Arachidonic acid- and prostaglandin E₂-induced cerebral vasodilation is mediated by carbon monoxide, independent of reactive oxygen species in piglets

Abstract

Arachidonic acid (AA) and prostaglandin (PG) E₂ stimulate carbon monoxide (CO) production, and AA metabolism is known to be associated with the generation of reactive oxygen species (ROS). This study was conducted to address the hypothesis that CO and/or ROS mediate cerebrovascular dilation in newborn pigs. Experiments were performed on anesthetized newborn pigs with closed cranial windows. Different concentrations of AA (10⁻⁸-10⁻⁶ M), PGE₂ (10⁻⁸-10⁻⁶ M), iloprost (10⁻⁸-10⁻⁶ M), and their vehicle (artificial cerebrospinal fluid) were given. Piglets with PGE₂ and iloprost received indomethacin (5 mg/kg iv) to inhibit cyclooxygenase. AA, PGE₂, and iloprost caused concentration-dependent increases in pial arteriolar diameter. The effects of both AA and PGE₂ in producing cerebral vascular dilation and associated CO production were blocked by the heme oxygenase inhibitor chromium mesoporphyrin (2 × 10⁻⁵ M), but not by the prostacyclin analog, iloprost. ROS inhibitor tempol (SOD mimetic) (1 × 10⁻⁵ M) and the H₂O₂ scavenger catalase (1,000 U/ml) also do not block these vasodilator effects of AA and PGE₂. Heme-l-lysinate-induced cerebrovascular dilation and CO production was blocked by chromium mesoporphyrin. Hypoxanthine plus xanthine oxidase, a combination that is known to generate ROS, caused pial arteriolar dilation and CO production that was inhibited by tempol and catalase. These data suggest that AA- and PGE₂-induced cerebral vascular dilation is mediated by CO, independent of ROS.

in the cerebral microcirculation, products of arachidonic acid (AA) formed through cyclooxygenase (COX) are important contributors to the regulation of cerebral vasodilation (17). Cerebral blood vessels from newborn pigs produce mainly prostaglandin (PG) E₂ and, to a lesser extent, PGI₂, thromboxane A₂, and PGF_2α (6). Prostanoids contribute to a variety of cerebral vascular responses in newborn pigs, including vasodilation in response to hypercapnia and hypotension. PGE₂, the major prostanoid formed in cerebral microvessels, is a potent vasodilator in the cerebral circulation of newborn pigs (18), but little is known about the mechanisms underlying its vasodilator effect in cerebral vessels.

COX, a PG-endoperoxide synthase, is the rate-limiting enzyme in the synthesis of prostanoids from AA (31) into PGH₂, which is subsequently processed by different enzymes into various prostanoids. Two isoforms of COX have been identified: COX-1, a constitutive form; and COX-2, which is induced by various agents, including growth factors, cytokines, and endotoxins. COX-2, however, is also constitutively expressed in some cells as well (23). Both COX-1 and COX-2 are expressed in cerebral microvessels and cerebral endothelia of porcine and human neonates (24, 27, 29), and they contribute to basal prostanoid synthesis in cerebral microvascular endothelial cells of porcine and human neonates (23, 28). COX is associated with the generation of superoxides (3, 14, 30), which can stimulate carbon monoxide (CO) production.

CO is a gasotransmitter that can be related to neural function and blood flow regulation in the brain (2, 19). CO is produced from heme by the action of heme oxygenase (HO), which results in the formation of iron and biliverdin as byproducts (22). In vivo, topical application of CO causes dose-dependent vasodilation of cerebral arterioles in newborn pigs and adult rats (5, 19). However, the vasodilator effects of CO in cerebral vessels are developmentally regulated and time-dependent and may vary between different animal species (5, 13). In the adult rat and newborn pigs, prolonged exposure to CO causes vasoconstriction (8, 13). Moreover, pial arterioles and small cerebral arteries (30–200 μm) may have greater sensitivity to CO than large arteries.

