichi (Kresge Eye Institute,Detroit, Mich); and human EP3� cDNA was obtained from Merck(Pointe-Claire, Québec, Canada). The following agents were pur-chased: 16,16-dimethyl-PGE2, U46619, sulprostone, and 17-phenyltrinor-PGE2 (Cayman); ibuprofen, glibenclamide, soybean trypsininhibitor, bromcresol green, bromosulfophthalein, substance P, ara-chidonic acid, phenylmethylsulfonyl fluoride, actinomycin D,3-aminopropyltriethoxysilane, 17�-estradiol, Nonidet P-40, and pyr-rolidine dithiocarbamate (PDTC) (Sigma); iberiotoxin, cromakalim,EGTA, pertussis toxin (PTX), fura-2-AM, BAPTA-AM, wortman-nin, and PD 98059 (Calbiochem); SK&F96365, methylcarbamyl-platelet-activating factor (C-PAF) (Biomol); NS1619 (ResearchBiochemicals International); PGE2 RIA kits (Advanced Magnetics);anti-Big KCa polyclonal antibody (Alomone Labs); anti-FLAGmonoclonal antibody (Santa Cruz Biotechnology); anti–phospho-MAP kinase (Erk1/2) polyclonal antibody (Promega); anti-MAPkinase (Erk1/2) polyclonal antibody (Upstate Biotechnology); anti–phospho-Akt (Ser473) and anti-Akt polyclonal antibodies (New En-

gland BioLabs); horseradish peroxidase-conjugated goat anti-rabbitIgG (Pierce); styryl dye RH421 (Molecular Probes); pCMV-Tag2vector containing FLAG (Strategene); RNA guard RNase inhibitor(Amersham Pharmacia Biotech Inc); human pulmonary artery ECsand brain microvessel endothelial growth media (BioWhittaker);Dulbecco’s modified Eagle’s medium (Life Technologies);[3H]PGE2 (Amersham); and all other chemicals were purchased fromFisher Scientific.

AnimalsExperiments were performed on cells from brain microvasculaturefrom Yorkshire piglets (L’Ange Gardien, Quebec, Canada) anesthe-tized with halothane (2%) and euthanized with intracardiac pento-barbital (120 mg/kg) in accordance with regulations of the CanadianCouncil of Animal Care Committee and approval of the Sainte-Justine Hospital Animal Care Committee.

Cell Culture and TransfectionCerebral microvessels were isolated and primary EC cultures estab-lished as previously described.2,11 ECs from passages 5 to 13 wereused in the present study. The full-length human EP3� receptorcDNA was subcloned into the pCMV-Tag2 vector downstream to theFLAG epitope; framing and orientation were confirmed by sequenc-ing of the construct. Positioning of the small FLAG epitope at theN-terminus does not alter ligand-induced function12 (see Results).ECs (60% confluence) grown on glass coverslips were transfectedwith the plasmid using the FuGENE 6 Transfection Reagent (Roche)or PolyFect (Qiagen) according to manufacturer’s instructions.

Cell Fractionation and Nuclear IsolationCell fractionation was achieved by the hypotonic/Nonidet P-40 lysismethod. Briefly, ECs were washed 3 times with ice-cold PBS, gentlyscraped, and pelleted at 500g for 5 minutes. The cell pellet (for�50�106 cells as starting material) was resuspended in 2 mL lysisbuffer (10 mmol/L Trizma/Base, pH 7.4, 10 mmol/L NaCl,3 mmol/L MgCl2, 100 �g/mL soybean trypsin inhibitor, and1 mmol/L phenylmethylsulfonyl fluoride [PMSF]), homogenized(100 gentle strokes) with a Dounce tissue grinder (tight pestle; BellcoGlass), and then centrifuged at 600g for 10 minutes at 4°C. The pelletwas resuspended in 2 mL lysis buffer containing 0.1% (v/v) NP-40,left on ice for 5 to 10 minutes and sedimented thereafter at 600g for10 minutes at 4°C. Nuclear pellet was washed 3 times with 10-mLlysis buffer. The morphological integrity and purity (�98%) wasassessed by light microscopy after trypan blue staining and byelectron microscopy (Figure 1). Essentially, isolated nuclei werefixed for 4 hours at 4°C in Tris-HCl buffer (50 mmol/L) pH 7.0containing 3% glutaraldehyde, MgCl2 (5 mmol/L), and sucrose(250 mmol/L). Samples were then washed, fixed with osmiumtetroxide (1%), dehydrated with ethyl alcohol, and embedded inEpon. Ultrathin sections were examined using a transmission elec-tron microscope (Philips 410 LS). The nuclear fraction contained�7% of the total cellular activity of plasma membrane marker 5�nucleotidase (Sigma assay kit).

Figure 1. Electron micrographs of isolatednuclei from brain microvascular ECs. Insert in(A) is magnified in (B). Note preservation ofnuclear envelope.

