INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

IN THE CEREBRAL CIRCULATION of newborn pigs, dilator prostanoids, but not nitric oxide (NO), are predominant vasorelaxant factors that contribute to the vasodilation of pial arterioles in response to hypercapnia, acetylcholine, and histamine (1, 15, 22). In older pigs, NO-mediated responses also contribute to the regulation of cerebral vascular tone (1, 31, 32). The implication of these physiological findings is that the production of endothelium-derived vasorelaxant factors, prostanoids, and NO is developmentally regulated during the postnatal period. NO synthase and cyclooxygenase are the key enzymes in synthesis of the vasorelaxant factors. Endothelial Ca²⁺-dependent NO synthase could be regulated not only at transcriptional and translational levels but also posttranslationally by protein phosphorylation (2, 17, 18).

Cyclooxygenase is represented by two isoforms encoded by different genes, COX-1 and COX-2. COX-1 is a constitutive "housekeeping" enzyme in a variety of prostanoid-producing cells, whereas COX-2 as an early response gene product is largely induced by mitogens (8, 28). However, in the cerebral circulation of newborn pigs, COX-2 protein is constitutively expressed in microvessels (20, 23), in vascular endothelium in the choroid plexus (29), and in cultured microvascular cells (19, 20). In addition, COX-2 provides a major functional contribution to dilator prostanoid synthesis (19, 23). In cerebral vascular cells from newborn pigs, constitutively expressed COX-2 can be posttranslationally activated by tyrosine phosphorylation (19).

Multiple pathways for the regulation of NO synthase and cyclooxygenase at transcriptional, translational, and posttranslational levels can account for developmental changes in the expression and activity of both enzymes. In the present paper, we tested the hypothesis that in the cerebral circulation of pigs endothelial NO synthase (eNOS) activity is increased, whereas cyclooxygenase activity is inhibited, during the postnatal development. To test this hypothesis, we compared the expression and activities of eNOS and cyclooxygenase in isolated microvessels and in cultured endothelial cells from the cerebral cortex of newborn and adult pigs. The results of the present study demonstrated that our initial hypothesis should be modified. The expression of eNOS is upregulated in adult pigs and accounts for the increased production of endothelial NO in the cerebral microcirculation. However, the expression of endothelial cyclooxygenase, the functional role of COX-1 and COX-2 isoforms, and the production of endothelium-derived dilator prostanoids in cerebral circulation in pigs appear to remain unaltered on maturation.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Protocols involving animals were approved by the Animal Care and Use Committee at the University of Tennessee, Memphis. All procedures were done using aseptic techniques.

Materials. Cell culture reagents were obtained from Life Technologies (Gaithersburg, MD) and Sigma (St. Louis, MO). Matrigel (growth factor reduced) was purchased from Becton Dickinson (Bedford, MA). N^G-monomethyl-L-arginine (L-NMMA) was from Sigma. L-[U-¹⁴C]arginine was from DuPont (Wilmington, DE); NS-398 and arachidonic acid (peroxide-free) were from Cayman (Ann Arbor, MI). Iloprost was a gift from Schering, Germany.

Isolation of cerebral microvessels. Newborn piglets (3-5 days) were anesthetized with ketamine-acepromazine (33 and 3.3 mg/kg), the brain was removed, and the cortex was dissected. Brains from adult pigs were freshly collected at a local abattoir. Brain cortex was gently homogenized using a Dounce homogenizer with a loose-fitting pestle in culture medium 199 containing 100 U/ml penicillin, 100 µg/ml streptomycin, and 2.5 µg/ml amphotericin B. Cerebral microvessels (60-300 µm) were isolated by consequent filtration of the brain homogenate through 300- and 60-µm nylon mesh screens.

Cultured cells. Primary cultures of cerebral microvascular cells were established as previously described (19). For smooth muscle cell explants, cerebral microvessels were plated directly onto 12-well Costar plates coated with Matrigel. Endothelial cells were isolated by collagenase-dispase treatment (1 mg/ml for 2 h at 37°C) followed by Percoll density gradient centrifugation and plated onto Matrigel-coated Costar plates. Cells were cultured in Dulbecco's modified Eagle's medium with 20% fetal bovine serum, 30 µg/ml endothelial cell growth supplement, 1 U/ml heparin, and antibiotic-antimycotic mixture in a 5% CO₂ air incubator at 37°C until confluence (5-7 days for endothelial cells and 13-15 days for smooth muscle cells).