Recently, we proposed that superoxide and/or a subsequent reactive oxygen species (ROS) generated by COX-catalyzed AA metabolism increase cerebrovascular CO production in cerebral microvessels in vitro (9). In that study, however, we did not examine the effects of inhibitors of ROS on AA- or PGE₂-induced CO production. Moreover, the role of ROS in AA- and PGE₂-induced cerebrovascular dilation is not known. The goal of the present study was to test the hypothesis that CO mediates cerebral vasodilation produced by AA and PGE₂ in the newborn piglet brain, independent of ROS.

MATERIALS AND METHODS

All procedures that involved animals were reviewed and approved by the Animal Care and Use Committee of the University of Tennessee Health Science Center. Newborn pigs (1–3 days old) were anesthetized with ketamine hydrochloride (5 mg/kg im) and xylazine (2.0 mg/kg im) and maintained on α-chloralose (50 mg/kg iv). A catheter was inserted into a femoral artery to monitor blood pressure and heart rate and to collect blood for measurements of Pco₂, Po₂, and pH. A second catheter was placed in a femoral vein for anesthetic and fluid administration. The trachea was cannulated, and the animals were mechanically ventilated and supplemented with O₂, if needed, to maintain arterial pH, Pco₂, and Po₂ within the normal range. A heating pad was used to maintain the animals at 37.5–38.5°C, and temperature was monitored with a rectal probe.

Cranial window placement and pial arteriolar monitoring.

The scalp was surgically removed, and a 2-cm diameter craniotomy was made over the parietal cortex. The dura was incised and reflected over the bone to prevent contact between the brain source and the cut edge of the bone. A stainless steel and glass window was implanted into the hole and cemented sequentially with bone wax and dental acrylic. This window consisted of three parts: a stainless steel ring, a circular glass coverslip, and three ports consisting of 17-gauge hypodermic needles attached to three precut holes in the stainless steel ring. The space under the window was filled with artificial cerebrospinal fluid (aCSF; 150 meq/l Na⁺, 3 meq/l K⁺, 2.5 meq/l Ca²⁺, 1.2 meq/l Mg²⁺, 132 meq/l Cl⁻, 3.7 mM glucose, 6 mM urea, 25 meq/l HCO₃⁻; pH ≈ 7.33, Pco₂ 46 mmHg, Po₂ 43 mmHg, 37°C) through needles incorporated into the sides of the window. aCSF was slowly infused into one side of the window, and samples for measurement of CO were collected from the needle ports on the sides of the window. The volume of the fluid directly beneath the window was 350 μl and was contiguous with the periarachnoid space. Pial vessels were observed through the window with a stereomicroscope. Pial arteriolar diameter was measured with a video micrometer coupled to a television camera mounted on the microscope and a video monitor. Using a stage linear micrometer ranging from 0 to 1,000 μm, aCSF sample for CO analysis was collected at the end of each 10-min exposure period by slowly infusing aCSF under the window. The aCSF dripped freely into a brown amber collecting vial, the internal standard (³¹CO) was injected into the bottom of the vial, and the vial was immediately sealed with a rubberized Teflon-lined cap. The samples were then heated in a hot water bath (70°C), and CO head space gas was determined using gas chromatography/mass spectroscopy (5975C series; Agilent Technologies, Santa Clara, CA), as previously described (11, 16).

Chemicals.

The HO substrate heme-l-lysinate (HLL) was prepared using methods described by Tenhunen et al. (34) and was stored at −30°C. HLL stock solution (10⁻² M) in H₂O adjusted to pH 10.0 with 0.1 NaOH was diluted with aCSF to 2 × 10⁻² M. The HO inhibitor chromium mesoporphyrin (CrMP) was purchased from Frontier Scientific (Logan, UT); iloprost, prostacyclin analog was purchased from Tocris Bioscience (Ellisville, MO); and AA and PGE₂ were purchased from Cayman Chemical (Ann Arbor, MI). Both HO and CrMP are photosensitive and were protected from light. Water-soluble indomethacin (indomethacin trihydrate) was a gift from Merck Sharp & Dohme Research Laboratories (Rahway, NJ). Tempol was purchased from Santa Cruz Biotechnology (Santa Cruz, CA). Xanthine oxidase, hypoxanthine, catalase, and all other chemicals were of analytic grade and purchased from Sigma Chemical (St. Louis, MO).