Gobeil et al Perinuclear EP3 Receptors 683


Dow

nloaded from


Gene Transcription Assays and Detection ofeNOS RNAPorcine primary cerebral ECs (80% confluence) were incubated for18 hours in Dulbecco’s modified Eagle’s medium in the absence orpresence of the following agents: ibuprofen (10 �mol/L), 16,16dimethyl PGE2 (1 �mol/L, stable PGE2 analog), prostaglandintransporter inhibitor bromcresol green (50 �mol/L),6,13 and/or selec-tive EP3 agonist M&B28767 (1 �mol/L).14 Ribonuclease protectionassays were performed to detect eNOS and destrin (control) mRNAas previously described.2,15

For experiments using isolated nucleus, eNOS and 18S nuclearRNA was quantified by reverse transcriptase–polymerase chainreaction (RT-PCR) method as utilized.15 For this purpose, isolatednuclei (200 �g protein, 50 �L) were placed in buffer of the followingcomposition: Tris-HCl (10 mmol/L; pH 7.4), KCl (135 mmol/L),MgCl2 (3 mmol/L), CaCl2 (100 nmol/L), ATP, UTP, GTP, and CTP(500 �mol/L), and RNase guard (100 U). Nuclei were incubated at37°C for 60 minutes with saline or 17-phenyl trinor-PGE2 (1�mol/L) or sulprostone (1 �mol/L) or M&B28767 (0.1 �mol/L) inthe presence of membrane permeable and impermeable Ca2� chela-tors EGTA (100 �mol/L) and BAPTA-AM (100 �mol/L), preacti-vated PTX (20 �g/mL, 60 minutes preincubation),9 nonspecificreceptor-operated Ca2� channel blocker SK&F96365 (10 �mol/L),selective Big Ca2�-sensitive K� channel blocker [BKCa] iberiotoxin(0.1 �mol/L), ATP-sensitive K� channel [KATP] blocker gliben-clamide (10 �mol/L), PI-3-kinase inhibitor wortmannin (50 nmol/L),MEK inhibitor PD 98059 (10 �mol/L), or with NF-�B bindinginhibitor PDTC (100 �mol/L).16 Total RNA was purified by stan-dard guanidine isothiocyanate method and subjected to RT-PCR foreNOS and 18S RNA detection using methods and primersreported.2,15

Effect of EP3 Receptor Agonist M&B28767on NO-Dependent Vasorelaxation ofCerebral MicrovesselsTo assess whether modulation of M&B28767-induced eNOS expres-sion by prostaglandin transporter inhibitor bromcresol green wasreflected into function, vasorelaxation to eNOS-dependent substanceP was studied17 on cerebral microvessels in situ using video-imagingtechnique.2,15

Western Blot of MAP Kinases and Akt ActivationIsolated nuclei (100 �g protein) from ECs were treated or not withM&B28767 (0.1 �mol/L) for 0 and 15 minutes in the above-mentioned buffer. Freezing samples in liquid N2 terminated thereaction. Proteins were resolved by SDS-PAGE on 9% gel, trans-ferred onto PVDF membranes, and then probed in immunoblots withErk, phospho-Erk (1/2), Akt, and phospho-Akt antibodies (diluted1:1000) according to manufacturer’s instructions. Autoradiogramswere scanned and analyzed by densitometry (ImagePro 4�software).

Detection of EP3 Receptors and KCa ChannelsEP3 receptors and BKCa channels were immunodetected on ECs andderived isolated nuclei as described8,9; immunogold technique forEP3 was used as reported.8 Briefly, after fixation, cells and nucleiwere incubated for 1 hour with appropriate antibody diluted in PBScontaining 5% goat serum and 5% fetal calf serum. Fluoresceinisothiocyanate (FITC)–conjugated goat anti-rabbit or Texas Red–conjugated donkey anti-rabbit IgG was used as the secondaryantibody (1:200 dilution) (Santa Cruz Biotechnology). In separateexperiments, omission of primary antibodies was used as negativecontrols. Nuclei were stained with propidium iodide (PI; 3 ng/mL) orSytox green (100 nmol/L). Samples were then mounted on slideswith fluoroguard solution (Bio-Rad) and analyzed with fluorescentmicroscope (Zeiss) or a Multi Probe 2001 confocal argon laserscanning system (Molecular Dynamics). In other experiments,[3H]PGE2 binding and displacement by specific EP ligands wasperformed as described.8

Measurement of Calcium Signals andPotentiometric Changes in NucleiNuclear calcium [Ca2�] signals were measured by fura-2-acetoxymethyl ester technique as described.8,9 The fluorescent po-tentiometric styryl dye RH421 was used to ascertain change ofnuclear membrane potential and fluorescence was measured asreported.18 Nuclei were placed in HEPES/Tris (20 mmol/L, pH 7.0)containing CaCl2 (10 nmol/L), sucrose (300 mmol/L), and either100 mmol/L K2SO4 or 1 mmol/L K2SO4. Nuclei were stimulated withNS1619 or with M&B28767 with or without iberiotoxin. Excitationand emission wavelengths were 475 nm and 645 nm, respectively.

cAMP and Inositol 1,4,5-Triphosphates (IP3) AssaysMembrane preparations (200 �g protein) were incubated in theabsence or presence of M&B28767 (0.1 �mol/L),11,19 and cAMP andinositol 1,4,5-triphosphates (IP3) generation was measured on nu-clear and plasma membranes as described.11

Statistical AnalysisData were analyzed by 1-way ANOVA factoring for treatments withthe exception of vasomotor responses, which were analyzed by2-way ANOVA factoring for concentration and treatment groups.Comparison among means was performed by Tukey-Kramermethod. Statistical significance was set at P�0.05. Data are present-ed as mean�SEM.