All experiments were performed on confluent quiescent cells. To achieve confluence, cells were exposed to a serum-depleted medium (0% fetal bovine serum, 0% endothelial cell growth supplement) for 15-20 h before the experiment.

Prostanoid production. Cells were rinsed with Dulbecco-modified phosphate-buffered saline (PBS) and incubated in 1 ml of artificial cerebrospinal fluid (aCSF), an incubation medium similar to cortical cerebrospinal fluid (in mM, 3.0 KCl, 1.5 MgCl₂, 1.5 CaCl₂, 132 NaCl, 6.6 urea, 3.7 dextrose, and 24.6 NaHCO₃ equilibrated with 5% CO₂ and air to pH 7.4-7.5; PCO₂, 32-36 mmHg; PO₂, 100-120 mmHg). When indicated, ionomycin (10⁶ M), acetylcholine (10⁵ M), or bradykinin (10⁵ M) was added to the incubation medium. To determine the contribution of COX-2 to prostanoid synthesis, cells were pretreated for 20 min before the experiment with NS-398 (10⁵ M). As we have demonstrated before, 10⁵ to 10⁴ M NS-398 effectively inhibited cyclooxygenase activity in COX-2 expressing cells (endothelial and smooth muscle cells from cerebral microvessels of newborn pigs) (19, 20) but not in COX-1 expressing cells [subcultured human umbilical vein endothelial cells (HUVECs) and Swiss 3T3 fibroblasts] (19). After 30 min of incubation at 37°C, the medium was aspirated and stored at 20°C for prostanoid determination. For protein determination, cells were extracted with 0.1 N HCl; protein was detected using the Pierce micro-BCA assay (Pierce Chemical, Rockford, IL).

Cyclooxygenase activity. Cyclooxygenase activity was detected as prostanoid production from arachidonic acid (19). Freshly isolated cerebral microvessels or cultured endothelial cells were washed twice with PBS and incubated with 10-30 µM arachidonic acid in 1 ml of aCSF for 10-20 min at 37°C. The incubation medium was aspirated and stored at 20°C for prostanoid determination. To determine the functional contribution of COX-2, microvessels and cells were pretreated for 20 min with the COX-2 selective inhibitor NS-398 (10⁵ M).

Prostanoid assays. Concentrations of 6-keto-PGF₁ (the stable hydrolysis product of prostacyclin) and PGE₂ in the cell incubation medium were determined by radioimmunoassay.

Nitrite-nitrate assays. Endothelial cells were incubated in Krebs-Ringer buffer for 1 h at 37°C without and with 10⁵ M bradykinin or acetylcholine. To control the specificity of NO synthase-derived nitrite-nitrate detection, we added the NO synthase inhibitor L-NMMA (1 mM) or EDTA (5 mM) to the incubation medium; heat-inactivated cell homogenate (100°C, 5 min) was used as an additional control. The amount of the final products of NO metabolism in the incubation medium was detected by a nitrate-nitrite assay kit (Cayman Chemical). Cell medium (80 µl) pretreated with nitrate reductase to convert nitrate to nitrite was mixed with Griess reagents (0.5% sulfanilamide and 0.05% naphthylethylenediamine dihydrochloride in 5% phosphoric acid); the absorbance at 540 nm was measured.

NO synthase activity. NO synthase activity was detected in intact endothelial cells and in cell homogenates by the conversion of L-[¹⁴C]arginine to L-[¹⁴C]citrulline (25). To determine total NO synthase activity, cells were incubated with L-[¹⁴C]arginine (0.5 µCi/ml) for 30 min at 37°C. When indicated, 10⁵ M bradykinin or 10⁵ M acetylcholine was added to the incubation medium. L-NMMA (1 mM) and 5 mM EDTA were used for the specificity of Ca²⁺-dependent NO synthase detection. At the end of the incubation, the cells were disrupted by sonication in a cold 50 mM HEPES buffer (pH 7.5) containing 5 mM EDTA and clarified by centrifugation (5,000 g, 10 min). Excess [¹⁴C]arginine was removed by binding to Dowex-50W-X8 cation exchange resin (Na⁺ form) in 50 mM HEPES (pH 7.5). The amount of [¹⁴C]citrulline in supernatant was detected by liquid scintillation.