Experimental design.

After implantation of the cranial window, ∼30 min were allowed before experimentation was begun. The baseline pial arteriolar diameter was recorded, along with heart rate, mean arterial blood pressure, and core temperature. The experimental design consisted of measurements of arteriolar diameter, mean arterial pressure, arterial blood gases, and pH. Before and after each challenge, aCSF for CO measurement was collected from beneath the cranial window.

The newborn pigs were randomly divided into six groups. The first two groups of animals were used to determine the effect of different concentrations of AA (10⁻⁸-10⁻⁶ M), PGE₂ (10⁻⁸-10⁻⁶ M), iloprost (10⁻⁸-10⁻⁶ M), and their vehicle (aCSF) in the presence of the HO inhibitor CrMP (2 × 10⁻⁵ M). In experiments with PGE₂ and iloprost, the animals were administered indomethacin (5 mg/kg iv) as a single intravenous bolus 15 min before topical application of the stimuli (the challenges) to block endogenous production of COX products, whereas the AA set of animals did not receive indomethacin. We have shown that this dose of indomethacin inhibits endogenous cerebral PG production and prevents conversion of exogenous AA to PGs by the cerebral vessels (1, 11, 15, 20). AA and PGE₂ and iloprost were each dissolved in ethanol and diluted in aCSF solution. PGE₂ was used in these experiments because it is the prostanoid produced in the greatest amount from exogenous AA by the cerebral microvessels. In the third group of animals, the catalytic activity of HO in intact cerebral arterioles was determined by examining the effect of exogenous HLL (2 × 10⁻⁶ M) before and after the administration of CrMP (2 × 10⁻⁵ M). This test was done to confirm the efficacy of CrMP in inhibiting HO. In another group of piglets, we further examined the effects of ROS and superoxide anion on pial arteriolar diameters by administering hypoxanthine and xanthine oxidase, a superoxide-generating system. We topically applied hypoxanthine (1 × 10⁻⁵ M) and xanthine oxidase (1 × 10⁻⁵ M) before and after administration of the inhibitors tempol (SOD mimetic) (1 × 10⁻⁵ M) or catalase (H₂O₂ scavenger) (1,000 U/ml). Measurements of diameters of pial arterioles were taken at 1, 3, 5, and 10 min after administration, and an aCSF sample for CO analysis was collected at the end of each 10-min exposure period by slowly infusing aCSF under the window; the aCSF dripped freely into a collecting aliquot tube. At the end of each tested response, an arterial blood-gas sample was taken, and the area under the window gently flushed with fresh aCSF for 30 s to remove the previous stimulus through needles incorporated into the side of the window. Heart rate, mean arterial pressure, temperature, and pH were continuously monitored.

The role of the COX product PGE₂, which may increase CO production, was also examined. COX was inhibited with indomethacin, as stated above. Three concentrations of PGE₂ and iloprost were used that were equimolar to the exogenous AA. Each concentration of PGE₂, iloprost, AA, or their vehicle was randomly and topically applied for 10 min. Changes in diameters of pial arterioles were measured at 1, 3, 5, and 10 min after application, and an aCSF sample for CO analysis was collected at the end of each 10-min exposure period. We decided to use PGE₂ at maximal vasodilator concentration (determined to be 10⁻⁵ M) to obtain more than 50% vasodilation so as to determine the effect of various inhibitors more accurately on the changes in PGE₂ and arachidonic-induced vasodilation. PGE₂ in lower concentrations (10⁻¹⁰-10⁻⁸ M) produce moderate vasodilation with a maximal increase of only 5–20% above the basal diameter (26).