ResultsIntracellular Role of PGE2 in Regulation ofeNOS ExpressionInhibition of endogenous prostaglandin formation with ibu-profen (12 to 18 hours) caused a marked reduction of eNOSmRNA in microvascular ECs (Figure 2A). Concurrent treat-ment with stable PGE2 analog 16,16-dimethyl PGE2 orselective EP3 agonist M&B28767 prevented the effect ofibuprofen, as documented.2 Prostaglandin transporter inhibi-tor bromcresol green prevented 16,16-dimethyl PGE2- andM&B28767-induced upregulation of eNOS expression; com-parable effects were observed with a distinct prostaglandintransporter inhibitor, bromosulfophthalein (100 �mol/L).6

These prostaglandin transporter inhibitors per se had no effecton eNOS mRNA expression. Data suggest that PGE2 andanalogs do not seem to induce eNOS mRNA expression byactivating the readily accessible plasma membrane EP3 re-ceptor but rather requires intracellular transport for thispurpose.

To test whether the effect of PGE2 analogs on eNOSexpression modulated by prostaglandin transporter inhibitorsis reflected functionally, we exposed porcine brain slices toM&B28767 (4 to 6 hours) and measured the NO-dependentvasorelaxant responses of microvessels to substance P. Sub-stance P–induced vasorelaxation, which as anticipated couldbe fully inhibited by the NOS blocker L-nitro-arginine (L-NA) (1 mmol/L), was augmented in tissues incubated withM&B28767 (Figure 2B). This enhanced vasorelaxation tosubstance P was prevented by concomitant treatment (4 to 6hours) with bromcresol green or transcription inhibitor acti-nomycin D; treatment simply with bromcresol green (withoutM&B28767) did not alter vasorelaxation to substance P. Incontrast, treatment with bromcresol green or actinomycin Ddid not interfere with vasoconstriction to M&B28767 (Figure2C); M&B28767 does not evoke relaxation of brain paren-chymal vessels.19 Hence, findings imply distinct functions for

684 Circulation Research April 5, 2002


Dow

nloaded from


intracellular and plasma membrane EP3 receptors, such thatthe former induces eNOS expression and the latter mediatesvasoconstriction.

Nuclear Immunolocalization of EP3 Receptors andTheir Effect on eNOS ExpressionWe investigated if direct stimulation of nuclear EP3 receptorscan elicit eNOS expression. EP3 immunoreactivity on plasmamembrane and perinuclear region of ECs was confirmed(Figures 3A and 3B) and similarly shown on isolated nuclei(Figure 3C); furthermore, high-resolution electron micros-copy revealed dominance of EP3 more specifically on theouter nuclear membrane (Figure 3A). Stimulation with

Figure 3. A, EP3 immunolocalization on cerebral EC plasmamembrane (left) and nuclear envelope (right) by electron micros-copy (bar�0.5 �m). B, Indirect immunofluorescence (Texas red)detection of EP3 receptor; nuclei were stained with Sytox green,and visualized by confocal microscopy. C and D, Localization ofEP3� and of FLAG-tagged EP3� receptors in isolated nuclei ofECs. Immunofluorescence (FITC) of EP3 and FLAG, and nuclearstaining with PI was performed on isolated nuclei, subsequentlyvisualized by fluorescence microscopy. Nontransfected cells didnot exhibit immunoreactivity to FLAG, as expected. Note perinu-clear immunostaining of EP3 (A, B, and C), which is furtherappreciated by transverse (z) section (insert, B). E, eNOS RNAexpression in nuclei of nontransfected and FLAG-EP3–trans-fected ECs during resting state or stimulated for 60 minuteswith M&B28767 (0.1 �mol/L) or C-PAF (0.1 �mol/L). eNOS RNAwas determined by RT-PCR and normalized to 18S RNA. Valuesare mean�SEM of 3 to 4 experiments. *P�0.05 compared withcorresponding control; †P�0.01 compared with correspondingvalue from nuclei of nontransfected cells.

Figure 2. A, Effects of ibuprofen and PGE2 analogs on eNOSexpression in brain microvascular ECs. ECs were incubated (18hours) in the absence (control) or presence of ibuprofen (10�mol/L) with or without 16,16 dimethyl PGE2 (1 �mol/L), prosta-glandin transporter inhibitor bromcresol green (50 �mol/L; BCG),and/or M&B28767 (0.1 �mol/L); other cells were treated simplywith BCG. RNA (10 �g) was subjected to RNase protectionassays. The unprotected (blots at far right) and protected frag-ments are 414 and 356 nucleotides for eNOS and 237 and 165nucleotides for destrin, respectively. eNOS expression in cellstreated with saline without ibuprofen did not differ from that incontrol untreated cells (not shown). Values in histogram aremean�SEM of 3 to 4 experiments. *P�0.01 compared withcontrol. B, Modulation of vasorelaxation to substance P after 4to 6 hours treatment with BCG or actinomycin D. Brain slicesfrom piglets were treated for 4 to 6 hours with saline, BCG (50�mol/L), and/or M&B28767 (0.1 �mol/L) in absence or presenceof BCG, L-NA (1 mmol/L), or actinomycin D (25 �mol/L).Vasorelaxation to endothelium-dependent substance P wasstudied on U46619 (0.1 �mol/L)-preconstricted vessels in situby video-imaging technique. Values are mean�SEM of 3 to 4experiments. *P�0.05 compared with other treatments (2-wayANOVA factoring for substance P concentration and treatmentgroups). C, Effect of BCG and actinomycin D on vasomotoreffects of M&B28767. Brain slices were treated for 4 to 6 hourswith BCG or actinomycin D as in (B), and thereafter, vasocon-striction to M&B28767 studied by video imaging. Values aremean�SEM of 3 to 4 experiments.