To determine NO synthase activity in cell homogenates, endothelial cells were disrupted by sonication in cold 50 mM HEPES buffer (pH 7.5) containing protease inhibitors (1 mM EDTA, 1 mM dithiothreitol, 80 mg/l leupeptin, 40 mg/l aprotinin, 40 mg/l phenylmethylsulfonyl fluoride, and 35 mg/l sodium orthovanadate). Cell homogenates were incubated for 15-60 min at 37°C in a reaction mixture (100 µl total volume, 30-50 µg protein) containing 50 mM HEPES (pH 7.5), 2 mM CaCl₂, 1 mM NADPH, 10 µM FAD, 100 µM tetrahydrobiopterin, 10 µM L-arginine, and L-[¹⁴C]arginine (10⁵ count/min per tube) (25, 27). Unreacted [¹⁴C]arginine was removed by binding to 0.5 ml of Dowex-50W-X8 cation exchange resin (Na⁺ form) in 50 mM HEPES (pH 7.5; ratio 1:1, wt/vol). The Dowex pellet was washed twice by centrifugation using 0.5 ml of cold 50 mM HEPES, pH 7.5. Supernatants were combined, and the amount of [¹⁴C]citrulline was detected by liquid scintillation counting. To control the specificity of NO synthase activity detection, we added L-NMMA (1 mM) or EDTA (5 mM) to the incubation medium; heat-inactivated cell homogenate (100°C, 5 min) was used as an additional control. To calculate NO synthase activity, we subtracted nonspecific background (detected as [¹⁴C]citrulline formation resistant to 1 mM L-NMMA) from the total amount of [¹⁴C]citrulline.

cAMP and cGMP formation by vascular smooth muscle cells. Smooth muscle cells were washed twice with PBS and incubated in aCSF with the stable prostacyclin analog iloprost (10¹⁰-10⁶ M) or the NO donor sodium nitroprusside (10¹⁰-10⁶ M) at 37°C for 15 min. Cyclic nucleotides were extracted from the cells by sonication in 0.1 N HCl (21). Cell homogenates were clarified by centrifugation, neutralized by 1 N NaOH, and stored at 60°C. Aliquots were used for protein determination by the Pierce micro-BCA method (Pierce Chemical). cAMP and cGMP in cell extracts were determined by radioimmunoassay (21). Acetylation of samples was performed immediately prior to assay to increase the sensitivity of the method as described (21). Cyclic nucleotide content was normalized to cell protein.

Western blotting. Cells were solubilized with Laemmli sample buffer (10 min, 100°C), and extracts were clarified by centrifugation. Samples (20-50 µg protein/lane) were separated by 7.5% polyacrylamide SDS gel electrophoresis as described (20). Biotinylated molecular weight protein standards (Bio-Rad) and prestained protein standards (Sigma) were used. For the specific standards, we used COX-2 from sheep placenta (Cayman Chemical), COX-1 from ram seminal vesicles (Oxford Biomed Research, Oxford, MI), and eNOS (Transduction Laboratories; Lexington, KY). The separated proteins were electrotransferred to polyvinylidene difluoride-membranes (Micron Separations). Membranes were probed with 1) COX-2 (human) polyclonal antiserum (Cayman, at 1:10,000 dilution) (20); 2) COX-1 peptide polyclonal antiserum (PG-16 from Oxford Biomed, at 1:1,000 dilution) (20); and 3) eNOS monoclonal antibodies (Transduction Laboratories, at 1:2,000 dilution). As secondary antibody, we used peroxidase-conjugated donkey anti-rabbit or anti-mouse IgG (dilution 1:10,000; Jackson Immunoresearch Laboratories, West Grove, PA). Streptavidin-biotin-horseradish peroxidase complex (Amersham, at 1:1,000 dilution) was added to the second antibody incubation medium to develop biotinylated standards. Blots were developed using a chemiluminescence detection system (DuPont).

