A Major Role for Prostacyclin in Nitric Oxide–Induced Ocular Vasorelaxation in the Piglet

From the Centre de Recherche de l'Hôpital Sainte-Justine, Department of Pediatrics and Pharmacology, Université de Montréal (P.H., D.A., X.H., I.L., K.G.P., P.A., S.C.), and the Department of Pharmacology and Therapeutics, McGill University (D.R.V., S.C.), Montréal, Québec, Canada.

Correspondence to Sylvain Chemtob, MD, PhD, Research Center, Hôpital Sainte-Justine, Departments of Pediatrics, Ophthalmology, and Pharmacology, 3175 Côte Sainte-Catherine, Montréal, Québec, Canada H3T 1C5. E-mail chemtobs{at}ere.umontreal.ca

	Abstract

Top Abstract Introduction Materials and Methods Results Discussion References

 
Abstract—We studied the mechanisms of retinal and choroidal vasorelaxation elicited by nitric oxide (NO) using piglet eyes. The NO donors sodium nitroprusside (SNP) and diethylamine-NONOate caused comparable concentration–dependent relaxation that was partially (40%) attenuated by the guanylate cyclase inhibitors methylene blue and LY83583 and reduced to a lesser extent (25%) by the inhibitor of cGMP–dependent kinase, KT 5823. In contrast, NO-induced dilatation (by NO donors and endogenous NO after stimulation with bradykinin) was substantially (70%) diminished by the K_Ca channel blockers tetraethylammonium (TEA), charybdotoxin, and iberiotoxin; by the cyclooxygenase inhibitors indomethacin and ibuprofen; by the prostaglandin I (PGI₂) synthase inhibitor trans-2-phenyl cyclopropylamine (TPC); and by the removal of endothelium; whereas relaxation of endothelium-denuded vasculature to SNP was unaltered by indomethacin, TPC, and charybdotoxin but was nearly nullified by methylene blue and the K_v channel blocker 4-aminopyridine. NO donors significantly increased PGI₂ synthesis and the putative PGI₂ receptor–coupled second messenger cAMP, from ocular vasculature (retinal microvessels and choroidal perfusate), and this increase in PGI₂ formation was markedly reduced by TPC, tetraethylammonium, charybdotoxin, and/or the removal of endothelium, but it was only slightly reduced by methylene blue and LY83583. Also, SNP and K_Ca channel openers NS1619 and NS004 caused an increase in PGI₂ synthesis in cultured endothelial cells, which was virtually abolished by K_Ca blockers. Finally, vasorelaxation to a cGMP analogue, 8-bromo cGMP, and protein kinase G stimulant ß-phenyl-1,N²-etheno-8-bromoguanosine 3':5'-cyclic monophosphate was mostly K_v dependent and, in contrast to NO, largely unrelated to PGI₂ formation. In conclusion, data indicate that NO-induced ocular vasorelaxation is partly mediated by cGMP through its action on smooth muscle, and more importantly, by stimulating PGI₂ formation of endothelial origin via a mechanism mostly independent of guanylate cyclase, which involves the opening of a K_Ca channel.

Key Words: nitric oxide • sodium nitroprusside • prostacyclin • cGMP • K⁺ channel

	Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

 
Nitric oxide (NO) is a potent vasodilator that activates guanylate cyclase resulting in the generation of cGMP, which is presumed to be the principal effector of NO-induced vasorelaxation in various tissues.^{1 2 3 4} However, recent studies on pulmonary, aortic, and intestinal tissue suggest that NO can also cause smooth muscle relaxation by direct activation of K⁺ channels mostly of the calcium-dependent type (K_Ca).^{5 6} NO has also been reported to interact with prostaglandin G/H synthase independent of cGMP, perhaps directly,⁷ although this remains controversial.^{8 9} Bidirectional interactions between prostaglandins and K⁺ channels have been demonstrated such that in some vascular tissue, prostaglandins activate K⁺ channels,^{10 11} but also the opening of K⁺ channels has been reported to stimulate prostaglandin production¹² found to mediate a portion of the vasorelaxation induced by the resulting hyperpolarization.¹³ The physiological significance of this potential interaction between NO, K⁺ channels, and prostaglandins in mediating NO effects on vasculature is far from clear.¹⁴ Regardless, data from current literature suggest that the mechanisms that mediate vasomotor effects of NO in 1 type of vasculature may not necessarily apply to other vascular beds.

As seen in nearly all vasculatures studied to date, NO has been shown to control the tone of the ophthalmic, retinal, and choroidal vasculature.^{15 16 17 18} NO has been implicated in the control of retinal and choroidal blood flow autoregulation.^{17 18} Recently we reported that ocular vasorelaxation to some autocoids was NO-dependent but independent of cGMP^{19 20} ; however, this NO-induced relaxation was significantly reduced by indomethacin,¹⁷ suggesting a possible role for prostaglandins. In the present study, we explored the role of prostaglandins, specifically that of prostaglandin I (PGI₂), in mediating relaxant actions of NO on retinal and choroidal vasculature and in this process investigated the relative contribution of cGMP and K⁺ channels including their potential involvement in the formation of PGI₂.¹² Findings indicate that although cGMP participates in NO-induced ocular vasodilatation through its action on smooth muscle, a more important role is contributed by PGI₂ of endothelial origin via a mechanism mostly independent of guanylate cyclase activity, which involves a K_Ca channel.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

 
Animals
Animals were used according to a protocol approved by the Animal Care Committee of Hôpital Sainte-Justine in accordance with the principles of the Guide to the Care and Use of Experimental Animals and guidelines of the Canadian Council on Animal Care. Yorkshire piglets (2 to 4 days old, n=95) obtained from Fermes Ménard Inc (L'Ange-Gardien, Québec) were anesthetized with halothane (2.5%) and killed by intracardiac injection of pentobarbital (120 mg/kg). Eyes were quickly removed and placed in ice-cold Krebs buffer (pH 7.35 to 7.45) of the following composition (mmol/L): NaCl 120, KCl 4.5, CaCl₂ 2.5, MgSO₄ 1.0, NaHCO₃ 27, KH₂PO₄ 1.0, and glucose 10, to which was added 1.5 U/mL heparin.

Vasomotor Responses of Retinal Vessels
Eyecups were prepared to study the relatively undisturbed retinal vasculature, as previously described.^{21 22} An incision was made at the level of the ora serrata, and the anterior segment and vitreous of the eye were removed. Vertical and horizontal incisions directed toward the optic nerve were made, and the eyecup was pinned to a wax support in a bath containing 20 mL Krebs buffer equilibrated with 21% O₂, 5% CO₂, and 74% N₂ and maintained at 37°C. Eyecups were washed 2 to 3 times with fresh Krebs buffer and allowed to equilibrate for 30 to 45 minutes before starting the experiment.

