Isomer-specific contractile effects of a series of synthetic F2-isoprostanes on retinal and cerebral microvasculature

Free Radical Biology & Medicine, Vol. 36, No. 2, pp. 163 –172, 2004Copyright D 2003 Elsevier Inc.

Printed in the USA. All rights reserved0891-5849/$-see front matter

doi:10.1016/j.freeradbiomed.2003.10.024

Original Contribution

ISOMER-SPECIFIC CONTRACTILE EFFECTS OF A SERIES OF SYNTHETIC

F2-ISOPROSTANES ON RETINAL AND CEREBRAL MICROVASCULATURE

XIN HOU,* L. JACKSON ROBERTS II,y FERNAND GOBEIL Jr.,z DOUGLAS F. TABER,§ KAZUO KANAI,§ DANIEL ABRAN,O

SONIA BRAULT,*,b DANIELLA CHECCHIN,*,b FLORIAN SENNLAUB,* PIERRE LACHAPELLE,b

DAYA R. VARMA,b and SYLVAIN CHEMTOB*,b

*Centre de Recherche de l’Hopital Sainte-Justine, Department of Pediatrics and Department of Pharmacology, Universite de Montreal, 3175 CoteSainte-Catherine, Montreal, PQ, Canada H3T 1C5; yDepartment of Pharmacology and Department of Medicine, Vanderbilt University, Nashville, TN37232, USA; z Institut de Pharmacologie, Universite de Sherbrooke, Sherbrooke, PQ, Canada J1H 5N4; §Department of Chemistry, University of

Delaware, Newark, DE 19716, USA; OTheratechnologies Inc., St. Laurent, PQ, Canada H4S 2A4; and bDepartment of Pharmacology and Departmentof Ophthalmology, McGill University, Montreal, PQ, Canada H3G 1Y6

(Received 27 June 2003; Revised 30 October 2003; Accepted 31 October 2003)

Ad

Hopita

Univer

Canada

E-mail

Abstract—F2-isoprostanes (F2-IsoP’s) are biologically active prostanoids formed by free radical-mediated peroxidation

of arachidonic acid. Four different F2-IsoP regioisomers (5-, 8-, 12-, and 15-series), each comprising eight racemic

diastereomers, total 64 compounds. Information regarding the biological activity of IsoP’s is largely limited to 15-F2t-

IsoP (8-iso-PGF2a). We recently demonstrated that 15-F2t-IsoP and its metabolite, 2,3-dinor-5,6-dihydro-15-F2t-IsoP,

evoked vasoconstriction and TXA2 generation in retina and brain microvasculature. We have now examined and

compared the biological activities of a series of recently synthesized new 5-, 12-, and 15-series F2-IsoP isomers in pig

retinal and brain microvasculature. We hereby show that other 15-series F2-IsoP isomers, 15-epi-15-F2t-IsoP, ent-15-F2t-

IsoP, and ent-15-epi-15-F2t-IsoP, are also potent vasoconstrictors. The 12-series isomers tested, 12-F2t-IsoP and 12-epi-

12-F2t-IsoP, also caused marked vasoconstriction. Of the 5-series isomers tested, 5-F2t-IsoP and 5-epi-5-F2t-IsoP

possessed no vasomotor properties, whereas ent-5-F2t-IsoP caused modest vasoconstriction. The vasoconstriction of ent-

5-F2t-IsoP, 12-F2t-IsoP, and 12-epi-12-F2t-IsoP was abolished by removal of the endothelium, by TXA2 synthase and

receptor inhibitor (CGS12970, L670,596), and by receptor-mediated Ca2+ channel blockade (SK&F96365);

correspondingly, these isomers increased TXB2 formation by activating Ca2+ influx (detected with fura 2-AM) through

non-voltage-dependent receptor-mediated Ca2+ entry (SK&F96365 sensitive) in endothelial cells. In conclusion, as seen

with 15-F2t-IsoP, ent-5-F2t-IsoP, 12-F2t-IsoP, and 12-epi-12-F2t-IsoP constricted both retinal and brain microvessels by

inducing endothelium-dependent TXA2 synthesis. These new findings broaden the scope of our understanding regarding

the potential involvement of F2-IsoP’s as mediators of oxidant injury. D 2003 Elsevier Inc. All rights reserved.

Keywords—Isoprostane, Thromboxane, Calcium, Vascular endothelium, Ischemia, Free radicals

INTRODUCTION

F2-isoprostanes (F2-IsoP’s) are prostaglandin F2-like com-

pounds that are produced in vivo by nonenzymatic free

radical-mediated peroxidation of arachidonic acid [1,2].

Depending on the site of hydrogen abstraction and oxygen

dress correspondence to: Dr. S. Chemtob, Research Center,

l Sainte-Justine, Departments of Pediatrics and Pharmacology,

site de Montreal, 3175 Cote Sainte-Catherine, Montreal, PQ,

H3T 1C5; Fax: +1-514-345-4801;

: [email protected].

163

insertion, four different prostaglandin H2-like isomers

(H2-IsoP’s) are formed. These four intermediates are then

reduced to form four F2-IsoP regioisomers [1,2]. Each of

these regioisomers comprises eight racemic diastereomers

for a total of 64 different compounds. A nomenclature was

established for the IsoP’s, which was approved by the

Eicosanoid Nomenclature Committee, sanctioned by

JCBN of IUPAC [3]. The nomenclature is based on the

designation of the regioisomers according to the number

of the carbon on which the side chain hydroxyl is attached.

Accordingly, there are 5-, 8-, 12-, and 15-series IsoP

X. Hou et al.164

regioisomers. In addition to F-ring IsoP’s, E- and D-ring

IsoP’s and isothromboxanes are also formed in vivo [4,5].