Statistical analysis. Data are presented as means ± SE of absolute values or percentage of control. Data were analyzed by ANOVA for repeated measurements, followed by Fisher's test for protected least-significant difference to isolate differences among groups. P < 0.05 was considered significant in all statistical tests.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

NO synthase and cyclooxygenase expression in cerebral microvessels and in cultured cerebral microvascular endothelial cells. eNOS is expressed in freshly isolated cerebral microvessels and in cultured cerebral microvascular endothelial cells from newborn and adult pigs. The expression of eNOS in microvessels and cultured cells from adult pigs is two- to threefold higher than in newborn pigs (Fig. 1). As we have found previously, COX-1 and COX-2 isoforms are constitutively expressed in cerebral microcirculation of newborn pigs (20). We compared COX-1 and COX-2 protein expression in freshly isolated cerebral microvessels and in cultured microvascular endothelial cells from newborn and adult animals. No differences were found in COX-2 (Fig. 1) and COX-1 (data not shown) expression between the age groups.

View larger version (33K): [in this window] [in a new window]   Fig. 1.   Immunodetection of endothelial nitric oxide synthase (eNOS) and COX-2 in cerebral microvessels and microvascular endothelial cells from newborn and adult pigs. Freshly isolated cerebral microvessels and confluent quiescent cells were lysed in Laemmli sample buffer, electrophoresed in 7.5% PAGE-SDS, transferred to polyvinylidene difluoride membranes, probed with monoclonal anti-eNOS (Transduction Laboratories, 1:1,000) and polyclonal anti-COX-2 (Cayman, 1:10,000), and reprobed for actin (monoclonal anti-actin, Boehringer, 1:50,000). A: representative blots. B: quantification of eNOS and COX-2 expression in adult pigs compared with newborn animals; each point represents means ± SE of 4 determinations.

NO synthase activity in cerebral microvascular endothelial cells. Cerebral microvascular endothelial cells from newborn and adult pigs demonstrate Ca²⁺-dependent NO synthase activity as measured by [¹⁴C]citrulline formation from [¹⁴C]arginine by intact cells and cell homogenates. L-NMMA (1 mM) inhibited 60-70% of the total NO synthase activity in intact cells and in cell homogenates from both newborn and adult pigs (data not shown). In intact endothelial cells from adult pigs, basal and stimulated NO synthase activity was two- to threefold higher than in cells from newborns (Fig. 2). Endothelial cells from both newborn and adult pigs responded to acetylcholine (10⁵ M) and bradykinin (10⁵ M) by increasing NO synthase activity 1.3- to 1.5-fold above the basal level (Fig. 2). Increased NO synthase activity in adult pigs was confirmed using endothelial cell homogenates: total NO synthase activity was 20 ± 5 × 10³ and 120 ± 30 × 10³ counts/min [¹⁴C]citrulline/mg protein in newborns and adults, respectively (P < 0.005); L-NMMA-inhibited activity was 8 ± 2 × 10³ and 25 ± 5 × 10³ counts/min [¹⁴C]citrulline/mg protein in newborns and adults, respectively (P < 0.005). Nitrate-nitrite production by endothelial cells from adult pigs was elevated under basal conditions and in the presence of 10⁵ M bradykinin (data not shown); however, the low sensitivity of the assay did not allow parametric statistical analysis.

View larger version (28K): [in this window] [in a new window]   Fig. 2.   Nitric oxide (NO) synthase activity in cerebral microvascular endothelial cells from newborn and adult pigs. NO synthase activity was detected as NG-monomethyl-L-arginine (L-NMMA)-inhibited formation of [14C]citrulline from L-[14C]arginine. Confluent quiescent cells were incubated for 30 min at 37°C with L-[14C]arginine (0.5 µCi/ml), artificial cerebrospinal fluid (aCSF, control), acetylcholine (105 M), or bradykinin (105 M) in absence or presence of 1 mM L-NMMA. [14C]citrulline formation was measured as described in MATERIALS AND METHODS. Each point represents means ± SE of 3 experiments. * P < 0.05 compared with control values in each group.  P < 0.05 between newborns and adults.