The effects of the NO donors sodium nitroprusside (SNP) and the NO adduct diethylamine NONOate (NONOate)²³ ; the effects of bradykinin, which are mostly NO-dependent in ocular vasculature²⁴ ; and the effects of the stable analogues of cGMP 8-bromo cGMP and ß-phenyl-1,N²-etheno-8-bromoguanosine-3':5'-cyclic monophosphate (8-bromo PET cGMP), a selective protein kinase G stimulant,²⁵ were studied on the diameter of unperfused retinal arterioles and venules (100 to 200 µmol/L) selected in the field of a dissecting microscope (Zeiss M-400), as previously described.^{21 22} The vasodilator effects of these agents were determined on preparations preconstricted submaximally with the thromboxane A₂ (TXA₂) mimetic U46619 (0.2 µmol/L),^{21 22} which decreased the vessel diameter by 24.6±2.1%. The vascular diameter was recorded with a video camera before and after topical application of increasing concentrations of the agents. Concentrations were increased every 7 minutes, at which time a stable response had been reached. The digitized images were analyzed using a commercial software (Sigma Scan, Jandel Scientific). Each measurement of the diameter was repeated 3 times and had a variability of <1%. Cumulative dose-responses (10^-12 to 10^-5 mol/L) of the different agents were constructed on retinal vessels from eyes of different animals in the absence or the presence of the prostaglandin G/H synthase inhibitors indomethacin (1 µmol/L) and ibuprofen (100 µmol/L), the PGI₂ synthase blocker trans-2-phenylcyclopropylamine (TPC, 5 µmol/L),^{21 26} the guanylate cyclase inhibitors methylene blue (1 µmol/L) and LY83583 (10 µmol/L),²⁷ the protein kinase G antagonist, KT 5823 (1 µmol/L),²⁸ the K_v channel blocker 4-aminopyridine (4-AP, 3 mmol/L),²⁹ and the K_Ca channel blocker tetraethylammonium (TEA, 1 mmol/L).^{5 29} Concentrations of all blockers are consistent with those that inhibit targeted enzymes and channels.^{5 27 28 29}

Measurement of Choroidal Vascular Perfusion Pressure
The mechanisms of action of NO were also evaluated on the choroid, which is a vascular tissue, using a preparation we previously described.^{19 20} A vorticose vessel was catheterized to immediately beyond the sclera using a 27-gauge butterfly needle held in place with cyanoacrylate glue. The catheterized eyeball was placed in a bath containing Krebs buffer (pH 7.4); the buffer was bubbled with a mixture of 21% O₂, 5% CO₂, and 74% N₂ and maintained at 37°C. The choroid was perfused by means of a pulsatile minipump (Gilson) with Krebs buffer at a physiological constant flow rate of 0.20 mL/min to produce a perfusion pressure of 60 mm Hg.^{19 20} Perfusion pressure immediately proximal to the eyeball was continuously recorded using a pressure transducer (Perceptor DT) connected to a Gould multichannel amplifier-recorder (TA 240).

The choroidal vascular bed (with and without endothelium) was perfused for 30 minutes with Krebs buffer for stabilization of the preparation, and endothelium was removed by infusing air in the vasculature, which no longer relaxed to acetylcholine²⁰ but responded normally to endothelium-independent stimulants U46619 and papaverine. Thereafter, Krebs containing SNP or bradykinin (10^-12 to 10^-5 mol/L) was infused with or without pretreatment (30 minutes) with indomethacin (1 µmol/L), TPC (5 µmol/L), methylene blue (1 µmol/L), the K_Ca blockers, charybdotoxin (100 nmol/L) and iberiotoxin (100 nmol/L),^{5 29} the K_ATP blocker glibenclamide (10 µmol/L),^{5 29} the K_v blocker 4-AP (3 mmol/L),²⁹ or N^G-nitro-L-arginine methyl ester (L-NAME, 1 mmol/L); detection of decreases in perfusion pressure (vasorelaxation) did not require pretreatment with U46619. Indomethacin, TPC, methylene blue, L-NAME, and 4-AP produced a small increase (5±2 mm Hg) in perfusion pressure in choroids with intact endothelium. The removal of endothelium also caused a slight increase in perfusion pressure (4±1 mm Hg), which was raised by methylene blue another 3±1 mm Hg. These small increases in perfusion pressure did not alter maximal relaxation to papaverine (0.1 µmol/L). Vasomotor responses were recorded continuously and concentration of stimulants was increased every 10 minutes when responses reached a plateau.

Preparation of Retinal Microvessels
Retinal microvessels were prepared as previously described.³⁰ Retinas were gently homogenized with a Wheaton pestle in 5 mmol/L Tris-HCl buffer (pH 7.4) containing 1.1 mmol/L acetylsalicylic acid, 0.5 mmol/L EGTA, 1 mmol/L benzamidine, 0.1 mmol/L PMSF, and 100 µg/mL of a soybean trypsin inhibitor. The homogenate was mixed with Ficoll 400 (40%) at a 1:1 vol/vol ratio and centrifuged at 20 000g for 20 minutes at 4°C. The pellet, which contains the microvessels was washed in the above buffer 3 times. Purity of the microvessel preparation was confirmed by high-power microscopy and by -glutamyltranspeptidase activity, which was higher in the vessel (5.6 to 6.1 mU/mg protein) than in neural parenchyma (0.3 to 0.35 mU/mg protein).³¹

Retinal Microvascular Endothelial Cell Culture
Endothelium dependence of NO effect on PGI₂ production was equally studied on cultured endothelial cells. Retinal microvessels were suspended in endothelium growth medium (Clonetics) containing gentamicin (5 µg/mL), kanamycin (20 µg/mL), and nystatin (10 U/mL) and placed in a humidified atmosphere with 95% O₂ and 5% CO₂ at 37°C. After a first passage, 80% of the confluent cells were factor VIII (FVIII) positive, and by the second passage virtually 100% were FVIII positive. Other characteristics used for identification of endothelial cells and to differentiate them from smooth muscle cells were their cobblestone morphology at confluence and the negative staining for smooth muscle–specific actin. Cell viability was verified by trypan blue exclusion. Formation of 6-keto-prostaglandin F (PGF₁) was measured on third-passage endothelial cells stimulated with SNP.

Immunostaining for FVIII and smooth muscle actin was performed by fixing cells on cover slips with acetone for 10 seconds and subsequently rehydrating them in PBS for 20 minutes. Fixed cells were incubated for 60 minutes to FVIII or smooth muscle actin antibody (1:50) diluted in PBS containing 10% fetal calf serum and 5% goat serum with 0.1% Triton X-100. The cells were washed 2 to 3 times with PBS and blocked for 15 minutes in PBS containing 0.2% BSA, 5% goat serum, and 0.2% Triton X-100. After 5 washes in PBS, the secondary antibody FITC-conjugated goat anti-rabbit (1:100) antibody was applied under the same conditions, and cells were washed again in PBS and water. Coverslips were then mounted in aqueous mounting medium (Immuno-Mount, Shandon) and examined under an epifluorescent microscope (Leitz Diaplan).

Prostanoid Assays
Retinal microvessels (600 to 800 µg of protein) were suspended in 50 mmol/L Tris-HCl buffer (pH 7.4) containing 1 mmol/L PMSF, 1.5 µmol/L pepstatin A, 0.2 mmol/L leupeptin, and 100 µg/mL soybean trypsin inhibitor. The tissues were preincubated for 20 minutes with arachidonic acid (5 µmol/L) at 37°C in the absence or presence of indomethacin (1 µmol/L), ibuprofen (100 µmol/L), TPC (5 µmol/L), methylene blue (1 µmol/L), LY83583 (10 µmol/L), KT 5823 (1 µmol/L), 4-AP (3 mmol/L), and TEA (1 mmol/L) before the addition of SNP (0.1 µmol/L), NONOate (0.1 µmol/L), 8-bromo cGMP (10 µmol/L), or 8-bromo PET cGMP (10 µmol/L). The dose response to stimulants (SNP, NONOate, 8-bromo cGMP, and 8-bromo PET cGMP) was also studied. The reaction was stopped in boiling water (5 minutes). Preparations were then centrifuged at 2000g for 20 minutes. Using a similar protocol, prostaglandin production was also measured on cultured endothelial cells. 6-Keto-prostaglandin F (PGF₁₎ (stable prostaglandin I (PGI₂) metabolite), prostaglandin E (PGE₂), and prostaglandin D (PGD₂₎ were determined on the supernatant by radioimmunoassay, as previously described in detail,^{19 32} and protein was determined on the pellet. As expected, basal levels of both PGE₂ and PGD₂ (in the absence of stimulation with SNP or NONOate) were reduced only by indomethacin and ibuprofen (<70 pg/mg protein per minute for PGE₂ and <3 pg/mg protein per minute for PGD₂), and 6-keto PGF₁ only by indomethacin, ibuprofen, and TPC (<20 pg/mg protein per minute).