To further investigate the effects of ROS and superoxide anion, the superoxide-generating system hypoxanthine (1 × 10⁻⁵ M) and xanthine oxidase (1 × 10⁻⁵ M) was administered topically before and after administration of tempol (1 × 10⁻⁵ M), a SOD mimetic, or the H₂O₂ scavenger catalase (1,000 U/ml). Changes in diameters of pial arterioles were measured at 1, 3, 5, and 10 min. At the end of each tested response, an arterial blood-gas sample was collected, as previously described for CO measurements. Heart rate, mean arterial pressure, temperature, and pH were continuously monitored.

The effects of topical application of the ROS inhibitor tempol or the H₂O₂ scavenger catalase on AA (1 × 10⁻⁵ M), PGE₂ (1 × 10⁻⁵ M), and their vehicle in piglet pial arterioles was investigated in the last group of animals.

Data analysis.

Data are presented as means ± SE. Results were analyzed by ANOVA with repeated measures, followed by Tukey's post hoc test to isolate differences between groups. A level of P < 0.05 was considered significant

RESULTS

There were no significant changes in arterial blood-gas levels, pH, blood pressure, or body temperature when the values were compared at the beginning and end of the experiments.

Effects of topical application of AA and PGE₂ on pial arteriolar diameters.

Figure 1, A and B, show changes in diameter of cerebral arterioles in response to AA (10⁻⁶-10⁻⁵ M), PGE₂ (10⁻⁸-10⁻⁶ M), and their vehicle after administration of CrMP (2 × 10⁻⁵ M). AA and PGE₂ in the presence of indomethacin dilated pial arterioles in a concentration-dependent manner. The effect of AA and PGE₂ to increase pial arteriolar diameter was blocked by CrMP (2 × 10⁻⁶ M).

Fig. 1.

A: changes in diameter of cerebral arterioles in response to arachidonic acid (AA; 10⁻⁶-10⁻⁵ M), before and in the presence of chromium mesoporphyrin (CrMP; 2 × 10⁻⁵ M). The cerebral vasodilator effects of AA were blocked by CrMP. The cranial window was implanted, and each concentration of AA or their vehicle (Veh) was randomly and topically applied for 10 min. The maximal change in the diameter of pial arterioles (45–150 μm) was observed with a dissecting microscope and measured with a video micrometer, coupled to a television camera, mounted on the microscope and a video monitor. After application of each concentration, the arteriolar diameter was allowed to return to basal values. Values are means ± SE (n = 6). The increase in pial arteriolar diameter produced by each concentration of AA was significant compared with the corresponding value obtained with Veh. *P < 0.05, AA treatment vs. its Veh. †P < 0.05, CrMP vs. no CrMP. B: effects of topical application of prostaglandin (PG) E₂ in the presence of indomethacin (5 mg/kg iv) on diameters of pial arterioles of piglets without and with CrMP (2 × 10⁻⁶ M). PGE₂, in the presence of indomethacin, dilated pial arterioles in a concentration-dependent manner. The cerebral vasodilator effects of PGE₂ were blocked by CrMP (2 × 10⁻⁶ M). After application of each concentration, the arteriolar diameter was allowed to return to basal values. Values are means ± SE (n = 6). The increase in pial arteriolar diameter produced by each concentration of PGE₂ was significant compared with the corresponding value obtained with Veh. *P < 0.05, PGE₂ treatment vs. its Veh. †P < 0.05, CrMP vs. no CrMP. C: effect of topical application of the prostacyclin receptor agonist, iloprost, in the absence and presence of CrMP (2 × 10⁻⁵ M) on piglet arteriolar diameters. Topical application of iloprost on piglet cerebral arterioles caused dilation that was not blocked by the metal porphyrin inhibitor of heme oxygenase (HO), CrMP. Values are means ± SE; n = 6. *P < 0.05, iloprost vs. its Veh. †P < 0.05, CrMP vs. no CrMP.