Cyclooxygenase activity and prostanoid production in cerebral microvessels and in cultured microvascular endothelial cells. Cyclooxygenase activity (as prostanoid production from exogenous arachidonic acid) in freshly isolated cerebral microvessels and in cultured endothelial cells from adult pigs was similar to or even higher than that of newborn pigs (Figs. 3 and 4). Cultured cerebral microvascular endothelial cells produce vasodilator prostanoids, prostacyclin (as 6-keto-PGF₁), and PGE₂ under basal conditions and on stimulation (Fig. 4, A and B). We found no differences in the ability of cells from newborn and adult pigs to produce prostanoids from endogenous substrate under basal conditions or when ionomycin was added to stimulate phospholipase A₂ (PLA₂) (Fig. 4, A and B). We also found no age-related differences in receptor-mediated responses of endothelial cells; acetylcholine (10⁵ M) and bradykinin (10⁵ M) stimulated PGE₂ production two- to threefold above the basal level in newborn and adult pigs (Fig. 5).

View larger version (21K): [in this window] [in a new window]   Fig. 3.   Cyclooxygenase activity in cerebral microvessels from newborn and adult pigs as production of 6-keto-PGF1 (A) and PGE2 (B) from exogenous arachidonic acid. Freshly isolated cerebral microvessels were untreated (control) or pretreated for 20 min with NS-398 (105 M) or indomethacin (105 M). To determine cyclooxygenase activity, microvessels were incubated with fresh medium (aCSF) containing 30 µM arachidonic acid for 20 min at 37°C. Each point represents means ± SE of 2 separate experiments. * P < 0.05 compared with control values in each group.

View larger version (25K): [in this window] [in a new window]   Fig. 4.   Production of 6-keto-PGF1 (A) and PGE2 (B) by cerebral microvascular endothelial cells from newborn and adult pigs. Confluent quiescent cells were incubated with aCSF (control), 10 µM arachidonic acid, or 2 × 106 M ionomycin for 15 min at 37°C. Each point represents means ± SE of 4 separate experiments. * P < 0.05 compared with control values in each group.

View larger version (28K): [in this window] [in a new window]   Fig. 5.   Acetylcholine and bradykinin stimulate prostanoid production by cerebral microvascular endothelial cells from newborn and adult pigs. Confluent quiescent cells were incubated with aCSF (control), acetylcholine (105 M), or bradykinin (105 M) for 15 min at 37°C. Each point represents means ± SE of 3 separate experiments. * P < 0.05 compared with control values in each group.

We investigated the functional contribution of COX-1 and COX-2 to prostanoid synthesis in cerebral microcirculation of newborn and adult pigs. As we have reported before, NS-398 (10⁵-10⁴ M) effectively inhibited cyclooxygenase activity in COX-2 expressing cells (in this case, newborn pig vascular endothelial and smooth muscle cells in primary culture) (19, 20) with no effect in COX-1 expressing cells (subcultured HUVECs and Swiss 3T3 fibroblasts) (19). These data indicate that NS-398 can be used to probe for functional contributions of COX-2 to prostanoid production. NS-398 inhibited total cyclooxygenase activity in cerebral microvessels (30-50%; Fig. 3, A and B) and in endothelial cells from newborn and adult pigs (50-70%; Fig. 6B), indicating functional recruitment of COX-2 in both age groups. COX-1 (as reflected by NS-398-resistant prostanoid synthesis) contributed to 30-70% of the total prostanoid synthesis-COX activity in cerebral microvessels and in endothelial cells from both newborn and adult pigs (Figs. 3 and 6).

View larger version (18K): [in this window] [in a new window]   Fig. 6.   Functional contribution of cyclooxygenase isoforms to prostanoid synthesis in cerebral microvascular endothelial cells. Confluent quiescent cells were untreated (control) or pretreated at 37°C for 20 min with NS-398 (105 M). Cells were incubated with aCSF in absence (A) or presence of 10 µM arachidonic acid (B) or 2 × 106 M ionomycin (C) for 20 min at 37°C. Each point represents means ± SE of 3 experiments. * P < 0.05 compared with control values in each group.