One important aspect related to the discovery of

IsoP’s is that measurement of F2-IsoP’s has emerged as

probably the most reliable approach to assess oxidative

stress status in vivo [6]. However, not only have they

been found to be reliable biomarkers of oxidative stress

but also they have been shown to exert potent biological

actions [2]. This possibility was first demonstrated in

1990, when 15-F2t-IsoP (8-iso-PGF2a) was shown to be

Fig. 1. Structures of F

potent renal vasoconstrictor [1]. The interest in testing

this compound stemmed from the fact that the side chains

in prostaglandin-like compounds produced by the free

radical mechanism are predominantly oriented in cis in

relation to the cyclopentane ring [7], whereas side chains

in cyclooxygenase-derived prostaglandins are exclusive-

ly oriented in trans.

The vast majority of what is known about the biolog-

ical actions of IsoP’s is limited to the two IsoP’s 15-F2t-

IsoP and 15-E2t-IsoP (reviewed in [2]). This is largely

2-isoprostanes.

F2-isoprostane isomer actions 165

due to the fact that these two IsoP’s have been available

in synthetic form for several years. Only recently have

additional IsoP isomers been synthesized for biological

testing. This has revealed that IsoP’s are not homoge-

neous in their actions, and individual F2-IsoP’s exhibit

different biological profiles. For instance, 15-F2c-IsoP

can activate prostaglandin F2a receptors and induce

cardiomyocyte hypertrophy but utilizes intracellular sig-

naling pathways different from those of PGF2a [8,9]. 15-

F2t-IsoP has been documented to be produced in vivo

[10] and exerts a number of potent biological effects such

as stimulation of endothelial and smooth muscle cell

proliferation and endothelin-1 gene and protein expres-

sion [11], induction of endothelial barrier dysfunction

[12], and, most notably, potent and marked vasoconstric-

tion in numerous vascular beds [13–17]. The precise

nature of the receptor(s) mediating these effects of 15-

F2t-IsoP remains to be defined. Nonetheless, the modes

of action of 15-F2t-IsoP on vasculature have been studied

and are shown to differ between species and vascular

beds. For example, 15-F2t-IsoP induces renal vasocon-

striction in rat, which is not modified by cyclooxygenase

inhibitors [18], whereas aortic constriction is partly

dependent on cyclooxygenase products [19]. In the pig

retina and brain, 15-F2t-IsoP and its metabolite (2,3-

dinor-5,6-dihydro-15-F2t-IsoP [20]) evoke vasoconstric-

tion by stimulating thromboxane A2 (TXA2) formation

from neurovascular cells [15,17,21]. In addition to these

acute effects, long-term in vivo actions of 15-F2t-IsoP in

vascular degeneration with implications in ischemic

neuropathies have recently been reported [22,23]. In

contrast, no data are available concerning the vascular

properties of other F2-IsoP’s, especially the 5- and 12-

series (Fig. 1). We therefore proceeded to investigate the

effects and modes of actions of a number of newly

synthesized 5-, 12-, and 15-series F2-IsoP’s on ocular

and cerebral vasomotricity of pigs.

MATERIAL AND METHODS

Tissue preparation

Animals were used according to a protocol of the

Animal Care Committee of Hopital Sainte-Justine along

the principles of the Guide for the Care and Use of

Experimental Animals of the Canadian Council on An-

imal Care. Piglets (1–3 days of age) were acquired from

Fermes Menard, Inc. (L’Ange-Gardien, PQ, Canada).

Animals were anesthetized with halothane (f2.5–5%)

and killed by intracardiac injection of pentobarbital (120

mg/kg). The brains and eyes were removed and placed

immediately in ice-cold Krebs buffer (pH 7.4) of the

composition (mM) NaCl 120, KCl 4.5, CaCl2 2.5,

MgSO4 1.0, NaHCO3 27, KH2PO4 1.0, and glucose

10; 1.5 U/ml heparin was added to the buffer. For

biochemical measurements, tissues were frozen in liquid

N2 and stored at �80jC until assayed.

Vasomotor response of retinal and intraparenchymal

brain microvessels

Eyecups and brain slices (1 mm thick) were prepared

as previously described [15,17,24] to study retinal sur-

face (50–80 Am) and brain intraparenchymal cortical

microvessels (30–50 Am) in situ; these auxotonic prep-

arations minimize vascular injury and reflect more ap-

propriately in vivo physiological conditions [25].

Microvessels were visualized using a video camera

(Model CCD72; MTI, USA) mounted on a dissecting

microscope (Model M-400; Nikon, Japan), as previously

reported [15,17,24]. Vascular diameter was measured

using a digital image analyzer (Sigma Scan software;

Jandel Scientific, Corte Madera, CA, USA), and each

measurement was repeated three times with a variability

of <1%. Vascular diameter was recorded before and after

the topical application of increasing concentrations of

test agents (5-, 12-, and 15-series F2-IsoP’s) in the

presence or the absence of a 20 min pretreatment with

the following agents at known effective concentrations

[15,26]: TXA2 synthase inhibitor CGS12970 (1 AM),

TXA2 receptor antagonist L670596 (0.1 AM), and non-

voltage-dependent Ca2+ entry and receptor-mediated

Ca2+ channel blocker SK&F96365 (20 AM) [27]. Results

were expressed as percentage reduction in vascular

diameter from basal values, which were 65.3 F 5.8

and 35.1 F 3.8 Am, respectively for retinal and intra-

parenchymal brain vessels. Cumulative concentration–

vasorelaxant curves to 5-series F2-IsoP’s (5-F2t-IsoP and

5-epi-5-F2t-IsoP) and substance P were determined on

tissues precontracted with U46619 (0.3 AM) to 50% of

maximal contraction; vasorelaxant response was calcu-

lated as percentage reversal of U46619-induced constric-

tion [28].

Removal of the endothelium

The endothelium of retinal and cerebral microvessels

was chemically removed by intracarotid perfusion with

3-[(3-cholamidopropyl)dimethylammonio]-2-hydroxy-1-

propane sulfonate (Chaps) (5 mg/l for 2 min) [29,30].

The removal of the endothelium was considered success-

ful if the vasorelaxant response to substance P (1 AM)

[28] was abolished, whereas responses to U46619 (0.2

AM) and sodium nitroprusside (1 AM) were unaffected.