Dow

nloaded from


M&B28767 elicited eNOS RNA expression in isolated nuclei(Figure 3E). Similar results were obtained using sulprostone,an EP1/EP3 receptor agonist, whereas the EP1 selective ago-nist 17-phenyl trinor-PGE2 was ineffective.

To further establish that the effect of M&B28767 wasspecifically dependent on stimulation of EP3 receptors, thiseffect was tested on nuclei of ECs in which we overexpressedEP3 by transfection with a cDNA of EP3� fused to FLAG(Figure 3D); nontransfected cells do not exhibit immunore-activity to FLAG (not shown). Nuclear eNOS expressioninduced by M&B28767 (0.1 �mol/L) was significantly en-hanced in transfected cells (Figure 3E); unrelated lipid,platelet-activating factor was ineffective. In addition, on ECsfrom pulmonary artery where EP3 receptor binding sites werenot detectable, but which do contain the other 3 EP receptors,M&B28767 did not induce upregulation of eNOS RNA asopposed to the well known eNOS inducer 17�-estradiol (10nmol/L),20 which caused a 65�5% increase in eNOS RNA.

Effect of EP3 Stimulation on Nuclear Ca2�

Transients and eNOS Transcription: Role for GProteins, KCa Channels, and KinasesPTX inhibited nuclear eNOS expression evoked by EP3

agonist M&B28767, suggesting coupling to Gi/o (Figure 4A).Despite presence of functional phospholipase C and adenyl-ate cyclase at the nuclear membrane,21,22 EP3 stimulation didnot reduce cAMP generation induced by forskolin or elicit IP3

formation, whereas on plasma membrane, M&B28767 de-creased net forskolin-induced cAMP formation (from 39�3to 30�2 pmol/mg protein/min, P�0.05). Because G proteinsignaling can occur through Ca2� channels23 and nuclear Ca2�

participates in controlling gene transcription,24 we investi-gated if changes in nuclear Ca2� can be elicited by EP3

stimulation. Incubation of isolated nuclei from ECs withM&B28767 induced a concentration-dependent increase innuclear Ca2� levels (Figure 4B). Ca2� chelators EGTA andBAPTA and nonspecific Ca2� channel blocker SK&F96365prevented M&B28767-induced Ca2� transients and inductionof eNOS RNA, as seen with PTX (Figures 4A and 4B).

Upregulation of eNOS gene expression by PI-3 kinase/Aktand MEK-1–dependent pathways has recently been reportedin human ECs.25 These protein kinases have been localized inthe nucleus where they can regulate binding of transcriptionfactors onto promoter regions of targeted genes.24,25,26 Ourfindings show phosphorylation (and resultant activation) ofAkt and Erk1/2 by direct treatment of nuclei with M&B28767and suppression of eNOS gene with PI-3 kinase–activatedprotein kinase Akt27 and MEK inhibitors wortmannin andPD98059, respectively (Figure 4A); NF-�B binding inhibitorPDTC also prevented M&B28767-induced eNOS expression.

We also tested if EP3-induced Ca2� transients can beelicited by opening ion channels23 identified at the nuclearenvelope.28 We focused on K� channels because (1) they arepresent in endothelium including of neurovascular tissue,15

(2) their activation can lead to Ca2� transients in ECs, and (3)they have been detected at the nuclear membrane.28 Isolatednuclei of neurovascular ECs exhibited perinuclear immuno-staining to BKCa detected by confocal microscopy imaging(Figure 5A). On nuclei, KCa channel opener NS1619 stimu-

lated Ca2� transients and potentiometric changes (in high K�

buffer) detected with the styryl dye RH421 (Figures 5B and5C); these signals were inhibited by iberiotoxin. EP3 stimu-lation induced nuclear Ca2� transients, fluorescence-detectedpotentiometric changes, and nuclear eNOS expression, whichwere all prevented by iberiotoxin but not by KATP blockerglibenclamide (Figures 4 and 5C).