Prostanoid production and cyclooxygenase activity in cerebral microvascular smooth muscle cells. PGE₂ is a major prostanoid produced by cerebral microvascular smooth cells (22). In smooth muscle cells from adult pigs, PGE₂ production from endogenous substrate under basal conditions was significantly decreased compared with that in cells from newborn animals (Fig. 7A). However, in the presence of ionomycin, the functional difference between the two age groups was eliminated (Fig. 7A). When exogenous arachidonic acid was provided to stimulate the cyclooxygenase activity, no differences were observed between cells from newborn and adult pigs (Fig. 7A). The dynamics of 6-keto-PGF₁ in smooth muscle cells from newborn and adult pigs revealed a similar pattern under both basal and stimulated conditions (data not shown). These data indicate that prostanoid production by cerebral microvascular smooth muscle cells is decreased in adult animals due to decreased PLA₂ activity, whereas cyclooxygenase activity is similar in both age groups.

View larger version (20K): [in this window] [in a new window]   Fig. 7.   Functional contribution of cyclooxygenase isoforms to prostanoid synthesis in cerebral microvascular smooth muscle cells. Confluent quiescent cells were untreated (control) or pretreated at 37°C for 20 min with NS-398 (105 M). Cells were incubated with aCSF (control), 2 × 106 M ionomycin, or 10 µM arachidonic acid for 20 min at 37°C. A: total prostanoid synthesis. B: COX-2 contribution (as NS-398-inhibited prostanoid synthesis). C: COX-1 contribution (as NS-398-resistant prostanoid synthesis). Each point represents means ± SE of 3 experiments. * P < 0.05 compared with control values in each group.  P < 0.05 between newborns and adults.

We investigated the functional contribution of COX-1 and COX-2 to prostanoid production by cerebral smooth muscle cells. We found no difference between NS-398-inhibited prostanoid production from exogenous arachidonic acid (indicator of COX-2 activity) and NS-398-resistant prostanoid production (indicator of COX-1 activity) in newborn and adult pigs (Fig. 7, B and C). Under all experimental conditions, COX-2 accounted for 40-70% of total prostanoid synthesis in smooth muscle cells of newborn and adult pigs.

Responsiveness of cerebral microvascular smooth muscle cells to vasodilator agents. We compared the ability of cultured cerebral microvascular smooth muscle cells from newborn and adult pigs to respond to prostacyclin and NO by activation of an appropriate second messenger system (cAMP and cGMP, respectively). Basal production of cAMP and cGMP was similar in quiescent confluent smooth muscle cells from newborns and adults (9.1 ± 2.2 and 13.1 ± 2.8 pmol cAMP/mg protein, respectively; P > 0.05; 3.2 ± 0.2 and 3.8 ± 0.2 pmol cGMP/mg protein, respectively, P > 0.05; N = 6 independent cultures). We found no age-related differences in responsiveness of vascular smooth muscle to the prostacyclin receptor agonist iloprost (10¹⁰-10⁶ M) and to the NO donor sodium nitroprusside (10¹⁰-10⁶ M). Iloprost caused similar dose-dependent increase in smooth muscle cAMP level in both age groups (3- to 4-fold increase at 10⁶ M, P > 0.05, Fig. 8A). Smooth muscle cells from newborns and adults also demonstrated equal potencies to respond to sodium nitroprusside by increasing cGMP level (2- to 2.5-fold increase at 10⁶ M, P > 0.05, Fig. 8B).

View larger version (15K): [in this window] [in a new window]   Fig. 8.   Responses of cerebral microvascular smooth muscle cells to iloprost (A) and sodium nitroprusside (B) by activation of second messenger systems. Confluent quiescent cells were incubated with aCSF containing indicated concentrations of iloprost or sodium nitroprusside for 15 min at 37°C. A: cAMP formation in response to iloprost. B: cGMP formation in response to sodium nitroprusside. Each point represents means ± SE of 6 experiments.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

In vivo physiological studies in pigs conducted in our laboratory for several years demonstrate that characteristics of endothelial regulation of cerebral blood flow are age dependent (1, 31, 32). In newborn pigs, endothelium-dependent cerebral vascular reactivity to such physiologically important stimuli as hypercapnia and acetylcholine is mediated exclusively by prostanoids but not NO (1, 22, 32), whereas both prostanoids and NO contribute to vascular tone in older pigs (1, 31, 32). In the present study, we compared the expression and activities of NO synthase and cyclooxygenase in isolated cerebral microvessels and in cultured microvascular endothelial cells from newborn and adult pigs. Our results demonstrate that the expression and the activity of eNOS, as well as the receptor-mediated activation of the enzyme by bradykinin and acetylcholine, are increased in adults. No differences between newborn and adult animals were found in the cyclooxygenase isoforms (COX-1 and COX-2) expression or in their activities and functional contribution to the basal or the stimulated endothelial prostanoid synthesis. These data demonstrate that in cerebral microvasculature in pigs, NO production is developmentally upregulated due to the increased expression of eNOS on pig maturation.