Measurement of thromboxane generation

Effects of ent-5-F2t-IsoP, 12-F2t-IsoP, and 12-epi-12-

F2t-IsoP on TXA2 formation were studied in the retina

X. Hou et al.166

and brain slices stimulated (15 min) with test agents at 0.1

and 1 AM; the reaction was terminated with liquid N2.

Thromboxane B2 (stable TXA2 metabolite) was measured

in homogenized tissues by radioimmunoassay as previ-

ously described [15,17]. TXB2 concentration was also

measured in the tissues stimulated for 15 min with ent-

5-F2t-IsoP, 12-F2t-IsoP, and 12-epi-12-F2t-IsoP (1 AM)

in the presence of CGS12970 (1 AM) and SK&F96365

(20 AM).

Brain microvessel endothelial cell culture

Microvessels were suspended in selective endothelial

growth medium (Clonetics, CA, USA). Confluent indi-

vidual endothelial cells were trypsinized, centrifuged,

reseeded in culture flasks, and subcultured as described

in detail [15,17,22]; cell viability was verified by trypan

blue exclusion and was >90%. Endothelial cells were

identified by their cobblestone morphology at conflu-

ence, positive reactivity to Factor VIII antibody, and

negative reactivity to smooth muscle-specific actin and

glial fibrillar acidic protein antibodies (Dako, Carpinte-

Fig. 2. Effects of (top) 5- and 12-series and (bottom) 15-series F2-IsoP’agents were studied in situ on retinal surface and brain slices as describseparate experiments.

ria, CA, USA). Confluent cultures of endothelial cells

from passages 5–15 were used for experiments.

Ca2+ signals

Intracellular Ca2+ ([Ca2+]i) signals were measured

using the fluorescent indicator fura 2-AM as we have

reported [15,17]. For this purpose, confluent brain

microvessel endothelial cells of newborn pigs were

trypsinized in a solution containing 0.05% trypsin and

0.02% EDTA for 2 min, then 5 ml of HBSS was added.

Cells were centrifuged at 250g for 10 min and resus-

pended in a buffer containing (in mM) 20 Hepes, 10 D-

glucose, 4.6 KCl, 118 NaCl, and 0.5 CaCl2, as well as

1% fetal bovine serum. Cell viability was determined

by trypan blue exclusion and was >90%. Fura 2-AM (2

AM) was added to cell suspensions, which were incu-

bated at 37jC for 30 min. The loaded cells were then

washed twice and resuspended in HBSS with Ca2+ (2.5

mM) and 1% fetal bovine serum with or without a 15

min pretreatment with SK&F96365 (20 AM) followed

by stimulation with 5- and 12-series of F2-IsoP’s (0.1

s on vasoconstriction of retinal and brain microvessels. Effects ofed under Material and Methods. Data are means F SEM of five

Table 1. EC50 and maximum constriction values of different agents on retinal and brain microvessels

Agent Retina Brain

EC50 (nM) Maximum

constriction (%)

EC50 (nM) Maximum

constriction (%)

U46619a 33.1 F 3.4 53.4 F 2.9 49.3 F 5.8 34.9 F 3.75-F2t-IsoP n.d. 3.2 F 0.9 n.d. 2.2 F 0.55-Epi-5-F2t-IsoP n.d. 5.2 F 1.1 n.d. 3.2 F 1.2Ent-5-F2t-IsoP 49.3 F 3.1 16.4 F 2.1 53.5 F 6.9 14.1 F 2.112-F2t-IsoP 15.1 F 1.6 24.5 F 3.2 54.8 F 3.4 19.1 F 3.212-Epi-12-F2t-IsoP 27.0 F 3.2 23.3 F 4.2 18.3 F 3.9 22.3 F 2.515-F2t-IsoP 14.7 F 0.9 26.3 F 2.9 22.8 F 3.6 22.4 F 1.32,3-Dinor-5,6-dihydro-15-F2t-IsoP 12.8 F 1.7 21.1 F 2.0 18.5 F 2.7 17.5 F 1.615-Epi-15-F2t-IsoP 54.1 F 2.4 14.4 F 2.3 21.4 F 4.4 10.2 F 2.0Ent-15-F2t-IsoP 30.6 F 4.1 19.2 F 0.9 53.5 F 5.7 15.0 F 0.5Ent-15-epi-15-F2t-IsoP 15.4 F 2.6 28.1 F 3.2 23.7 F 1.4 20.2 F 2.4

Values are means F SEM of five separate experiments. EC50 values were calculated from dose– response curves of concentrations of agents ranging

from 10�12 to 10�5 M. n.d., not detectable.a Data from Ref. [21].

F2-isoprostane isomer actions 167

and 1 AM). The [Ca2+]i was determined in 2 ml of fura

2-AM-loaded cell suspension (f2 � 106 cells/ml)

continuously stirred and measured with a spectrofluo-

rometer (Model LS 50; Perkin–Elmer, Beaconsfield,

UK) by using excitation wavelengths of 340 and 380

nm and emission at 510 nm. Calibration of the fluores-

cent signal was determined using 10 mM ionomycin

and 5 mM EGTA plus 0.2% Triton X-100 to obtain a

maximal and minimal fluorescence ratio. [Ca2+]i was

calculated as reported [31].

Chemicals

F2-IsoP’s tested were synthesized by Dr. Douglas

Taber [32–34]. L670596 and CGS12970 were generous

gifts from Merck–Frosst (Pointe-Claire, PQ, Canada)

and Ciba–Geigy (Summit, NJ, USA), respectively. The

Fig. 3. Vasorelaxant response to 5-F2t-IsoP, 5-epi-5-F2t-IsoP, and suRelaxation is expressed as percentage reversal of U46619 (0.3 AM)-indthose for Fig. 2. Data are means F SEM of five separate experiments

following products were purchased: SK&F96365 (Bio-

Mol, Plymouth Meeting, PA, USA); U46619 (Cayman

Chemicals, MI, USA); ATP, EDTA, EGTA, ionomycin,

Triton X-100, Tris–HCl, and Chaps (Sigma Chemical,

St. Louis, MO, USA); fura 2-AM (Calbiochem, La Jolla,

CA, USA); TXB2 radioimmunoassay kits (Amersham,

Oakville, ON, Canada); all other chemicals (Fisher

Scientific, Montreal, PQ, Canada).