DiscussionThe present study was intended to determine the cellular siteof action of PGE2 in regulating eNOS expression in brainmicrovascular ECs. In this process, we have unveiled apreviously undescribed concept for the G protein–coupledreceptor EP3, which exhibits on the nuclear envelope func-tions that are different from those on plasma membrane. Wehave provided direct evidence that stimulation of the perinu-

Figure 4. G protein–, K� channel–, and protein kinase–depen-dence of EP3 stimulation-induced (A) eNOS expression and (B)Ca2� transients in brain EC nuclei. Isolated nuclei were stimu-lated with M&B28767 (0.1 �mol/L) in presence or absence ofEGTA (100 �mol/L), BAPTA (100 �mol/L), PTX (20 �g/mL),SK&F96365 (1 �mol/L), iberiotoxin (0.1 �mol/L), glibenclamide(10 �mol/L), wortmannin (50 nmol/L), PD 98059 (10 �mol/L), orPDTC (100 �mol/L). eNOS nuclear RNA expression was mea-sured by RT-PCR and normalized to 18S RNA. Inset in (A)reveals Western blot of total and phosphorylated Erk1/2 and Aktin absence and presence of M&B28767, representative of 3experiments. Mean fold-increments of phosphorylated Erk-1,Erk-2, and Akt from baseline were 2.24�0.10, 1.54�0.23, and5.12�0.54, respectively; total kinases were stable. Ca2� tran-sients (B) were measured spectrofluorometrically using the indi-cator fura-2/AM. Saline treatment in absence of M&B28767 didnot induce eNOS expression and Ca2� transients (not shown).Values in histograms are mean�SEM of 3 to 4 experimentseach; *P�0.01 compared with values without asterisks.



Dow

nloaded from


clear EP3 receptor can induce the expression of a constitutivegene, namely eNOS. This EP3 induction of eNOS expressionis coupled to a BKCa channel, formerly never shown innuclear envelope of ECs.

Evidence for perinuclear G protein–coupled receptors isaccumulating. Receptors for PTH, endothelin, angiotensin II,opioids, and prostanoids have been found at the perinuclearmembrane.8,9,29–31 Physiological relevance of perinuclearprostanoid receptors is strengthened by localization of ligand-generating enzymes, namely cPLA2, COX-1, and COX-2, inthe vicinity of receptor sites at the nuclear envelope.4,5

Furthermore, functionality of perinuclear G protein–coupledreceptors has only been described for PGE2 receptors includ-ing EP3, which was reported to evoke immediately early genetranscription.8,9 The present findings extend the scope ofnuclear EP3 actions to include the regulation of constitutivegenes, namely of eNOS.

A major feature of this study is the suggestion for differentroles for EP3 receptors on nuclear envelope and plasmamembrane. This inference is supported by a number ofobservations. (1) The presence of EP3 receptors on plasmaand nuclear membranes of brain ECs has previously beendocumented by binding and immunoreactivity9; these obser-vations have been substantiated (Figure 3A). (2) Effects ofibuprofen on eNOS expression in ECs imply a role for

endogenous prostaglandins; accordingly, an inhibitable netsynthesis of PGE2 (1.8�0.4 pg/mg protein/min) by nuclearmembranes was observed during incubation with arachidonicacid (1 �mol/L). Inhibition of endogenous prostaglandinswith ibuprofen allowed to investigate the effects of exoge-nous PGE2 analogs on eNOS expression, which were reversedby PGE2 and selective EP3 agonist M&B28767.14 But pros-taglandin transporter inhibitor bromcresol green (and bromo-sulfophthalein) prevented the PGE2- and M&B28767-inducedincrease of eNOS expression and associated NO-dependentrelaxation to substance P (Figure 2). The prostaglandintransporter is present on ECs.32 The transporter inhibitorsutilized block uptake of prostaglandins into cells but per se donot interfere with prostanoid receptor–mediated events7,13 orwith eNOS expression (Figures 2A through 2C). (3) Vaso-constrictor effects evoked by direct stimulation of EP3 recep-tors were unaltered by prostaglandin transporter inhibitors(Figure 2C), suggesting effects on plasma membrane recep-tors.11 (4) EP3 receptors on nuclear envelope and plasmamembrane are coupled to different signals: adenylate cyclasein the latter but not in the former. (5) More convincingly, arole for nuclear EP3 receptors in inducing eNOS expressionwas obtained by direct stimulation of isolated nuclei fromECs with the EP3 agonist M&B28767; as anticipated from thereceptor occupancy theory, these effects were augmented byoverexpressing the EP3 receptor (Figure 3D). Collectively,these findings suggest a major role for perinuclear EP3

receptors in the regulation of eNOS expression, which seemsdistinct from actions mediated by plasma membrane EP3.

The nucleus is a dynamic calcium and ion barrier28; in theunstimulated state, our nuclear preparations exhibited stableCa2� concentrations and potentiometry (Figure 5). Nuclearcalcium plays an instrumental role in DNA repair, chromatincondensation and regulation of gene transcription.24 A num-ber of Ca2� channels and pumps have been identified on thenuclear envelope. IP3, IP4 and ryanodine receptors, andCa2�-ATPase pump, are some of these nuclear Ca2� traffick-ing mechanisms.33,34 Our findings conform to these reports.Specifically, chelation of extranuclear and intranuclear Ca2�

or blockade of Ca2� channels prevented EP3 stimulation–induced transcription of eNOS in isolated nuclei.