NO plays an important role in the regulation of endothelium-dependent responses of cerebral blood flow to hypercapnia and acetylcholine in different adult animal species (6, 24, 26) and in juvenile pigs (1, 31, 32). In juvenile pigs, inhibition of NO production by L-NMMA attenuated the cerebral vasodilation response to hypercapnia (31, 32) and acetylcholine (32), whereas in newborn pigs, L-NMMA did not affect vascular responses to these stimuli (22, 31, 32). These data indicate that endothelial NO-mediated cerebral vascular reactivity in pigs develops postnatally. The absence of NO-mediated responsiveness in the cerebral circulation of newborn pigs could be associated with 1) a low level of basal NO synthesis by the vascular endothelium (decreased expression and/or activity of NO synthase, the key enzyme in NO synthesis or low efficiency of receptor-mediated activation of NO synthase); 2) altered responsiveness of vascular smooth muscle to endothelium-derived NO (decreased activity of NO-activated soluble guanylyl cyclase and/or ineffective coupling between cGMP formation and vascular smooth muscle relaxation). However, we may reject the latter possibility on the basis of our in vivo data that sodium nitroprusside, a NO donor, effectively dilated pial arterioles and increased cortical cGMP in both newborn and juvenile pigs (1). Our present data demonstrate that cerebral microvascular smooth muscle cells from newborn and adult pigs equally respond to a NO donor by activation of cGMP second messenger system. These findings in vivo and in vitro indicate that NO transduction mechanisms are completely developed in the cerebral microvasculature in newborn pigs.

Our present data demonstrate that the production of NO in the cerebral microcirculation is increased in adult pigs compared with newborns. In isolated cerebral microvessels from adult pigs, eNOS-immunoreactive protein and the basal NO synthase activity were two- to threefold higher than in newborn pigs; the differences in NO production between adults and newborns were also retained in primary cultures of microvascular endothelial cells. eNOS is activated by such vasoactive substances as acetylcholine and bradykinin via receptor-mediated mechanisms (3-5, 7, 9, 10, 30). We found that cerebral microvascular endothelial cells from both newborn and adult animals responded to acetylcholine or bradykinin by increasing NO synthase activity 1.3- to 2-fold. Although we have found no differences between the two age groups in relative magnitude of receptor-mediated responses, the values of acetylcholine- and bradykinin-stimulated NO production were significantly higher in adult animals due to the elevated basal NO synthase activity. Quantitative differences between adult and newborn animals were also retained in primary cultures of endothelial cells from cerebral microvessels. Our results suggest that the expression and activity of eNOS are upregulated in the cerebral microcirculation of adult pigs compared with neonatal pigs.

We investigated whether endothelial prostanoid production decreases with age. In vivo data in pigs indicate that the transition from newborn to juvenile age is accompanied by a decreased contribution of dilator prostanoids to endothelium-dependent cerebral vascular responses (1, 31, 32). Prostanoid synthesis is determined by the activities of two rate-limiting enzymes, PLA₂, which selectively cleaves arachidonic acid from membrane phospholipids, and cyclooxygenase (prostaglandin G/H synthase), which catalyzes the first committed step in the conversion of arachidonic acid into prostanoids (8, 28). We have found no differences in the ability of endothelial cells from cerebral microvessels of newborn and adult pigs to produce vasodilator prostanoids, prostacyclin, and PGE₂ during basal conditions. Ionomycin, which stimulates prostanoid synthesis in a Ca²⁺-dependent manner by activating PLA₂ and increasing endogenous arachidonic acid (16), caused a similar increase in prostanoid synthesis in newborn and adult animals. Acetylcholine and bradykinin, which increase PLA₂ activity via receptor-mediated mechanisms (11-14), stimulated prostanoid production by cerebral microvascular endothelial cells to the same degree in newborns and adults. We also found no age-related differences in total cyclooxygenase activity; intact cerebral microvessels and cultured endothelial cells from newborn and adult pigs effectively converted exogenous arachidonic acid to prostacyclin and PGE₂. Taken together, our data indicate that in the cerebral circulation in pigs, endothelial prostanoid synthesis, including activities of rate-limiting enzymes, cyclooxygenase, and, apparently, PLA₂, as well as receptor-mediated responsiveness to acetylcholine and bradykinin, does not decline on transition from newborn to adult age.