The effect of the NO donor SNP on prostaglandins was also measured on perfusate from choroids with intact and denuded of endothelium in the presence or absence of K⁺ channel blockers. For this purpose, prostaglandins were measured on 10-minute collections of perfusate and expressed as a function of choroid protein content. Net production of prostaglandins was calculated after correction for basal synthesis in the absence of the NO donor.

cAMP and cGMP Assays
For cAMP and cGMP assays, retinal microvessels were incubated at 37°C for 5 minutes with SNP (10 nmol/L) and diethylamine (DEA) (100 nmol/L) in Tris-HCl buffer 10 mmol/L (pH 8.0) containing (mmol/L) ATP 1, MgCl₂ 7.5, creatine phosphate 15, EGTA 0.5, isobutyl methylxanthine 0.5, DTT 1, benzamidine 1, PMSF 0.1, creatine phosphokinase 185 U/mL, acetylsalicylic acid 200 µg/mL, and soybean trypsin inhibitor 100 µg/mL; tissues were pretreated or not (20 minutes) with TPC (5 µmol/L). Microvessels were then homogenized (Omni) to measure cAMP and cGMP by radioimmunoassay using commercial kits, as previously reported.^{19 20 30} Net cAMP and cGMP production stimulated by test agents was calculated after correction for basal production in the absence of stimulants, which was for cAMP 8.1±1.4 pmol/mg protein per minute and for cGMP 7.0±1.9 pmol/mg protein per minute.

Chemicals
NS004 was generously provided by Dr Søren-Peter Olesen (NeuroSearch, Glostrup, Denmark). The following agents or items were purchased: diethylamine NONOate (Cayman Chemicals); 8-bromo cGMP and 8-bromo PET cGMP (Biolog); SNP, L-NAME, indomethacin, bradykinin, ibuprofen, TPC, tetraethylammonium chloride (TEAC), charybdotoxin, iberiotoxin, 4-AP, acetylsalicylic acid, U46619, acetylcholine, papaverine, arachidonic acid, soybean trypsin inhibitor (type II-S), benzamidine, PMSF, DTT, creatine phosphokinase, EGTA, EDTA, and isobutyl methylxanthine (Sigma Chemical); LY83583 and KT 5823 (Calbiochem); NS1619 (Research Biochemicals International); radioimmunoassay kits for 6-keto PGF₁ and PGE₂ (Advanced Magnetics); cGMP assay kits (Amersham); cAMP assay kits (Diagnostic Product Corporation); endothelial cell medium (Clonetics); [FITC]-conjugated goat anti-rabbit antibody, fetal calf serum, goat serum (Jackson Immunoresearch Laboratories); FVIII antibody and smooth muscle–specific actin antibody (Dako); and all other high-purity chemicals (Fisher Scientific).

Data Analysis
Data were analyzed using a Student t test or 2-way ANOVA, including factoring for concentrations and drugs. Post-ANOVA comparisons among means were performed using the Tukey-Kramer method. Statistical significance was set at P<0.05. Data are presented as mean±SEM.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

 
Effects of NO Donors on Retinal Vessel Diameter
SNP and NONOate produced a concentration-dependent relaxation of retinal arterioles and venules (Figure 1). The maximal response to both NO donors was comparable on arterioles and venules, but SNP was more potent than NONOate, with EC₅₀ values in arterioles and venules, respectively, of 0.16±0.02 and 0.13±0.01 nmol/L for SNP and 12.6±0.1 and 7.9±0.1 nmol/L for NONOate.

View larger version (18K): [in this window] [in a new window]   Figure 1. Vasorelaxant concentration-response to SNP (filled symbols) and NONOate (open symbols) of retinal arterioles (squares) and venules (circles). Relaxant effects are expressed as percent reversal of U46619 (0.2 µmol/L)-induced vasoconstriction. Data are mean±SEM of 4 to 6 experiments.

Indomethacin, ibuprofen, the PGI₂ synthase inhibitor TPC, the K_Ca blocker TEA, and the cGMP-dependent protein kinase inhibitor KT 5823, did not alter the basal diameter of retinal vessels. Guanylate cyclase inhibitors methylene blue and LY83583 and the K_v channel blocker 4-AP produced slight constriction (3.1±0.4%). All 8 blockers partially inhibited vasorelaxant effects of SNP and NONOate comparably on arterioles and venules (Figure 2). KT 5823 inhibited 25% to 30% of NO-induced dilatation, and methylene blue and LY83583 inhibited 35% to 45% of the relaxation by NO; indomethacin, ibuprofen, and TPC blocked 65% to 75% of the relaxation. A combination of guanylate cyclase and prostaglandin G/H or PGI₂ synthase inhibitors blocked nearly 90% of NO-induced dilatation.

View larger version (37K): [in this window] [in a new window]   Figure 2. Vasorelaxant effects of SNP (open bars) and NONOate (hatched bars) on retinal arterioles (A) and venules (B). Tissues were preconstricted with U46619 (0.2 µmol/L) and pretreated with the following agents before the addition of SNP (10 nmol/L) and NONOate (100 nmol/L): methylene blue (MB, 1 µmol/L), LY83583 (LY, 10 µmol/L), KT 5823 (KT, 1 µmol/L), tetraethylammonium (TEA, 1 mmol/L), indomethacin (Indo, 1 µmol/L), ibuprofen (Ibu, 100 µmol/L), and/or trans-2-phenyl cyclopropylamine (TPC, 5 µmol/L), and saline. Data are mean±SEM of 4 to 6 experiments expressed as percent reversal of U46619-induced constriction. *P<0.01 compared with other values.

Effects of NO on Prostaglandin Production in Retinal Microvessels
SNP and NONOate caused a concentration-dependent increase in 6-keto PGF₁ production by isolated retinal microvessels (Figure 3). In contrast, PGE₂ synthesis slightly decreased after stimulation with SNP and NONOate; likewise, PGD₂ synthesis was not stimulated by SNP and NONOate (basal, 15.0±2.0 pg/mg protein per minute, and after NONOate (100 nmol/L), 13.8±2.3 pg/mg protein per minute). Prostaglandin G/H synthase inhibitors indomethacin and ibuprofen reduced the synthesis of prostaglandins, and the PGI₂ synthase inhibitor TPC as well as the K_Ca channel blocker TEA selectively and equivalently decreased only that of 6-keto PGF₁ after stimulation by NO donors. Of all the inhibitors, and as expected, only prostaglandin synthase blockers reduced the basal levels (in the absence of stimulation with SNP or NONOate) of prostaglandins. Guanylate cyclase inhibitors methylene blue (1 µmol/L) and LY83583 (10 µmol/L) slightly decreased NO donor–induced production of 6-keto PGF₁ but not of PGE₂; whereas consistent with its vasomotor effects, the cGMP-dependent protein kinase inhibitor KT 5823 (1 µmol/L), did not affect prostaglandin formation; 5-fold higher concentrations of methylene blue, LY83583, and KT 5823 did not reduce further 6-keto PGF₁ generation (data not shown). Thus, in retinal vasculature, NO donors induce PGI₂ synthesis mostly independent of cGMP and protein kinase G but apparently largely dependent of K_Ca.