It has been suggested that ion currents regulate nuclearCa2� channels.28 A variety of ion channels have been foundlocalized at the nuclear envelope mostly on the outer mem-brane.28 We have identified functional KCa channels on thenuclear envelope, which are coupled to EP3 receptors, regu-late nuclear Ca2� channels, and in turn affect eNOS expres-sion (Figures 4 and 5). This deduction is supported by effectsof KCa channel openers on nuclear Ca2� transients andpotentiometry, and more importantly by effects of selectiveBKCa blockers on these parameters and eNOS expressioninduced by EP3 stimulation (Figures 4 and 5); of relevance,changes in nuclear membrane potential have been associatedwith DNA replication.24 The actions, we observed for EP3

mediated by BKCa, are dependent on G protein activation butindependent of cAMP and IP3 formation. EP3 receptors cancouple to Gq and/or PTX-sensitive Gi/Go proteins.35 Thissuggests that in brain vascular ECs, induction of eNOS geneby PGE2 via EP3 receptors seems linked to subsequent

Figure 5. A, Localization of BKCa channels on isolated nuclei ofcerebral ECs. Nuclei were immunostained for BKCa (FITC-conjugated), counterstained with PI, and visualized by confocalmicroscopy; in the absence of primary antibody, no stainingwas seen. Perinuclear localization is clearly seen, especially inthe further magnified nucleus in the insert. B and C, Effect ofEP3 and KCa channel stimulation on nuclear Ca2� transients andpotentiometric changes. Ca2� transients were measured spec-trofluorometrically using fura-2/AM; arrow points to moment ofaddition of compounds. Potentiometric changes were assessedusing the styryl dye RH421 in high (100 mmol/L) and low(1 mmol/L) K� buffer; iberiotoxin was used at 0.1 �mol/L. Valuesin histogram are mean�SEM of 3 to 4 experiments; *P�0.05compared with values without asterisks.



Dow

nloaded from


activation of Go proteins; through their �/� subunits these Gproteins may directly control ion channels36 independently ofcAMP or IP3.37 In addition, this nuclear EP3 receptor–evokedincrease in eNOS expression was found to be mediated viaPI-3 kinase/Akt and Erk-MAP kinase–dependent pathwaysand via NF-�B activation (Figure 4), providing an alternateexplanation for similar finding on whole cells.25 In line withthese observations, transcription factors of the NF-�B as wellas AP-1 systems can be regulated by MAP and PI-3 ki-nases.25,38 These along with other factors (such as AP-2,CRE, ER, GATA-1, and SP-1) may bind onto cis elements ofknown consensus sequences on the human eNOS promoter.39

Altogether the findings set forth new perspectives in eluci-dating the signaling of nuclear G protein–coupled receptorssuch as EP3 in controlling gene expression.

In summary, the present study identifies for the first timedistinct functional roles for the nuclear envelope G protein–coupled EP3 receptor of PGE2 from those on plasma mem-brane. In so doing, the data provide a mechanism for theregulation of the constitutive eNOS gene on brain microvas-cular cells acting via nuclear EP3 receptors by PGE2, thatinvolves formerly undescribed perinuclear KCa channels aswell as PI-3 kinase, Erk-MAP kinase, and NF-�B. Thisconcept alluding to intracrine effects of PGE2 applies toconditions which exhibit high endogenous prostaglandinsynthesis via COX-2 pathways localized at the nuclearmembrane,4,5 such as in inflammation and in the developingsubject.10,15

AcknowledgmentsThis study was supported by grants from the Canadian Institute ofHealth Research, the Heart and Stroke Foundation of Québec, andthe March of Dimes. I. Dumont is a recipient of a studentship fromthe Ministry of Indian and Northern Affairs, Canada. F. Gobeil Jr,and A.M. Marrache are recipients of fellowship and studentshipawards, respectively, from the Canadian Institute of Health Re-search. S. Chemtob is recipient of a Canada Research Chair. We arethankful to Hendrika Fernandez for technical assistance and to LesFermes Ménard Inc (L’Ange Gardien, Québec, Canada) for theirgenerous supply of piglets.

References1. Moncada S, Palmer RM, Higgs EA. Nitric oxide: physiology, pathophys-

iology, and pharmacology. Pharmacol Rev. 1991;43:109–142.2. Dumont I, Hou X, Hardy P, Peri KG, Beauchamp M, Najarian T,

Molotchnikoff S, Varma DR, Chemtob S. Developmental regulation ofendothelial nitric oxide synthase in cerebral vessels of newborn pig byprostaglandin E2. J Pharmacol Exp Ther. 1999;291:627–633.

3. Northington FJ, Tobin JR, Harris AP, Traystman RJ, Koehler RC. Devel-opmental and regional differences in nitric oxide synthase activity andblood flow in the sheep brain. J Cereb Blood Flow Metab. 1997;17:109–115.

4. Spencer AG, Woods JW, Arakawa T, Singer II, Smith WL. Subcellularlocalization of prostaglandin endoperoxide H synthases-1 and -2 byimmunoelectron microscopy. J Biol Chem. 1998;273:9886–9893.

5. Schievella AR, Regier MK, Smith WL, Lin LL. Calcium-mediated trans-location of cytosolic phospholipase A2 to the nuclear envelope and endo-plasmic reticulum. J Biol Chem. 1995;270:30749–30754.