Effect of topical application of the prostacyclin receptor agonist iloprost (10⁻⁸-10⁻⁶ M) on pial arteriolar diameters.

Figure 1C shows topical application of iloprost on piglet cerebral arterioles, causing dilation that was not blocked by the inhibitor of HO, CrMP.

Effects of topical application of HLL on pial arteriolar diameters.

Figure 2 shows the change in piglet pial arteriolar diameter caused by the topical application of HLL (2 × 10⁻⁶ M) before and after administration of CrMP (2 × 10⁻⁵ M). Topical application of the HO substrate HLL on piglet cerebral arterioles caused dilation that was blocked by the metal porphyrin inhibitor of HO, CrMP.

Fig. 2.

Effect of heme-l-lysinate (HLL) (2 × 10⁻⁶ M) on piglet pial arteriolar diameters in the absence and presence of CrMP (2 × 10⁻⁵ M). Topical application of the HO substrate HLL on piglet cerebral arterioles caused dilation that was blocked by the metal porphyrin inhibitor of HO, CrMP. Values are means ± SE; n = 6. *P < 0.05, HLL vs. its Veh. †P < 0.05, CrMP vs. no CrMP (control).

Effects of topical application of the superoxide-generating system hypoxanthine and xanthine oxidase on pial arteriolar diameters.

Figure 3 shows the change in pial arteriolar diameter caused by the topical application of the superoxide-generating system hypoxanthine acting on xanthine oxidase. Because this reaction produces both superoxide and hydrogen peroxide, the superoxide-generating system increases piglet pial arteriolar diameter. This increase in pial arteriolar diameter was blocked by the ROS inhibitor tempol (SOD mimetic) or by the H₂O₂ scavenger catalase (1,000 U/ml). Catalase was used to eliminate the effect of hydrogen peroxide or via the dismutation of superoxide.

Fig. 3.

The superoxide-generating system hypoxanthine (HX; 1 × 10⁻⁵ M) and xanthine oxidase (1 × 10⁻⁵ M) increase piglet pial arteriolar diameter. This increase in pial diameter was blocked by the reactive oxygen species (ROS) inhibitor tempol (SOD mimetic; 1 × 10⁻⁵ M) and also by the H₂O₂ scavenger catalase (1,000 U/ml). Changes in diameter were measured as described in Fig. 1 legend. Each point is given as insert means ± SE; n = 6. *P < 0.05, HX with xanthine oxidase vs. its Veh. ‡P < 0.05, tempol vs. no tempol. †P < 0.05, catalase vs. no catalase.

Effects of topical application of the ROS inhibitor tempol (SOD mimetic), or the H₂O₂ scavenger catalase on AA, PGE₂, and their vehicle in piglet pial arterioles.

Figure 4 illustrates the effects of topical application of tempol and catalase for 10 min on AA (1 × 10⁻⁵ M), PGE₂ (1 × 10⁻⁵ M), and their vehicle in piglet pial arterioles. Topical application of tempol (1 × 10⁻⁵ M) and catalase (1,000 U/ml) on AA (1 × 10⁻⁵ M), PGE₂ (1 × 10⁻⁵ M), and their vehicle in piglet pial arterioles caused an increase in pial arteriolar diameter. Neither the ROS inhibitor tempol nor the H₂O₂ scavenger catalase blocked this increase in pial arteriolar diameter produced by either AA or PGE₂.

Fig. 4.