Statistics

Results are expressed as means F SEM. Data were

analyzed using Student’s t test and two-way ANOVA

factoring for concentrations and age or treatments; post-

ANOVA comparisons among means were performed

using the Tukey–Kramer method. p values of less than

.05 were considered to be significant.

bstance P on retinal and intraparenchymal brain microvessels.uced vasoconstriction. Experimental preparations were similar to.

Fig. 4. Contribution of thromboxane on ent-5-F2t-IsoP-, 12-F2t-IsoP-, and 12-epi-12-F2t-IsoP-induced vasoconstriction of retinal andintraparenchymal brain microvessels. Tissues were pretreated 20 min with saline (control) or (A) the thromboxane synthase inhibitorCGS12970 (1 AM), (B) the thromboxane receptor antagonist L670596 (0.1 AM), or (D) the non-voltage-dependent Ca2+ entry andreceptor-mediated Ca2+ channel blocker SK&F96365 (20 AM). Experimental preparations were similar to those for Fig. 2. Data aremeans F SEM of five separate experiments. *p < .01 compared with saline-treated preparations (two-way ANOVA). (C)Deendothelialization of retinal and brain vasculature was performed by intracarotid perfusion with Chaps (5 mg/l for 2 min; see Materialand methods). Data are means F SEM of five separate experiments. *p < .01 compared with values for control (two-way ANOVA).

X. Hou et al.168

e isomer actions 169

RESULTS

Effects of 5-, 12-, and 15-series F2-IsoP’s on retinal and

brain microvessels

12-F2t-IsoP and its epimer (12-epi-12-F2t-IsoP) caused

concentration-dependent constriction of retinal (Fig. 2A)

and brain (Fig. 2B) microvessels. Maximal effects (Emax)

and EC50 values of 12-F2t-IsoP and its epimer (12-epi-

12-F2t-IsoP) were comparable on retinal and brain micro-

vessels (Table 1). In contrast, vasoconstriction to 5-F2t-

IsoP and its epimer (5-epi-5-F2t-IsoP) was negligible on

both preparations; on the other hand the enantiomer ent-

5-F2t-IsoP caused constriction of both the retinal and the

brain microvessels. All 15-series F2-IsoP’s studied (15-

F2t-IsoP, ent-15-F2t-IsoP, 15-epi-F2t-IsoP, ent-15-epi-F2t-

IsoP, and 2,3-dinor-5,6-dihydro-15-F2t-IsoP) evoked sig-

nificant constriction of retinal and brain microvessels

(Fig. 2). 15-F2t-IsoP, ent-15-epi-15-F2t-IsoP, and 2,3-

dinor-5,6-dihydro-15-F2t-IsoP tended to be more effec-

tive than the other 15-series F2-IsoP’s on both prepara-

tions; EC50’s of the 15-series F2-IsoP’s studied were

comparable (Table 1). Because 5-F2t-IsoP and its epimer

(5-epi-5-F2t-IsoP) did not exhibit vasoconstrictor effects,

we tested whether they elicited vasorelaxation; however,

5-F2t-IsoP and 5-epi-5-F2t-IsoP did not cause vasorelax-

ation, whereas substance P evoked a robust relaxation

(Fig. 3).

TXA2-mediated vasoconstriction to F2-IsoP’s

We have previously shown that 15-F2t-IsoP and 2,3-

dinor-5,6-dihydro-15-F2t-IsoP stimulate TXA2 formation

F2-isoprostan

Fig. 5. Effects of ent-5-F2t-IsoP, 12-F2t-IsoP, and 12-epi-12-F2t-IsoP onwere treated with agents at 0.1 and 1 AM for 15 min in the presence ormeansF SEM of five separate experiments. *p < .01 compared with cocompared with agents at 0.1 AM (ANOVA).

on retinal and brain vasculature [15,17,21]. We deter-

mined if vascular effects of ent-5-F2t-IsoP, 12-F2t-IsoP,

and 12-epi-12-F2t-IsoP are also TXA2-dependent. Retinal

and cerebral vasoconstriction evoked by these F2-IsoP’s

was almost completely abolished by the TXA2 synthase

inhibitor CGS12970 and the TXA2 receptor blocker

L670596 (Figs. 4A and 4B). Correspondingly, ent-5-

F2t-IsoP, 12-F2t-IsoP, and 12-epi-12-F2t-IsoP caused

dose-dependent increase in TXB2 levels in retinal and

brain preparations; this effect was inhibited by TXA2

synthase inhibitor CGS12970 (Fig. 5). Ent-5-F2t-IsoP-,

12-F2t-IsoP-, and 12-epi-12-F2t-IsoP-induced increase in

TXB2 formation in cerebral microvascular endothelial

cells was also observed (data not shown).