6. Kanai N, Lu R, Satriano JA, Bao Y, Wolkoff AW, Schuster VL. Identi-fication and characterization of a prostaglandin transporter. Science.1995;268:866–869.

7. Schuster VL. Molecular mechanisms of prostaglandin transport. AnnuRev Physiol. 1998;60:221–242.

8. Bhattacharya M, Peri KG, Almazan G, Ribeiro-da-Silva A, Shichi H,Durocher Y, Abramovitz M, Hou X, Varma DR, Chemtob S. Nuclear

localization of prostaglandin E2 receptors. Proc Natl Acad Sci U S A.1998;95:15792–15797.

9. Bhattacharya M, Peri K, Ribeiro-da-Silva A, Almazan G, Shichi H, HouX, Varma DR, Chemtob S. Localization of functional prostaglandin E2

receptors EP3 and EP4 in the nuclear envelope. J Biol Chem. 1999;274:15719–15724.

10. Peri KG, Hardy P, Li DY, Varma DR, Chemtob S. Prostaglandin G/Hsynthase-2 is a major contributor of brain prostaglandins in the newborn.J Biol Chem. 1995;270:24615–24620.

11. Li DY, Varma DR, Chemtob S. Ontogenic increase in PGE2 and PGF2�

receptor density in brain microvessels of pigs. Br J Pharmacol. 1994;112:59–64.

12. Neuschafer-Rube F, Oppermann M, Moller U, Boer U, Puschel GP.Agonist-induced phosphorylation by G protein–coupled receptor kinasesof the EP4 receptor carboxyl-terminal domain in an EP3/EP4 prostaglandinE2 receptor hybrid. Mol Pharmacol. 1999;56:419–428.

13. Bito LZ, Baroody RA. Inhibition of pulmonary prostaglandin metabolismby inhibitors of prostaglandin biotransport (probenecid and bromcresolgreen). Prostaglandins. 1975;10:633–639.

14. Abramovitz M, Adam M, Boie Y, Carriere M, Denis D, Godbout C,Lamontagne S, Rochette C, Sawyer N, Tremblay NM, Belley M, GallantM, Dufresne C, Gareau Y, Ruel R, Juteau H, Labelle M, Ouimet N,Metters KM. The utilization of recombinant prostanoid receptors todetermine the affinities and selectivities of prostaglandins and relatedanalogs. Biochim Biophys Acta. 2000;1483:285–293.

15. Najarian T, Marrache AM, Dumont I, Hardy P, Beauchamp MH, Hou X,Peri K, Gobeil F, Jr, Varma DR, Chemtob S. Prolonged hypercapnia-evoked cerebral hyperemia via K� channels. Circ Res. 2000;87:1149–1156.

16. Kawai M, Nishikomori R, Jung EY, Tai G, Yamanaka C, Mayumi M,Heike T. Pyrrolidine dithiocarbamate inhibits intercellular adhesionmolecule-1 biosynthesis induced by cytokines in human fibroblasts.J Immunol. 1995;154:2333–2341.

17. Jansen I, Alafaci C, McCulloch J, Uddman R, Edvinsson L. Tachykinins(substance P, neurokinin A, neuropeptide K, and neurokinin B) in thecerebral circulation: vasomotor responses in vitro and in situ. J CerebBlood Flow Metab. 1991;11:567–575.

18. Laris PC, Bahr DP, Chaffee RR. Membrane potentials in mitochondrialpreparations as measured by means of a cyanine dye. Biochim BiophysActa. 1975;376:415–425.

19. Li DY, Hardy P, Abran D, Martinez-Bermudez AK, Guerguerian AM,Bhattacharya M, Almazan G, Menezes R, Peri KG, Varma DR, ChemtobS. Key role for cyclooxygenase-2 in PGE2 and PGF2� receptor regulationand cerebral blood flow of the newborn. Am J Physiol. 1997;273:R1283–R1290.

20. Kleinert H, Wallerath T, Euchenhofer C, Ihrig-Biedert I, Li H, For-stermann U. Estrogens increase transcription of the human endothelialNO synthase gene: analysis of the transcription factors involved. Hyper-tension. 1998;31:582–588.

21. Divecha N, Irvine RF. Phospholipid signaling. Cell. 1995;80:269–278.22. Yamamoto S, Kawamura K, James TN. Intracellular distribution of ade-

nylate cyclase in human cardiocytes determined by electron microscopiccytochemistry. Microsc Res Tech. 1998;40:479–487.

23. Wickman K, Clapham DE. Ion channel regulation by G proteins. PhysiolRev. 1995;75:865–885.

24. D’Santos CS, Clarke JH, Divecha N. Phospholipid signaling in thenucleus. Biochim Biophys Acta. 1998;1436:201–232.

25. Cieslik K, Abrams CS, Wu KK. Up-regulation of endothelial nitric-oxidesynthase promoter by the phosphatidylinositol 3-kinase gamma/Januskinase 2/MEK-1– dependent pathway. J Biol Chem. 2001;276:1211–1219.

26. Kim SJ, Kahn CR. Insulin regulation of mitogen-activated protein kinasekinase (MEK), mitogen-activated protein kinase and casein kinase in thecell nucleus: a possible role in the regulation of gene expression. BiochemJ. 1997;323:621–627.