A: effects of topical application of tempol (1 × 10⁻⁵ M) and catalase (1,000 U/ml) for 10 min on the effects of AA (1 × 10⁻⁵ M) and their Veh in piglet pial arterioles. Neither the ROS inhibitor tempol (SOD mimetic) nor the H₂O₂ scavenger catalase inhibited the increase in pial arteriolar diameter produced by either AA or PGE₂. Values are means ± SE; n = 6. *P < 0.05, AA vs. its Veh. ‡P < 0.05, tempol vs. no tempol. †P < 0.05, catalase vs. no catalase. B: effects of topical application of tempol (1 × 10⁻⁵ M) and catalase (1,000 U/ml) for 10 min on the effects of PGE₂ (1 × 10⁻⁵ M) and their Veh in piglet pial arterioles. Neither the ROS inhibitor tempol (SOD mimetic) nor the H₂O₂ scavenger catalase inhibited the increase in pial arteriolar diameter produced by PGE₂. Values are means ± SE; n = 6. *P < 0.05, PGE₂ vs. its Veh. ‡P < 0.05, tempol vs. no tempol. †P < 0.05, catalase vs. no catalase.

Effect of HLL (2 × 10⁻⁶ M) on CO production before and after administration of CrMP.

As expected, the substrate HLL increased CO production, and this effect was blocked by CrMP (Fig. 5).

Fig. 5.

Effect of HLL (2 × 10⁻⁶ M) on carbon monoxide (CO) production in the absence and presence of CrMP (2 × 10⁻⁵ M). Values are means ± SE; n = 6. *P < 0.05, HLL vs. no HLL. †P < 0.05, CrMP vs. no CrMP.

Effects of CrMP, the ROS inhibitor tempol, and the H₂O₂ scavenger catalase on PGE₂ (1 × 10⁻⁵ M)-induced CO production.

The HO inhibitor CrMP blocked the production of CO, whereas tempol and catalase did not inhibit the increase in PGE₂-induced CO production (Fig. 6).

Fig. 6.

Effect of PGE₂-induced CO production in the absence and presence of tempol and catalase. Tempol and catalase do not inhibit the increase in PGE₂-induced CO production. Values are means ± SE; n = 6. *P < 0.05, PGE₂ vs. its Veh. †P < 0.05, CrMP vs. no CrMP.

Effect of the superoxide-generating system hypoxanthine (1 × 10⁻⁵ M) and xanthine oxidase (1 × 10⁻⁵ M) on CO production in the absence and presence of tempol, catalase, and their vehicle.

Figure 7 shows that hypoxanthine and xanthine oxidase increased CO production. This increase in CO production was blocked by the ROS inhibitor tempol (SOD mimetic) and also by the H₂O₂ scavenger catalase (1,000 U/ml).

Fig. 7.

Effect of the superoxide-generating system of HX (1 × 10⁻⁵ M) and xanthine oxidase (1 × 10⁻⁵ M) on CO production in piglet pial arterioles in the absence and presence of tempol, catalase, and their Veh. The ROS increased CO production. This increase was blocked by the ROS inhibitor tempol (SOD mimetic) (1 × 10⁻⁵ M) and also by the H₂O₂ scavenger catalase. Values are means ± SE; n = 6. *P < 0.05, HX with xanthine oxidase vs. its Veh. ‡P < 0.05, tempol vs. no tempol. †P < 0.05, catalase vs. no catalase.

DISCUSSION

This study in piglets presents the following novel findings. 1) Topical application of AA and PGE₂ on pial arterioles produced vasodilation and CO production that was blocked by the HO inhibitor CrMP. 2) AA- and PGE₂-induced cerebral vasodilation and CO production were not inhibited by the ROS inhibitor tempol (SOD mimetic) or by the H₂O₂ scavenger catalase. 3) The standard active superoxide-generating system of hypoxanthine and xanthine oxidase increased pial arteriolar diameter and CO production, which were inhibited by tempol and catalase.