Involvement of Ca2+ in F2-IsoP-induced TXA2 formation

and vasoconstriction

Removal of endothelium abrogated the TXA2-depen-

dent action of ent-5-F2t-IsoP, 12-F2t-IsoP, and 12-epi-12-

F2t-IsoP (Fig. 4C); it thus seems that endothelial cells

exert a major contribution to the TXA2 formation evoked

by these F2-IsoP’s. Enzyme-catalyzed prostanoid forma-

tion is Ca2+-dependent via phospholipase A2. Endothelial

cells are generally devoid of voltage-gated Ca2+ channels

[35]. We attempted to demonstrate that a non-voltage-

gated Ca2+ channel was involved in inducing TXA2

generation and vasoconstriction in response to the F2-

IsoP’s of interest. Indeed, the ent-5-F2t-IsoP-, 12-

F2t-IsoP-, and 12-epi-12-F2t-IsoP-induced increase in

TXB2 formation in retinal and brain tissue (Fig. 5) and

the evoked vasoconstriction (Fig. 4D) were markedly

thromboxane formation in retinal and brain tissue. Preparationsabsence of CGS12970 (1 AM) or SK&F96365 (20 AM). Data arerresponding value for agents in the absence of inhibitors; zp < .05

Fig. 6. Intracellular peak calcium transients [Ca2+]i in neurovascularendothelial cells in response to 5- and 12-series F2-IsoP’s (0.1 and 1AM) using fura 2-AM (see Material and Methods). Cells were pretreated20 min with saline (control) or SK&F96365 (20 AM). Values are meansF SEM of five separate experiments. *p < .01 compared withcorresponding value for agents in the absence of SK&F96365; zp < .05compared with agents at 0.1 AM (ANOVA).

X. Hou et al.170

inhibited by the putative non-voltage-gated receptor-

operated Ca2+ channel blocker SK&F96365 (Fig. 5);

the L-voltage-gated Ca2+ channel blocker nifedipine

was ineffective (not shown).

The effects of ent-5-F2t-IsoP, 12-F2t-IsoP, and 12-epi-

12-F2t-IsoP on Ca2+ transients in endothelial cells cor-

roborated the data on vasomotor effect and TXB2 forma-

tion. Ent-5-F2t-IsoP, 12-F2t-IsoP, and 12-epi-12-F2t-IsoP

induced an increase in Ca2+ transients in endothelial cells,

which was significantly reduced by SK&F 96365 (Fig. 6).

In contrast, the negligible vasoconstrictors, 5-F2t-IsoP and

5-epi-5-F2t-IsoP, did not affect Ca2+ transients, consistent

with their vasomotor effects (Fig. 2).

DISCUSSION

Whereas the value and reliability of measuring F2-

IsoP’s to assess oxidative stress status in vivo are widely

appreciated [6], less is understood regarding the extent

to which F2-IsoP’s may participate as mediators of

oxidative stress, and most of what is known is largely

limited to the biological actions of 15-F2t-IsoP (8-iso-

PGF2a). 15-series F2-IsoP’s possess potent biological

activity, including vasomotor effects and platelet aggre-

gation, which may be implicated in the pathophysiology

of cardiovascular disorders [13]. However, because nu-

merous forms of isoprostanes can be produced simulta-

neously in vivo [1,2,4–6], we proceeded to explore the

actions of additional recently synthesized F2-IsoP’s of

the 5-, 15-, and 12-series [32–34] that are quantifiable in

humans [35]. Increased levels of 5-F2t-IsoP have been

found in atherosclerosis, myocardial reperfusion after

thrombolysis, and percutaneous transluminal coronary

angioplasty [33,34,36]. However, little is known about

the neurovasomotor effects of 5- and 12-series F2-IsoP’s.

The present study discloses robust neuromicrovascular

properties of 12-series F2-IsoP’s to a degree generally

equivalent to that of the 15-series, which in both cases

are nearly totally endothelium-dependent and mediated

via a non-voltage-gated Ca2+ channel-induced TXA2

formation; unlike 12- and 15-series F2-IsoP’s, 5-series

F2-IsoP’s were either less effective (ent-5-F2t-IsoP) or

ineffective (5-F2t-IsoP and 5-epi-5-F2t-IsoP) compared to

the former two series (Figs. 2 and 3). The present

findings extend our knowledge of the biological prop-

erties of major mediators of peroxidation, which have an

impact on the understanding of the complex process of

ischemic neuropathies.

Vascular effects of 15-F2t-IsoP have been found to be

markedly inhibited by TXA2 receptor blockers [2,13,

15,21], but it remains unclear whether 15-F2t-IsoP exerts

its action via the TXA2 receptor or a similar but distinct

‘‘IsoP’’ receptor [2]. Although 15-F2t-IsoP has been

proposed to interact with the TXA2 receptor [37,38], this

binding seems weak and modest [39]. In other instances,

certain 15 series IsoP’s have been found to interact in the

ciliary body with the FP receptor [9]. In neurovascular

tissue, however, the mechanism of action of 15-F2t-IsoP

is better understood. 15-F2t-IsoP and its metabolite 2,3-

dinor-5,6-dihydro-15-F2t-IsoP produce vasoconstriction

by stimulating TXA2 formation [15,17,21]; a similar

effect in neurovascular degeneration has been observed

[22,23]. In the present study, 12-F2t-IsoP, 12-epi-12-F2t-

IsoP, and ent-5-F2t-IsoP induced thromboxane formation

in retina, brain, and neurovascular endothelium and

elicited a TXA2-dependent constriction of microvessels

in these tissues (Figs. 4A, 4B, and 5). This observation

also applies to the previously tested [15,17,21] as well as

untested 15-series IsoP’s, namely ent-15-F2t-IsoP, 15-epi-

F2t-IsoP, and ent-15-epi-F2t-IsoP. In addition, both the F2-

IsoP’s that evoked TXA2 formation and the vasocon-

striction were calcium-channel-dependent (Figs. 4–6).

Because removal of endothelium, TXA2 synthase inhib-

itors, TXA2 receptor antagonists, and calcium channel

blockers all abolish constriction to F2-IsoP’s (Fig. 4), it is

reasonable to suggest that 12-F2t-IsoP-, 12-epi-12-F2t-

IsoP-, and ent-5-F2t-IsoP-induced constriction is mediat-

ed by a calcium-dependent TXA2 released from vascular

endothelium, as reported with 15-F2t-IsoP [15]. Taken

together, the data suggest that these F2-IsoP’s increase

calcium influx through receptor-operated channels in

endothelium, and this in turn enhances the formation of

TXA2. The commonality of this mechanism of action of

these IsoP isomers might suggest (but not prove) that

they all may exhibit actions via a similar receptor

pathway, which seems to involve that of the TXA2

F2-isoprostane isomer actions 171

receptor, consistent with the requirement of this receptor

to mediate most vasomotor effects evoked by F2-IsoP’s

[13]; on the other hand the molecular identity of the

binding site of F2-IsoP’s remains to be found.