View larger version (39K): [in this window] [in a new window]   Figure 3. Effects of SNP (open bars) and NONOate (hatched bars) on 6-keto PGF1 (A) and PGE2 (B) synthesis in isolated retinal microvessels. Dose response to SNP and NONOate and the effects of SNP and NONOate were studied in the absence or presence of the following agents: MB (1 µmol/L), LY (10 µmol/L), Indo (1 µmol/L), Ibu (100 µmol/L), TPC (5 µmol/L), TEA (1 mmol/L), and KT (1 µmol/L). As expected, basal levels (unstimulated with SNP or NONOate) of PGE2 were reduced only by Indo and Ibu (<70 pg/mg protein per minute), and 6-keto PGF1 only by Indo, Ibu, and TPC (<20 pg/mg per protein). Data are mean±SEM of 4 to 6 experiments. *P<0.05 compared with corresponding control values (SNP or NONOate after saline pretreatment); P<0.05 compared with basal values. Abbreviations as in Figure 2.

Effects of cGMP Analogues on Retinal Vascular Relaxation and PGI₂ Production
To further assess the role of cGMP and protein kinase G on retinal vasorelaxation and their interaction with prostaglandins, the effects of cGMP analogues were studied on vasomotricity and PGI₂ synthesis. 8-Bromo cGMP caused a dose-dependent vasorelaxation that was minimally attenuated by indomethacin and TPC, and was markedly diminished by the protein kinase G inhibitor KT 5823, the K_v blocker 4-AP, and the combination of 4-AP and KT 5823, but was unaffected by TEA (Table). 8-Bromo cGMP also caused a small stimulation of 6-keto PGF₁ production that was markedly reduced by indomethacin and TPC but was unaffected by KT 5823, 4-AP, and TEA. This suggested that PGI₂ synthesis by cGMP is independent of protein kinase G and activation of K_Ca and K_v channels. Along the same lines and consistent with results presented in Figures 2 and 3 regarding the role of protein kinase G, selective stimulation of this cGMP-dependent kinase with 8-bromo PET cGMP²⁵ caused negligible generation of 6-keto PGF₁, and a vasodilatation unaltered by prostaglandin G/H and PGI₂ synthase inhibitors, which was reversed by KT 5823 and/or 4-AP. Thus, results with analogues of cGMP suggest that ocular vasorelaxation to this cyclic nucleotide is largely unrelated to PGI₂ (and to K_Ca channels), in contrast to effects elicited by NO (Figures 2 and 3), but seems to depend on K_v channel activation (Table).

View this table: [in this window] [in a new window]   Table 1. Effects of 8-Bromo cGMP and 8-Bromo PET cGMP on Retinal Vessel Relaxation and Net Production of 6-Keto PGF1

Effects of NO on cAMP and cGMP Production in Retinal Vessels
Further evidence that NO mediates part of its ocular vascular effects via PGI₂ was provided by measuring the putative second messenger to the PGI₂ receptor, namely cAMP. SNP and NONOate stimulated cAMP production in retinal microvessels. This was markedly reduced by TPC (Figure 4), which did not affect basal production of cAMP (in the absence of NO donors). As expected, NO donors also increased cGMP formation.

View larger version (17K): [in this window] [in a new window]   Figure 4. Effects of SNP (open bars) and NONOate (hatched bars) on net cAMP and cGMP production in isolated retinal microvessels; tissues were pretreated with TPC (5 µmol/L) or saline before the addition of SNP (10 nmol/L) and NONOate (100 nmol/L). Net stimulated production by test agents was calculated after correction for basal production in the absence of NO donors, which was for cAMP 8.1±1.4 pmol/mg protein per minute, and for cGMP 7.0±1.9 pmol/mg protein per minute; TPC did not modify basal cAMP production in the absence of NO donors. Values are mean±SEM of 4 experiments, each performed in duplicate: * P<0.01 compared with corresponding control values (saline).

Role of PGI₂, Guanylate Cyclase, and K_Ca on NO-Induced Relaxation in Perfused Choroid
The role of PGI₂, guanylate cyclase, and K_Ca on SNP-induced vasorelaxation was also studied on a separate ocular tissue, specifically, the choroid, which is totally of a vascular nature. Consistent with observations made on the retinal vasculature (Figure 2), SNP-induced choroidal relaxation was inhibited by 75% by indomethacin, TPC, and the K_Ca blockers charybdotoxin and iberiotoxin. The SNP-induced choroidal relaxation was unaffected by the K_ATP blocker glibenclamide and was reduced by nearly 40% by methylene blue and 4-AP (Figure 5A). The removal of endothelium (by infusing air, which abolished relaxation to acetylcholine but not to papaverine and did not affect contraction to U46619) resulted in diminished relaxation to SNP, which under these conditions was unaltered by indomethacin, TPC, and charybdotoxin, but was markedly reduced further by methylene blue and 4-AP. Moreover, the absence of functional endothelium vasorelaxation to SNP was comparable to that in endothelialized choroid treated with prostaglandin synthesis inhibitors indomethacin and TPC or K_Ca blockers charybdotoxin and iberiotoxin (Figure 5A). In addition, on choroid denuded of endothelium, 8-bromo cGMP (10 µmol/L)–elicited vasorelaxation was unaffected by TPC and charybdotoxin but virtually abrogated by 4-AP (data not shown). Hence, (1) removal of endothelium eliminated the PGI₂ and K_Ca dependence of SNP-induced relaxation, and (2) the preponderant role of guanylate cyclase seems to be on the smooth muscle via a K_v channel (Table).

View larger version (30K): [in this window] [in a new window]   Figure 5. A, Effects of SNP on perfusion pressure of infused choroids containing (open bars) or devoid of (hatched bars) functional endothelium in the presence or absence of prostaglandin synthase, guanylate cyclase, and KCa, KV, and KATP channel blockers. Choroids were infused with Krebs buffer at a physiological constant flow rate of 0.2 mL/min, producing a perfusion pressure of 60 mm Hg19,20; a negative change in perfusion pressure reflects vasorelaxation. Endothelium was removed by infusing air that abolished relaxation to acetylcholine but not to papaverine and did not affect contraction to U46619. Krebs-containing SNP was infused with or without pretreatment (30 minutes) with Indo (1 µmol/L), TPC (5 µmol/L), MB (1 µmol/L), CTX (100 nmol/L), IBX (100 nmol/L), 4-AP (3 mmol/L), or glibenclamide (Glib, 10 µmol/L). Indo, TPC, MB, and 4-AP produced a small increase (5±2 mm Hg) in perfusion pressure in choroids with intact endothelium. Removal of endothelium also caused a slight increase in perfusion pressure (4±1 mm Hg) that was raised by MB by another 3±1 mm Hg. These small increases in perfusion pressure did not alter maximal relaxation to papaverine (0.1 µmol/L). Values are mean±SEM of 4 experiments. *P<0.01 compared to corresponding saline; P<0.01 compared to all other values of SNP (10 nmol/L)-treated preparations, except those with Glib. B, SNP-stimulated net production of 6-keto PGF1 in choroidal perfusate. 6-Keto PGF1 was measured on 10-minute collections of perfusate at infusion rate described in A; concentrations of agents and other details are described in A. Net production stimulated by SNP was calculated after correction for basal production in the absence of NO donor (67.4±9.8 pg/mg protein per minute). Values are mean±SEM of 4 experiments and are expressed as a function of choroid protein content. Endo refers to endothelium. *P<0.01 compared with corresponding saline. C, SNP-stimulated synthesis of 6-keto PGF1 on retinal endothelial cells. Endothelial cells were cultured from isolated retinal vessels. Dose response to SNP was studied on cells (third passage) loaded with arachidonic acid (5 µmol/L) for 20 minutes. Effects of CTX (100 nmol/L), IBX (100 nmol/L), 4-AP (3 mmol/L), or Glib (10 µmol/L) before the addition of SNP (10 nmol/L), NS1619 (1 µmol/L), or NS004 (1 µmol/L) was also studied. Net stimulated production by test agents (SNP, NS1619, or NS004) was calculated after correction for basal production in the absence of stimulants (43.5±1.2 pg/mg protein per minute); K+ channel blockers did not modify basal production of 6-keto PGF1 in the absence of SNP, NS1619, or NS004. Values are mean±SEM of 3 experiments. *P<0.01 compared with corresponding saline. Abbreviations as in Figure 2.