These observations and the demonstrations that 1) CO produces cerebral vasodilation, and 2) the HO substrate HLL also caused cerebral vasodilation that was blocked by CrMP raised the possibility that CO produced in cerebral microvessels might mediate AA- and PGE₂-induced cerebral vasodilation in piglets. Supporting this view were our observations that AA- and PGE₂-induced cerebral vasodilation was blocked apparently by inhibiting PGE₂ (EP) receptors (25). Interestingly, iloprost also produced cerebral vasodilation that was not altered by CrMP (Fig. 1C), suggesting that differences in the mechanism of action produce cerebral vasodilation. Further studies will be required to elucidate the differences in the cellular and molecular mechanisms involved in cerebral vasodilation produced by prostacyclin and its analog than PGE₂.

Cerebral microvessels contain both COX-1 and COX-2 (24, 29), but the relative contributions of COX-1 and COX-2 to the metabolism of PGE₂ that stimulates CO production and causes cerebral vasodilation are not known. The metabolism of AA by COX is associated with generation of ROS (4, 14, 30), which could stimulate CO production and cause vasodilation. Previously, we proposed that superoxide and/or a subsequent ROS generated by COX-catalyzed AA metabolism might increase cerebrovascular CO production in cerebral microvessels in vitro (9). In that study, however, we did not examine the effect of inhibitors of ROS on AA- or PGE₂-induced CO production. Therefore, it is possible that AA- and/or PGE₂-induced cerebral vasodilation is due to ROS. However, that appears to be unlikely for the following reasons: 1) the effects of AA and/or PGE₂ on CO production and pial arteriolar diameter were not altered by either tempol or catalase (Fig. 4, A and B); and 2) tempol and catalase inhibited dilation of pial arterioles produced by the ROS-generating system hypoxanthine and xanthine oxidase and CO production (Figs. 6 and 7). These observations, together with our demonstrations that AA increased CO production and that the COX inhibitor indomethacin blocked AA- but not HLL-induced CO production (9), suggested that endogenous PGs generated in the cerebral microvessels produce cerebral vasodilation via CO production. The principle mediators of AA-induced vasodilation, as supported by our previous data from cerebral microvessels studies, appear to be mainly PGE₂ and, to a lesser extent, PGI₂ (6).

PG formation occurs in the majority of mammalian organs and tissues, but not in all cell types of each organ, and the amounts produced by different cell types may vary (32). In the brain, PG hydrogen synthase protein immunoreactivity was shown to be associated with neurons and, to a lesser extent, glial cells in most areas (35). Such immunostaining can be accounted for by PG hydrogen synthase-2 gene expression that, unlike the situation in peripheral tissues, appears to be the constitutive isoform in neurons of the rat brain (36). In contrast, findings on cultured neurons and astrocytes showed that the latter produce and release far larger amounts of PGE₂ (12).

Recent studies have also shown that CO generated in astrocytes mediates glutamate-induced cerebral vasodilation (21). Moreover, we have also shown that adenosine diphosphate-induced cerebral vasodilation is mediated by astrocyte-dependent CO production (10). These observations, together with the demonstration that astrocytes can metabolize AA to PGE₂ (7, 33) and contain prostanoid receptors (37), raise the possibility that astrocytes may also contribute to AA- and PGE₂-mediated cerebral vasodilation that increases CO.

The present study provides the first evidence that endogenous PGs generated in the cerebral circulation produce cerebral vasodilation via generation of CO. These studies also demonstrate that PGE₂, the most abundant AA metabolite of COX in cerebral blood vessels of newborn pigs, produces its effect via generation of CO. However, we cannot rule out the contribution of other vasodilator prostanoids (e.g., PGI₂) to the cerebral vasodilator effect of AA. Moreover, our data suggest that the cerebral vasodilation produced by AA and PGE₂ is independent of ROS.

GRANTS

This project was supported by National Heart, Lung, and Blood Institute Grant 5K01HL0964.

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the author(s).

AUTHOR CONTRIBUTIONS

Author contributions: A.K. conception and design of research; A.K. performed experiments; A.K. analyzed data; A.K. interpreted results of experiments; A.K. prepared figures; A.K. drafted manuscript; A.K. and C.W.L. edited and revised manuscript; A.K. and C.W.L. approved final version of manuscript.