Although 5-F2t-IsoP, its epimer 5-epi-5-F2t-IsoP, and

the 12- and 15-series F2-IsoP’s tested are all formed by

free radical-mediated peroxidation of arachidonic acid,

they differ with regard to the site of the initial hydrogen

abstraction and oxygen insertion resulting in major

differences in the lateral chain structure. These structural

differences might contribute to the diminished or absent

efficacy of 5-F2t-IsoP and 5-epi-5-F2t-IsoP compared to

the other IsoP’s tested and are consistent with the

recently reported lack of response of rat thoracic aorta,

and human internal mammary artery and saphenous vein,

to 5-F2t-IsoP and 5-epi-5-F2t-IsoP [40]. Along the same

lines, related differences in structure are likely responsi-

ble for the decreased retinal and brain vasoconstriction of

15-epi-15-F2t-IsoP and ent-15-F2t-IsoP versus that seen

with the other 15-series F2-IsoP’s. It seems as though the

positioning of the side chains relative to the cyclopentane

ring for the 5-series compounds and the orientation of the

hydroxyl group on the side chains in the 15-series impact

on activity (Fig. 2). In addition, there seems to be some

enantiomer selectivity in the activities of the 5- and 15-

series of isomers.

Numerous forms of IsoP’s are concomitantly generat-

ed during oxidative stress. The preferential increase of

one form over another has been reported for IsoP’s

containing different prostane rings [41]. For instance,

D2/E2 IsoP levels have been found to augment to a

greater extent during hyperoxic stress than F2-IsoP’s

[41]; a similar pattern was also observed for hyperoxia-

induced increase in isofurans over F2-IsoP’s [42]. In

another situation, glutathione and other thiols favored

reduction of IsoP endoperoxides to F-ring IsoP’s over

D2/E2-ring IsoP’s [41]. On the other hand, conditions in

vivo that facilitate the formation of a specific form or

series of F2-IsoP (which differ by their side chains)

remain unknown. Nonetheless, antioxidants are effective

in limiting IsoP generation [43].

In conclusion, the present findings reveal so far

undescribed biological properties of 12-F2t-IsoP, 12-epi-

12-F2t-IsoP, ent-5-F2t-IsoP, 15-epi-15-F2t-IsoP, ent-15-

F2t-IsoP, and ent-15-epi-15-F2t-IsoP. These observations

broaden our understanding of the potential involvement

of F2-IsoP’s as mediators of vascular responses associat-

ed with oxidative stress. In particular in the present

physiological setting, 15-F2t-IsoP has been implicated

in the pathogenesis of ischemic retinopathies and ence-

phalopathies [2,13], especially in the immature subject

with incompletely developed antioxidant defenses

[15,17,22,23]. In ischemic neuropathies secondary to

systemic or local hypoxia, the vasoconstrictor effects of

generated IsoP’s would further compromise local circu-

lation [24]. Preventive, but not therapeutic (postinjury

induction), antioxidant treatment is beneficial in ischemic

encephalopathies [44]; but for therapy administered after

injury induction it may be preferable to target the

mechanisms activated by the generated longer lasting

stable products of oxidation, namely the IsoP’s acting via

TXA2 [44,45].

Acknowledgments—We thank Mrs. Hensy Fernandez for her technicalassistance. This work was supported by grants from the CanadianInstitutes of Health Research, the March of Dimes Birth DefectsFoundation, the Heart and Stroke Foundation of Quebec, the Fonds dela Recherche en Sante du Quebec (Reseau Vision), and the NationalInstitutes of Health Merit Award Grant GM 42056 to L. R. F. S. and S.B. are recipients of a Canadian Institutes of Health Research Fellowshipand Studentship award, respectively, D. C. is an awardee of the NationalScience and Engineering Research Council of Canada, and S. C. holds aCanada Research Chair.

REFERENCES

[1] Morrow, J. D.; Hill, K. E.; Burk, R. F.; Nammour, T. M.; Badr,K. F.; Roberts, L. J., II. A series of prostaglandin F2-like com-pounds are produced in vivo in humans by a non-cyclooxyge-nase, free radical-catalyzed mechanism. Proc. Natl. Acad. Sci.USA 87:9383–9387; 1990.

[2] Roberts, L. J., II; Morrow, J. D. Products of the isoprostane path-way: unique bioactive compounds and markers of lipid peroxida-tion. Cell. Mol. Life Sci. 59:808–820; 2002.

[3] Taber, D. F.; Morrow, J. D.; Roberts, L. J., II. A nomenclaturesystem for the isoprostanes. Prostaglandins 53:63–67; 1997.

[4] Morrow, J. D.; Minton, T. A.; Mukundan, C. R.; Campbell, M.D.; Zackert, W. E.; Daniel, V. C.; Badr, K. F.; Blair, I. A.;Roberts, L. J., II. Free radical induced generation of isoprostanesin vivo: evidence for the formation of D-ring and E-ring iso-prostanes. J. Biol. Chem. 269:4317–4326; 1994.

[5] Morrow, J. D.; Awad, J. A.; Wu, A.; Zackert, W. E.; Daniel, V. C.;Roberts, L. J., II. Free radical-induced generation of thromboxane-like compounds (isothromboxanes) in vivo. J. Biol. Chem. 271:23185–23190; 1996.

[6] Roberts, L. J., II; Morrow, J. D. Measurement of F2-isoprostanesas an index of oxidative stress in vivo. Free Radic. Biol. Med.28:505–513; 2000.

[7] O’Connor, D. E.; Mihelich, E. D.; Coleman, M. C. Stereochem-ical course of the autoxidative cyclization of lipid hydroperoxidesto prostaglandin-like bicylo endoperoxides. J. Am. Chem. Soc.106:3577–3584; 1984.