27. Toker A. Protein kinases as mediators of phosphoinositide 3-kinase sig-naling. Mol Pharmacol. 2000;57:652–658.

28. Mazzanti M, Bustamante JO, Oberleithner H. Electrical dimension of thenuclear envelope. Physiol Rev. 2001;81:1–19.

29. Ventura C, Maioli M, Pintus G, Posadino AM, Tadolini B. Nuclear opioidreceptors activate opioid peptide gene transcription in isolated myocardialnuclei. J Biol Chem. 1998;273:13383–13386.

30. Bkaily G, Pothier P, D’Orleans-Juste P, Simaan M, Jacques D, Jaalouk D,Belzile F, Hassan G, Boutin C, Haddad G, Neugebauer W. The use ofconfocal microscopy in the investigation of cell structure and function in



Dow

nloaded from


the heart, vascular endothelium and smooth muscle cells. Mol CellBiochem. 1997;172:171–194.

31. Massfelder T, Dann P, Wu TL, Vasavada R, Helwig JJ, Stewart AF.Opposing mitogenic and anti-mitogenic actions of parathyroid hormone-related protein in vascular smooth muscle cells: a critical role for nucleartargeting. Proc Natl Acad Sci U S A. 1997;94:13630–13635.

32. Topper JN, Cai J, Stavrakis G, Anderson KR, Woolf EA, Sampson BA,Schoen FJ, Falb D, Gimbrone MA, Jr. Human prostaglandin transportergene (hPGT) is regulated by fluid mechanical stimuli in cultured ECs andexpressed in vascular endothelium in vivo. Circulation. 1998;98:2396–2403.

33. Malviya AN, Rogue P, Vincendon G. Stereospecific inositol 1,4,5-[32P]trisphosphate binding to isolated rat liver nuclei: evidence for inositoltrisphosphate receptor-mediated calcium release from the nucleus. ProcNatl Acad Sci U S A. 1990;87:9270–9274.

34. Gerasimenko OV, Gerasimenko JV, Tepikin AV, Petersen OH. Calciumtransport pathways in the nucleus. Pflugers Arch. 1996;432:1–6.

35. Namba T, Sugimoto Y, Negishi M, Irie A, Ushikubi F, Kakizuka A, ItoS, Ichikawa A, Narumiya S. Alternative splicing of C-terminal tail ofprostaglandin E receptor subtype EP3 determines G-protein specificity.Nature. 1993;365:166–170.

36. Rosenthal W, Hescheler J, Trautwein W, Schultz G. Control of voltage-dependent Ca2� channels by G protein–coupled receptors. FASEB J.1988;2:2784–2790.

37. Kojima I, Shibata H, Ogata E. Pertussis toxin blocks angiotensin II–in-duced calcium influx but not inositol trisphosphate production in adrenalglomerulosa cell. FEBS Lett. 1986;204:347–351.

38. Garcia-Garcia E, Sanchez-Mejorada G, Rosales C. Phosphatidylinositol3-kinase and ERK are required for NF-�B activation but not for phago-cytosis. J Leukoc Biol. 2001;70:649–658.

39. Miyahara K, Kawamoto T, Sase K, Yui Y, Toda K, Yang LX, Hattori R,Aoyama T, Yamamoto Y, Doi Y. Cloning and structural characterizationof the human endothelial nitric- oxide-synthase gene. Eur J Biochem.1994;223:719–726.



Dow

nloaded from


Ribeiro-da-Silva, Daya R. Varma, Ghassan Bkaily and Sylvain ChemtobBouayad, Sanaa Choufani, Mousumi Bhattacharya, Stephane Molotchnikoff, Alfredo

G. Bernier, Daniel Abran, Xin Hou, Martin H. Beauchamp, Christiane Quiniou, Asmaa Fernand Gobeil, Jr, Isabelle Dumont, Anne Marilise Marrache, Alejandro Vazquez-Tello, Sylvie

Receptors3Regulation of eNOS Expression in Brain Endothelial Cells by Perinuclear EP

Print ISSN: 0009-7330. Online ISSN: 1524-4571 Copyright © 2002 American Heart Association, Inc. All rights reserved.is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231Circulation Research

doi: 10.1161/01.RES.0000013303.17964.7A2002;90:682-689; originally published online February 21, 2002;Circ Res.

http://circres.ahajournals.org/content/90/6/682World Wide Web at:

The online version of this article, along with updated information and services, is located on the

http://circres.ahajournals.org//subscriptions/

is online at: Circulation Research Information about subscribing to Subscriptions:

http://www.lww.com/reprints Information about reprints can be found online at: Reprints:

document. Permissions and Rights Question and Answer about this process is available in the

located, click Request Permissions in the middle column of the Web page under Services. Further informationEditorial Office. Once the online version of the published article for which permission is being requested is

can be obtained via RightsLink, a service of the Copyright Clearance Center, not theCirculation Researchin Requests for permissions to reproduce figures, tables, or portions of articles originally publishedPermissions:


Dow

nloaded from

http://circres.ahajournals.org/content/90/6/682

http://www.ahajournals.org/site/rights/

http://www.lww.com/reprints

http://circres.ahajournals.org//subscriptions/


regulation of enos expression in brain endothelial...

Documents