Choroidal production of 6-keto PGF₁ stimulated by SNP was dose dependent. The SNP-induced generation of 6-keto PGF₁ was markedly reduced to the same extent by the removal of endothelium and charybdotoxin but not by glibenclamide or 4-AP (Figure 5B), and charybdotoxin did not further reduce 6-keto PGF₁ formation in the absence of functional endothelium. Likewise, 6-keto PGF₁ synthesis by cultured retinal endothelial cells in response to SNP was nearly nullified by charybdotoxin and iberiotoxin but not by glibenclamide or 4-AP (Figure 5C). In addition, the specific K_Ca channel openers NS1619 and NS004³³ also stimulated synthesis of 6-keto PGF₁; this was virtually abolished by iberiotoxin. K⁺ channel blockers did not modify basal (in the absence of SNP, NS1619, or NS004) production of 6-keto PGF₁. Thus, it can be inferred that NO-induced production of PGI₂ by ocular vasculature involves the opening of a K_Ca channel present on endothelial cells.

Role of PGI₂ in Bradykinin-Elicited Ocular Vasorelaxation
Finally, the contribution of PGI₂ on vasomotor response to endogenously released NO was also studied. For this purpose, the effects of bradykinin, which exerts most of its ocular vasorelaxation via NO,²⁴ were tested on retinal and choroidal vasculature. Bradykinin caused a dose-dependent retinal and choroidal vasorelaxation that was nearly nullified by the NO synthase inhibitor L-NAME and markedly decreased to an equivalent degree by prostaglandin G/H and PGI₂ synthase inhibitors indomethacin and TPC (Figure 6A and 6B).

View larger version (17K): [in this window] [in a new window]   Figure 6. Contribution of prostacyclin in bradykinin (Bk)-evoked retinal (A) and choroidal (B) vasorelaxation. Dose response to bradykinin and the effects of pretreatment with Indo (1 µmol/L), TPC (5 µmol/L), and L-NAME (1 mmol/L) were studied as described in Figures 2 and 5A. Values are mean±SEM of 4 experiments. *P<0.01 compared with corresponding saline value; P<0.05 compared with all other values.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

 
NO plays a major role in governing important hemodynamic responses in the eye. Specifically, NO exerts a significant role in the autoregulation of retinal and choroidal blood in response to acute increases in perfusion pressure¹⁷ and to hyperoxia.¹⁸ Vasomotor responses to various endogenous agents in the eye have also been found to be NO-dependent.^{19 20 24} But the mechanisms of action of NO in ocular vasculature have not been elucidated. Although cGMP has been generally suggested to mediate the vasomotor actions of NO in vasculature,^{1 2 3 4} this cyclic nucleotide may not be a universal effector of NO across all vascular beds. For instance, it has been reported for aorta and lung vasculature that vasorelaxation to NO may be mediated by activating K⁺ channels,^{5 6 34} and for skin¹⁴ and eyes to perhaps depend on prostaglandins.¹⁷ Interactions between K⁺ channels and prostaglandins have also been documented.^{10 11 13 35 36} We, therefore, explored the relative contributions of interactions between K⁺ channels, prostaglandins, and cGMP in mediating NO-evoked dilatation of ocular vasculature. Using retinal and choroidal vessel preparations as well as isolated cultured endothelial cells, we found that NO-induced relaxation of these oculovascular beds is partly mediated by cGMP but more importantly is dependent on PGI₂ formation by the endothelium through a mechanism mostly independent of guanylate cyclase, which involves opening a K_Ca channel. In this regard, we are not aware of any report disclosing a comparably major role for PGI₂ in mediating NO-induced dilatation in other blood vessels.

The PGI₂-dependence of NO action was observed in both retinal arterioles and venules (Figure 2) as well as on choroidal vasculature (Figure 5A). The relaxant effects of NO in these vessel beds was tested using 2 distinct NO donors, SNP and NONOate, which exhibited comparable efficacy (Figure 1), as well as with an agent which releases endogenous NO, bradykinin (Figure 6). In addition, relaxant response to all 3 compounds was equivalently reduced by 70% by the molecularly unrelated inhibitors of prostaglandin G/H synthase indomethacin and ibuprofen, as well as specifically by the PGI₂ synthase blocker TPC^{21 26} (Figures 2, 5A, and 6).

The major finding in this study is the important role of PGI₂ in mediating NO-induced relaxation of ocular vessels. This inference is based on the observations (1) Prostaglandin G/H synthase inhibitors indomethacin and ibuprofen reduced NO donor–elicited retinal and choroidal vasodilatation by 70%, which is similar to the effect produced by the specific PGI₂ synthase blocker TPC^{21 26} (Figures 2 and 5A). (2) Vasodilatory response to endogenously-released NO, after stimulation with bradykinin (inhibitable by L-NAME) was also blunted by indomethacin as well as TPC (Figure 6). This was similar to the extent that relaxation to NO donors was reduced by these prostaglandin synthase inhibitors (Figures 2 and 5A). (3) The effect of NO donors on the generation of cAMP, a second messenger for the PGI₂ receptor,³⁷ was markedly reduced by TPC (Figure 4). (4) NO donors stimulated dose-dependently the production of PGI₂ (measured by its stable metabolite 6-keto PGF₁) by retinal and choroidal vasculature as well as by retinal endothelial cells, and this formation of PGI₂ was inhibited by all 3 prostaglandin synthase blockers, indomethacin, ibuprofen, and TPC (Figures 3, 5B, and 5C). Thus, NO causes the formation of PGI₂,_, which contributes significantly to NO-induced ocular vasorelaxation.

Because in vasculature PGI₂ is believed to originate mostly from endothelium,³⁸ we verified this conjecture. Indeed, SNP-induced 6-keto PGF₁ production was virtually abolished by endothelium-denuded choroidal vasculature and was found to be produced directly by retinal-vessel endothelial cells (Figure 5B and 5C). At the functional level, the removal of endothelium eliminated PGI₂-dependent vasorelaxation induced by NO donors (Figure 5A). Our findings are consistent with the NO-induced formation of PGI₂ by coronary endothelial cells,³⁹ although mechanisms of NO-stimulated PGI₂ synthesis in these reports were not elucidated.