[8] Kunapuli, P.; Lawson, J. A.; Rokach, J. A.; Meinkoth, J. L.;FitzGerald, G. A. Prostaglandin F2a (PGF2a) and the isoprostane,8,12-iso-isoprostane F2a-III, induce cardiomyocyte hypertrophy:differential activation of downstream signaling pathways. J. Biol.Chem. 273:22442–22452; 1998.

[9] Kunapuli, P.; Lawson, J. A.; Rokach, J.; FitzGerald, G. A. Func-tional characterization of the ocular prostaglandin F2a (PGF2a)receptor. Activation by the isoprostane, 12-iso-PGF2a. J. Biol.Chem. 272:27147–27154; 1997.

[10] Morrow, J. D.; Minton, T. A.; Badr, K. F.; Roberts, L. J., II.Evidence that the F2-isoprostane, 8-epi-prostaglandin F2a, isformed in vivo. Biochim. Biophys. Acta 1210:244–248; 1994.

[11] Fukunaga, M.; Makita, N.; Roberts, L. J., II; Morrow, J. D.;Takahashi, K.; Badr, K. F. Evidence for the existence of F2-iso-prostane receptors on rat vascular smooth muscle cells. Am. J.Physiol. 264:C1619–C1624; 1993.

[12] Hart, C. M.; Karman, R. J.; Blackburn, T. L.; Gupta, M. P.;

X. Hou et al.172

Garcia, J. G.; Mohler, E. R., III. Role of 8-epi PGF2a, 8-isopros-tane, in H2O2-induced derangements of pulmonary artery endo-thelial cell barrier function. Prostaglandins Leukotrienes Essent.Fatty Acids 58:9–16; 1998.

[13] Cracowski, J. L.; Devillier, P.; Durand, T.; Stanke-Labesque, F.;Bessard, G. Vascular biology of the isoprostanes. J. Vasc. Res.38:93–103; 2001.

[14] Hoffman, S. W.; Moore, S.; Ellis, E. F. Isoprostanes: free radical-generated prostaglandins with constrictor effects on cerebral arte-rioles. Stroke 28:844–849; 1997.

[15] Hou, X.; Gobeil, F., Jr.; Peri, K.; Speranza, G.; Marrache, A. M.;Lachapelle, P.; Roberts, L. J., II; Varma, D. R.; Chemtob, S.Augmented vasoconstriction and thromboxane formation by 15-F(2t)-isoprostane (8-iso-prostaglandin F2a) in immature pig peri-ventricular brain microvessels. Stroke 31:516–524; 2000.

[16] Kromer, B. M.; Tippins, J. R. Coronary artery constriction bythe isoprostane 8-epi prostaglandin F2a. Br. J. Phamacol. 119:1276–1280; 1996.

[17] Lahaie, I.; Hardy, P.; Hou, X.; Hassessian, H.; Asselin, P.; La-chapelle, P.; Almazan, G.; Varma, D. R.; Morrow, J. D.; Roberts,L. J., II; Chemtob, S. A novel mechanism for vasoconstrictoraction of 8-isoprostaglandin F2a on retinal vessels. Am. J.Physiol. 274:R1406–R1416; 1998.

[18] Takahashi, K.; Nammour, T. M.; Fukunaga, M.; Ebert, J.; Mor-row, J.D.; Roberts, L. J., II; Badr, K. J. Glomerular actions of afree radical-generated novel prostaglandin, 8-epi-prostaglandinF2a, in the rat: evidence for interaction with thromboxane A2

receptors. J. Clin. Invest. 90:136–141; 1992.[19] Wagner, R. S.; Weare, C.; Jin, N.; Mohler, E. R.; Rhoades, R. A.

Characterization of signal transduction events stimulated by 8-epi-prostaglandin(PG)F2a in rat aortic rings. Prostaglandins 54:581–599; 1997.

[20] Morrow, J. D.; Zackert, W. E.; Yang, J. P.; Kurhts, E. H.; Call-ewaert, D.; Kanai, K.; Taber, D. F.; Moore, K. P.; Oates, J. A.;Roberts, L. J., II. Quantification of the major urinary metabolite of15-F2t-isoprostane (8-iso-PGF2a) by a stable isotope dilution massspectrometric assay. Anal. Biochem. 269:326–331; 1999.

[21] Hou, X.; Roberts, L. J. II; Taber, D. F.; Morrow, J. D.; Kanai, K.;Gobeil, F., Jr.; Beauchamp, M. H.; Bernier, S. G.; Lepage, G.;Varma, D. R.; Chemtob, S. 2,3-Dinor-5,6-dihydro-15-F(2t)-iso-prostane: a bioactive prostanoid metabolite. Am. J. Physiol.281:R391–R400; 2001.

[22] Beauchamp, M. H.; Martinez-Bermudez, A. K.; Gobeil, F., Jr.;Marrache, A. M.; Hou, X.; Speranza, G.; Abran, D.; Quiniou, C.;Lachapelle, P.; Roberts, J., II; Almazan, G.; Varma, D. R.; Chem-tob, S. Role of thromboxane in retinal microvascular degenerationin oxygen-induced retinopathy. J. Appl. Physiol. 90:2279–2288;2001.

[23] Brault, S.; Martinez-Bermudez, A. K.; Marrache, A. M.; Gobeil,F., Jr.; Hou, X.; Beauchamp, M. H.; Quiniou, C.; Almazan, G.;Lachance, C.; Roberts, L. J., II; Varma, D. R.; Chemtob, S. Selec-tive neuromicrovascular endothelial cell death by 8-iso-prosta-glandin F2alpha: possible role in ischemic brain injury. Stroke34:776–782; 2003.

[24] Chemtob, S.; Hardy, P.; Abran, D.; Li, D. Y.; Peri, K.; Cuzzani,O.; Varma, D. R. Peroxide–cyclooxygenase interactions in post-asphyxial changes in retinal and choroidal hemodynamics. J. Appl.Physiol. 78:2039–2046; 1995.

[25] Li, W.; Stephens, N. L. Auxotonic loading and airway smoothmuscle shortening. Can. J. Physiol. Pharmacol. 72:1458–1468;1994.

[26] Hardy, P.; Abran, D.; Hou, X.; Lahaie, I.; Peri, K.; Asselin, P.;Varma, D. R.; Chemtob, S. A major role for prostacyclin in nitricoxide-induced ocular vasorelaxation in the piglet. Circ. Res.83:721–729; 1998.

[27] Merritt, J. E.; Armstrong, W. P.; Benham, C. D.; Hallam, T. J.;Jacob, R.; Jaxa-Chamiec, A.; Leigh, B. K.; McCarthy, S. A.;Moores, K. E.; Rink, T. J. SK&F 96365, a novel inhibitor ofreceptor-mediated calcium entry. Biochem. J. 271:515–522; 1990.

[28] Dumont, I.; Hou, X.; Hardy, P.; Peri, K.; Beauchamp, M. H.;Najarian, T.; Molotchnikoff, S.; Varma, D. R.; Chemtob, S. De-velopmental regulation of endothelial nitric oxide synthase incerebral vessels of newborn pig by prostaglandin E2. J. Pharma-col. Exp. Ther. 291:627–633; 1999.

[29] Berthiaume, N.; Hess, F.; Chen, A.; Regoli, D.; D’Orleans-Juste,P. Pharmacology of kinins in the arterial and venous mesentericbed of normal and B2 knockout transgenic mice. Eur. J. Pharma-col. 333:55–61; 1997.

[30] Okazaki, K.; Endou, M.; Okumura, F. Involvement of barium-sensitive K+ channels in endothelium-dependent vasodilation pro-duced by hypercapnia in rat mesenteric vascular beds. Br. J. Phar-macol. 125:168–174; 1998.

[31] Grynkiewicz, G.; Poenie, M.; Tsien, R. Y. A new generation ofCa2+ indicators with greatly improved fluorescence properties.J. Biol. Chem. 260:3440–3450; 1985.

[32] Taber, D. F.; Kanai, K. Synthesis of the four enantiomerically-pureisomers of 15-F2t-isoprostane. Tetrahedron 54:11767–11782;1998.

[33] Taber, D. F.; Kanai, K.; Pina, R. 5-F2t-isoprostane, a human hor-mone? J. Am. Chem. Soc. 121:7773–7777; 1999.

[34] Taber, D. F.; Xu, M.; Hartnett, J. C. Synthesis of the eight enan-tiomerically-pure diastereomers of the 12-F2-isoprostanes. J. Am.Chem. Soc. 124:13121–13126; 2002.

[35] Li, H.; Lawson, J. A.; Reilly, M.; Adiyaman, M.; Hwang, S. W.;Rokach, J.; FitzGerald, G. A. Quantitative high performanceliquid chromatography/tandem mass spectrometric analysis ofthe four classes of F(2)-isoprostanes in human urine. Proc. Natl.Acad. Sci. USA 96:13381–13386; 1999.

[36] Himmel, H. M.; Whorton, A. R.; Strauss, H. C. Intracellular cal-cium, currents, and stimulus-response coupling in endothelialcells. Hypertension 21:112–127; 1993.

[37] Kinsella, B. T.; O’Mahony, D. J.; Fitzgerald, G. A. The humanthromboxane A2 receptor alpha isoform (TP alpha) functionallycouples to the G proteins Gq and G11 in vivo and is activated bythe isoprostane 8-epi prostaglandin F2 alpha. J. Pharmacol. Exp.Ther. 281:957–964; 1997.

[38] Yura, T.; Fukunaga, M.; Grygorczyk, R.; Makita, N.; Takahashi,K.; Badr, K. F. Molecular and functional evidence for the distinctnature of F2-isoprostane receptors from those of thromboxane A2.Adv. Prostaglandin Thromboxane Leukotriene Res. 23:237–239;1995.

[39] Pratico, D.; Smyth, E. M.; Violi, F.; FitzGerald, G. A. Localamplification of platelet function by 8-epi prostaglandin F2alphais not mediated by thromboxane receptor isoforms. J. Biol. Chem.271:14916–14924; 1996.

[40] Marliere, S.; Cracowski, J. L.; Durand, T.; Chavanon, O.; Bessard,J.; Guy, A.; Stanke-Labesque, F.; Rossi, J. C.; Bessard, G. The 5-series F2-isoprostanes possess no vasomotor effects in the ratthoracic aorta, the human internal mammary artery and the humansaphenous vein. Br. J. Pharmacol. 135:1276–1280; 2002.

[41] Morrow, J. D.; Roberts, L. J.; Daniel, V. C.; Awad, J. A.; Mir-ochnitchenko, O.; Swift, L. L.; Burk, R. F. Comparison of for-mation of D2/E2-isoprostanes and F2-isoprostanes in vitro and invivo—effects of oxygen tension and glutathione. Arch. Biochem.Biophys. 353:160–171; 1998.

[42] Fessel, J. P.; Porter, N. A.; Moore, K. P.; Sheller, J. R.; Roberts,L. J. Discovery of lipid peroxidation products formed in vivowith a substituted tetrahydrofuran ring (isofurans) that are favoredby increased oxygen tension. Proc. Natl. Acad. Sci. USA 99:16713–16718; 2002.

[43] Morrow, J. D.; Roberts, L. J. The isoprostanes: unique bioactiveproducts of lipid peroxidation. Prog. Lipid Res. 36:1–21; 1997.

[44] Hall, E. D. Lazaroids: efficacy and antioxidant mechanism inexperimental cerebral ischemia. In: Krieglstein, J.; Oberpichler,H., eds. Pharmacology of cerebral ischemia. Stuttgart: Wissen-schaftliche Verlagsgesellschaftm; 1990:343–350.

[45] Ogletree, M. L. Overview of physiological and pathophysiologi-cal effects of thromboxane A2. Fed. Proc. 46:133–138; 1987.