In contrast to PGI₂, the synthesis of PGE₂ and PGD₂ was not stimulated by NO donors (Figure 3). The precise reasons for this selective increase in NO-induced PGI₂ synthesis in ocular vasculature are not fully clear. Although stimulation of cyclooxygenase is generally associated with a rise in all prostanoids, the divergent production of prostanoids in response to various stimuli, including NO, has been reported.^{21 40 41} NO has also been shown to stimulate PGE₂ formation in certain cells^{42 43} but not in others.^{9 44} In vascular cells, the shear stress that releases NO was reported to elicit PGI₂ but not PGE₂ synthesis,⁴⁵ consistent with the present observations. Differences in the tissue expression and independent regulation of PGE₂, PGD₂, PGI_2, and TXA₂ synthases from that of cyclooxygenase may explain variable profiles of prostanoid formation.^{46 47}

A salient feature of this study is the significant contribution of calcium-dependent K⁺ channels in NO-induced ocular vasorelaxation, which requires the formation of PGI₂. Evidence for this inference is provided by various observations. First, 2 general blockers of K_Ca channels, TEA and charybdotoxin, as well as a specific blocker of the large conductance K_Ca channel, iberiotoxin, but not the K_ATP channel blocker glibenclamide²⁹ markedly reduced ocular vasorelaxation to NO donors to the same extent as blockers of prostaglandin G/H and PGI₂ synthase (Figures 2 and 5A). Secondly, SNP- and NONOate-stimulated formation of PGI₂ by retinal and choroidal vessels as well as by endothelial cells (the principal source of PGI₂) was significantly diminished by K_Ca but not K_ATP or K_v blockers (Figures 3, 5B, and 5C); accordingly, the removal of endothelium abolished the inhibitory effects of charybdotoxin on vasorelaxation to a NO donor (Figure 5B). More direct evidence that the opening of K_Ca channels leads to PGI₂ formation was obtained with the specific K_Ca openers NS1619 and NS004³³ and suggests that hyperpolarization evoked by these K_Ca channel openers induce PGI₂ generation from endothelium (Figure 5C). This role of K_Ca channels may also explain recently reported stimulation of PGI₂ formation by NO in coronary endothelial cells.³⁹ Third, in contrast to K_Ca channel blockers, PGI₂ synthesis was minimally reduced by the inhibition of guanylate cyclase (Figure 3) probably by endothelium.⁴⁸ In addition, K_Ca-dependent relaxation and PGI₂ formation by NO appear to be independent of cGMP, because cGMP analogue–elicited vasodilatation and PGI₂ synthesis was unaffected by K_Ca blockers (Table). These observations indicate that NO, but not cGMP, causes the opening of K_Ca channels, apparently of large conductance, that are present on the endothelium.⁴⁹ This results in PGI₂ formation largely responsible for the NO-evoked ocular vasorelaxation. Although the precise mode of interaction between NO and K_Ca was not investigated in the present study, a direct activation of K_Ca by NO has been suggested.^{5 6} Thus, our findings imply that NO elicits hyperpolarization of endothelium, which in turn leads to PGI₂ synthesis. This premise concurs with similar observations by others,¹² presumably by increasing intracellular calcium via non–voltage dependent calcium channels.^{49 50} On the basis of this inference, one cannot exclude activation of NO synthase by NO; however, this would suggest an unstable positive feedback physiological situation.

Although cGMP exhibits a minor role in NO-induced PGI₂ formation compared with K_Ca (Figure 3), it retains a contribution in NO-evoked ocular vasorelaxation (albeit small, relative to that by PGI₂) (Figure 2) by acting mostly on the smooth muscle. Indeed, on endothelialized vasculature, guanylate cyclase inhibitors methylene blue and LY83583 reduced vasorelaxation to NO by 40% (Figures 2 and 5A), and on vasculature denuded of functional endothelium, methylene blue nearly nullified the vasomotor effects of NO (Figure 5A). Because vasodilatation to cGMP analogues was virtually abolished by the cGMP-dependent kinase inhibitor KT 5823 (Table), it can be inferred from the data that cGMP exerts its major action on the smooth muscle of ocular vessels via protein kinase G. To further elucidate the action of NO and cGMP on ocular vascular smooth muscle, we examined the role of K⁺ channels.^{5 6 51 52 53 54} On endothelialized vasculature, the K_v channel blocker 4-AP (but not K_ATP blockers), like the guanylate cyclase inhibitors, reduced vasorelaxation to NO by 35% to 40% (Figures 2 and 5A) and almost abolished the vasodilatation induced by cGMP analogues (Table). On vasculature denuded of endothelium, 4-AP (but not K_Ca blockers), as seen with guanylate cyclase inhibitors, almost completely eliminated the vasomotor effects of NO (Figure 5A). Likewise, vascular smooth muscle relaxation to cGMP was found to be dependent of the K_v channels.⁵¹ These results indicate that in vascular smooth muscle of the retina and choroid, the NO-evoked ocular vasorelaxation is not due to activation of K_Ca channels but rather to that of K_v channels.

The absence of involvement of K_Ca channels on the relaxant response of smooth muscle of ocular vasculature to NO differs from that described in most other vascular beds.^{5 34 55 56} However, such a major role for K_Ca channels in mediating NO-induced vasorelaxation has not been universally observed,^{52 53 57} including in a comparable porcine model.^{53 58} Consistent with our findings in some tissues, the effect of NO on vasculature has been found to be K_v dependent.^{54 57} This diversity in mechanisms of action of NO may be due to tissue as well as regional specificity in the expression of ion channels in vascular smooth muscles.²⁹ Along the same lines, although in heart and lung a significant role for K_ATP channels has been reported in vasodilatation to PGI₂,^{10 11 59} our data with glibenclamide in ocular vasculature do not support an important role for K_ATP (Figure 5A and 5B).

In summary, our findings in the ocular vasculature disclose a previously undescribed cascade of events leading to relaxation in response to NO. Based on our data, we propose a model, depicted in Figure 7, in which NO activates mainly the opening of K_Ca channels and to a smaller extent cGMP generation in the endothelium, which result in PGI₂ synthesis released to act on its receptor on the smooth muscle³⁷ to evoke the predominant action of NO. Notwithstanding, in the smooth muscle, cGMP and its dependent kinase participate nonnegligibly in NO action mostly via K_v channels, although an effect of cGMP on ocular vasorelaxation independent of K_v channels cannot be excluded (Table). Together PGI₂ and cGMP elicit nearly all vasorelaxation to NO in the eye (Figure 2). Because NO and PGI₂ are important mediators of vasomotor tone, the similar marked improvement in ocular blood flow autoregulation observed after cyclooxygenase as well as NO synthase inhibitors in young animals^{17 18 41} may be explained by the intimate interaction between NO and PGI₂ in retinal and choroidal vasculature presented in the present study.

View larger version (18K): [in this window] [in a new window]   Figure 7. Model based on data presented, depicting mechanism of NO-induced ocular vasorelaxation involving interactions between cGMP, Kv channels, KCa channels, and PGI2. AC, GC, and PKG refer to adenylate cyclase, guanylate cyclase, and cGMP-dependent protein kinase, respectively.

	Acknowledgments

 
The authors thank Dr Søren-Peter Olesen from NeuroSearch (Glostrup, Denmark) for providing NS004. We are also grateful for technical assistance by Hendrika Fernandez. This work was supported by grants from the Medical Research Council of Canada, the Heart and Stroke Foundation of Quebec, the March of Dimes Birth Defects Foundation, the Hospital for Sick Children Foundation, the United Cerebral Palsy Foundation, and the Fonds de la Recherche en Santé du Québec. Pierre Hardy is the recipient of a fellowship award from the Medical Research Council of Canada, and Isabelle Lahaie received a studentship from Fight for Sight.

Received August 1, 1997; accepted July 15, 1998.

	References

Top Abstract Introduction Materials and Methods Results Discussion References

 

1. Nathan C. Nitric oxide as a secretory product of mammalian cells. FASEB J. 1992;6:3051–3064.[Abstract]
2. Lowenstein CJ, Dinerman JL, Snyder SH. Nitric oxide: a physiologic messenger. Ann Intern Med. 1994;120:121–134.
3. Ignarro LJ. Biosynthesis and metabolism of endothelium-derived nitric oxide. Annu Rev Pharmacol Toxicol. 1990;30:535–560.[Medline] [Order article via Infotrieve]
4. Moncada S, Palmer MJ, Higgs EA. Nitric oxide: physiology, pathophysiology, and pharmacology. Pharmacol Rev. 1990;43:109–142.[Medline] [Order article via Infotrieve]
5. Bolotina VM, Najibi S, Palacino JJ, Pagano PJ, Cohen RA. Nitric oxide directly activates calcium-dependent potassium channels in vascular smooth muscle. Nature. 1994;368:850–853.[Medline] [Order article via Infotrieve]
6. Koh SD, Campbell JD, Carl A, Sanders KM. Nitric oxide activate multiple potassium channels in canine colonic smooth muscle. J Physiol (Lond). 1995;489:735–743.[Medline] [Order article via Infotrieve]
7. Salvemini D, Currie MG, Mollace V. Nitric oxide-mediated cyclooxygenase activation: a key event in the antiplatelet effects of nitrovasodilators. J Clin Invest. 1996;7:2562–2568.
8. Tsai AL, Wei C, Kulmacz RJ. Interaction between nitric oxide and prostaglandin H synthase. Arch Biochem Biophys. 1994;313:367–372.[Medline] [Order article via Infotrieve]
9. Curtis JF, Reddy NG, Mason RP, Kalyanaraman B, Eling TE. Nitric oxide: a prostaglandin H synthase 1 and 2 reducing cosubstrate that does not stimulate cyclooxygenase activity or prostaglandin H synthase expression in murine macrophages. Arch Biochem Biophys. 1996;335:369–376.[Medline] [Order article via Infotrieve]
10. Bouchard JF, Dumont E, Lamontagne D. Evidence that prostaglandin I₂, E₂, and D₂ may activate ATP sensitive potassium channels in the isolated rat heart. Cardiovasc Res. 1994;28:901–905.[Abstract/Free Full Text]
11. Jackson WF, Konig A, Dambacher T, Busse R. Prostacyclin-induced vasodilation in rabbit heart is mediated by ATP-sensitive potassium channels. Am J Physiol. 1993;264:H238–H243.[Abstract/Free Full Text]
12. Cohen RA, Vanhoutte PM. Endothelium-dependent hyperpolarization: beyond nitric oxide and cyclic GMP. Circulation. 1995;92:3337–3349.[Free Full Text]
13. Siegel G, Mironneau J, Schnalke F, Schroder G, Schulz BG, Grote J. Vasodilation evoked by K⁺ channel opening. Prog Clin Biol Res. 1990;327:229–306.[Medline] [Order article via Infotrieve]
14. Di Rosa M, Ialenti A, Ianaro A, Sautebin L. Interaction between nitric oxide and cyclooxygenase pathways. Prostaglandins Leukot Essent Fatty Acids. 1996;54:229–238.[Medline] [Order article via Infotrieve]
15. Yao K, Tschudi M, Flammer J, Luscher, TF. Endothelium-dependent regulation of vascular tone of the porcine ophthalmic artery. Invest Ophthalmol Vis Sci. 1991;32:1791–1798.[Abstract/Free Full Text]
16. Donati G, Pournaras CJ, Munoz JL, Poitry S, Poitry-Yamate CL, Tsacopoulos M. Nitric oxide controls arteriolar tone in the retina of the miniature pig. Invest Ophthalmol Vis Sci. 1995;36:2228–2237.[Abstract/Free Full Text]
17. Hardy P, Nuyt AM, Abran D, St-Louis J, Varma DR, Chemtob S. Nitric oxide in retinal and choroidal blood flow autoregulation in newborn pigs: interaction with prostaglandins. Pediatr Res. 1996;39:487–493.[Medline] [Order article via Infotrieve]
18. Hardy P, Peri KG, Lahaie I, Varma DR, Chemtob S. Increased nitric oxide synthesis and action preclude choroidal vasoconstriction to hyperoxia in newborn pigs. Circ Res. 1996;79:504–511.[Abstract/Free Full Text]
19. Abran D, Varma DR, Chemtob S. Regulation of prostanoid vasomotor effects and receptors in choroidal vessels of newborn pigs. Am J Physiol. 1997;272:R995–R1001.[Abstract/Free Full Text]
20. Abran D, Dumont I, Hardy P, Peri K, Li D-Y, Molotchnikoff S, Varma DR, Chemtob S. Characterization and regulation of PGE₂ receptor and receptor-coupled functions in the choroidal vasculature of the pig during development. Circ Res. 1997;80:463–472.
21. Abran D, Hardy P, Varma DR, Chemtob S. Mechanism of the biphasic effects of peroxides on the retinal vasculature of newborn and adult pigs. Exp Eye Res. 1995;61:285–292.[Medline] [Order article via Infotrieve]
22. Abran D, Varma DR, Li D-Y, Chemtob S. Reduced responses of the newborn pig retinal vessels to prostaglandins but not to thromboxane. Can J Physiol Pharmacol. 1994;72:168–173.[Medline] [Order article via Infotrieve]
23. Maragos CM, Wang JM, Hrabie JA, Oppenheim JJ, Keefer LK. Nitric oxide/nucleophile complexes inhibit the in vitro proliferation of A375 melanoma cells via nitric oxide release. Cancer Res. 1993;53:564–568.[Abstract/Free Full Text]
24. Haefliger IO, Flammer J, Lüscher TF. Heterogeneity of endothelium-dependent regulation in ophthalmic and ciliary arteries. Invest Ophthalmol Vis Sci. 1993;34:1722–1730.[Abstract/Free Full Text]
25. Sekhar KR, Hatchett RJ, Shabb JB, Wolfe L, Franci SH, Wells JW, Jastorff B, Butt E, Chakinala MM, Corbin JD. Relaxation of pig coronary arteries by new and potent cGMP analogues that selectively activate the type 1a compared to type 1b cGMP-dependent protein kinase. Mol Pharmacol. 1992;42:103–108.[Abstract]
26. Ullrich V, Hecker M. A concept for the mechanism of prostacyclin and thromboxane A₂ biosynthesis. Adv Prostaglandin Thromboxane Leukot Res. 1990;20:95–101.[Medline] [Order article via Infotrieve]
27. Kawada T, Ishibashi T, Sasage H, Kato K, Imai S. Modification by LY83583 and methylene blue of relaxation induced by nitric oxide, glyceryl trinitrate, sodium nitroprusside and atriopeptin in aorta of the rat, guinea-pig and rabbit. Gen Pharmacol. 1994;25:1361–1371.[Medline] [Order article via Infotrieve]
28. Clementi E, Vecchio I, Sciorati C, Nistico G. Nitric oxide modulation of agonist-evoked intracellular Ca²⁺ release in neurosecretory PC-12 cells: inhibition of phospholipase C activity via cyclic GMP-dependent protein kinase 1. Mol Pharmacol. 1995;47:517–524.[Abstract]
29. Nelson MT, Quayle JM. Physiological roles and properties of potassium channels in arterial smooth muscle. Am J Physiol. 1995;268:C799–C822.[Abstract/Free Full Text]
30. Abran D, Li DY, Varma DR, Chemtob S. Characterization and ontogeny of PGE₂ and PGF₂ receptors on the retinal vasculature of the pig. Prostaglandins. 1995;50:253–267.[Medline] [Order article via Infotrieve]
31. 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.[Medline] [Order article via Infotrieve]
32. Peri KG, Hardy P, Li DY, Varma DR, Chemtob S. Prostaglandin G/H synthase-2 is a major contributor of brain prostaglandins in the newborn. J Biol Chem. 1995;270:24615–24620.[Abstract/Free Full Text]
33. Macmillan S, Sheridan RD, Chilvers ER, Patmore L. A comparison of the effects of SCA40, NS004, and NS1619 on large conductance Ca²⁺-activated K⁺ channels in bovine tracheal smooth muscle cells in culture. Br J Pharmacol. 1995;116:1656–1660.[Medline] [Order article via Infotrieve]
34. Bialecki RA, Stinson-Fisher C. K_Ca channel antagonists reduce NO donor-mediated relaxation of vascular and tracheal smooth muscle. Am J Physiol. 1995;268:L152–L159.[Abstract/Free Full Text]
35. Holzer P, Jocic M, Peskar BA. Mediation by prostaglandins of the nitric oxide-induced neurogenic vasodilatation in rat skin. Br J Pharmacol. 1995;116:2365–2370.[Medline] [Order article via Infotrieve]
36. Lebel M, Grose JH, Lacourciere Y. Effect of short-term administration of cromakalim on renal hemodynamics and eicosanoid excretion in essential hypertension. Am J Hypertens. 1991;4:740–744.[Medline] [Order article via Infotrieve]
37. Coleman RA, Smith WL, Narumiya S. International union of pharmacology classification of prostanoid receptors: properties, distribution, and structure of the receptors and their subtypes. Pharmacol Rev. 1994;46:205–229.[Medline] [Order article via Infotrieve]
38. Hyslop S, De Nucci G. Prostaglandin biosynthesis in the microcirculation: regulation by endothelial and non-endothelial factors. Prostaglandins Leukot Essent Fatty Acids. 1993;49:723–760.[Medline] [Order article via Infotrieve]
39. Davidge ST, Baker PN, Laughlin MK, Roberts JM. Nitric oxide produced by endothelial cells increases production of eicosanoids through activation of prostaglandin H synthase. Circ Res. 1995;77:274–283.[Abstract/Free Full Text]
40. Leffler CW, Busija DW, Armstead WM, Mirro R. H₂O₂ effects on cerebral prostanoids and pial diameter in piglets. Am J Physiol. 1990;258:H1382–H1387.[Abstract/Free Full Text]
41. Chemtob S, Beharry K, Rex J, Chatterjee T, Varma DR, Aranda JV. Ibuprofen enhances retinal and choroidal blood flow autoregulation in newborn piglets. Invest Ophthalmol Vis Sci. 1991;32:1799–1807.[Abstract/Free Full Text]
42. Corbett JA, Kwon G, Turk J, McDaniel ML. IL-1 beta induces the coexpression of both nitric oxide synthase and cyclooxygenase by islets of Langerhans: activation of cyclooxygenase by nitric oxide. Biochemistry. 1993;32:13767–13770.[Medline] [Order article via Infotrieve]
43. Salvemini D, Misko TP, Masferrer JL, Seibert K, Currie MG, Needleman P. Nitric oxide activates cyclooxygenase enzymes. Proc Natl Acad Sci U S A. 1993;90:7240–7244.[Abstract/Free Full Text]
44. Tetsuka T, Daphna-Iken D, Miller BW, Guan Z, Baier LD, Morrison AR. Nitric oxide amplifies interleukin 1-induced cyclooxygenase-2 expression in rat mesangial cells. J Clin Invest. 1996;97:2051–2056.[Medline] [Order article via Infotrieve]
45. Alshihabi SN, Chang YS, Frangos JA, Tarbell JM. Shear stress-induced release of PGE₂ and PGI₂ by vascular smooth muscle cells. Biochem Biophys Res Commun. 1996;224:808–814.[Medline] [Order article via Infotrieve]
46. Urade Y, Kitahama K, Ohishi H, Kaneko T, Mizuno N, Hayaishi O. Dominant expression of mRNA for prostaglandin D synthase in leptomeninges, choroid plexus, and oligodendrocytes of the adult and rat brain. Proc Natl Acad Sci U S A. 1993;90:9070–9074.[Abstract/Free Full Text]
47. Nüsing R, Ullrich V. Immunoquantitation of thromboxane synthase in human tissues. Eicosanoids. 1990;3:175–180.[Medline] [Order article via Infotrieve]
48. Heller R, Bussolino F, Ghigo D, Pescarmona GP, Calvino R, Gasco A, Till U, Bosia A. Activation of endothelial guanylate cyclase inhibits cellular reactivity. Agents Actions. 1995;45:177–181.
49. Marchenko SM, Sage SO. Calcium-activated potassium channels in the endothelium of intact rat aorta. J Physiol (Lond). 1996;492:53–60.[Medline] [Order article via Infotrieve]
50. Girard S, Clapham D. Acceleration of intracellular calcium waves in Xenopus oocytes by calcium influx. Science. 1993;260:229–232.[Abstract/Free Full Text]
51. Bkaily G. cGMP-sensitive K⁺ channels in vascular smooth muscle. In: Ionic Channels in Vascular Smooth Muscle. Boca Raton, Fla: CRC Press; 1994:87–102.
52. Clark SG, Fuchs LC. Role of nitric oxide and Ca²⁺-dependent K⁺ channels in mediating heterogeneous microvascular responses to acetylcholine in different vascular beds. J Pharmacol Exp Ther. 1997;282:1473–1479.[Abstract/Free Full Text]
53. Zhu P, Beny JL, Flammer J, Lüscher TF, Haefliger IO. Relaxation by bradykinin in porcine ciliary artery: role of nitric oxide and K⁺-channels. Invest Ophthalmol Vis Sci. 1997;38:1761–1767.[Abstract/Free Full Text]
54. Yuan XJ, Tod ML, Rubin LJ, Blaustein MP. NO hyperpolarizes pulmonary artery smooth muscle cells and decreases the intracellular Ca²⁺ concentration by activating voltage-gated K⁺ channels. Proc Natl Acad Sci U S A. 1996;93:10489–10494.[Abstract/Free Full Text]
55. Wellman GC, Bonev AD, Nelson MT, Brayden JE. Gender differences in coronary artery diameter involve estrogen, nitric oxide, and Ca²⁺-dependent K⁺ channels. Circ Res. 1996;79:1024–1030.[Abstract/Free Full Text]
56. Archer SL, Huang JM, Hampl V, Nelson DP, Shultz PJ, Weir EK. Nitric oxide and cGMP cause vasorelaxation by activation of a charybdotoxin-sensitive K⁺ channel by cGMP-dependent protein kinase. Proc Natl Acad Sci U S A. 1994;91:7583–7587.[Abstract/Free Full Text]
57. Quayle JM, McCarron JG, Brayden JE, Nelson MT. Inward rectifier K⁺ currents in smooth muscle cells from rat resistance-sized cerebral arteries. Am J Physiol. 1993;265:C1363–C1370.[Abstract/Free Full Text]
58. Armstead WM. Role of activation of calcium-sensitive K⁺ channels in NO- and hypoxia-induced pial artery vasodilation. Am J Physiol. 1997;272:H1785–H1790.