In the cerebral circulation of newborn pigs, COX-2 is a constitutively expressed cyclooxygenase isoform that provides for a major part of prostanoid synthesis (19, 23). COX-2 protein is immunodetected in isolated cerebral microvessels of newborn pigs (23) and in cultured microvascular endothelial and smooth muscle cells (19, 20). We compared the expression of COX-2 protein in endothelial cells from newborn and adult animals; no age-related changes were detected. To detect the functional contribution of COX-2 to prostanoid synthesis, we used NS-398, a selective inhibitor of COX-2. NS-398 (10⁵-10⁴ M) effectively inhibited cyclooxygenase activity in COX-2 expressing newborn pig cerebral microvascular cells in primary culture (19, 20) with no effect in COX-1 expressing cells (subcultured HUVECs and Swiss 3T3 fibroblasts) (19). In newborn pigs, COX-2 (as NS-398-inhibited prostanoid synthesis) accounts for 50-80% of total prostanoid synthesis in isolated cerebral microvessels (23) and in cultured microvascular endothelial and smooth muscle cells (19, 20). To investigate whether the functional contribution of endothelial COX-2 to the biosynthesis of vasodilator prostanoids is regulated age dependently, we compared the effects of NS-398 in cultured endothelial cells from newborn and adult animals. We found no age-related differences in the sensitivity of endothelial prostanoid synthesis to NS-398. COX-2 (as NS-398-inhibited prostanoid formation) accounted for 50-70% of the total amount of prostanoids produced by the vascular endothelium in newborn and adult pigs under basal conditions and on recruitment of PLA₂ by ionomycin. Similarly, the cyclooxygenase activity in endothelial cells (as prostanoid synthesis from exogenous arachidonic acid) was inhibited 50-70% by NS-398 in both newborn and adult animals. Altogether, these data demonstrate that in adult pigs, as well as in newborns, COX-2 is a constitutively expressed cyclooxygenase isoform that provides a major part of the cerebral vascular endothelial prostanoid synthesis.

It has been reported that in the brain cortex in vivo and in isolated cerebral microvasculature, NS-398 was less effective in inhibiting basal prostanoid synthesis in juveniles vs. newborn animals (23). Because smooth muscle cells have a high capacity for prostanoid biosynthesis (19, 20), we investigated whether a smooth muscle component of the cerebral microvasculature is altered on maturation. In cultured smooth muscle cells from cerebral microvessels of newborn pigs, the basal prostanoid synthesis was five- to sevenfold higher than in adult animals. However, cyclooxygenase activity was identical in smooth muscle cells from newborn and adult animals. Moreover, NS-398 inhibited 60-80% of the cyclooxygenase activity, indicating that COX-2 contributes to the majority of vascular smooth muscle prostanoid synthesis in both newborn and adult animals. The differences in smooth muscle prostanoid production between newborn and adult pigs were completely eliminated when ionomycin was added to stimulate PLA₂ and to increase the formation of endogenous arachidonic acid. These data indirectly indicate that PLA₂ appears to be the key enzyme that accounts for the decreased prostanoid synthesis in cerebral vascular smooth muscle in adult animals.

Physiological vascular responses may depend not only on production of endothelium-derived vasorelaxant factors but also on the responsiveness of vascular smooth muscle. To address this possibility, we investigated the responsiveness of vascular smooth muscle to prostacyclin and NO by activation of appropriate second messenger system (cAMP and cGMP, respectively). As we conclude from our data, newborn and adult animals have similar prostacyclin receptor-mediated responses as indicated by increased cAMP formation in vascular smooth muscle. We also found no differences between newborn and adult cGMP responses to the NO donor sodium nitroprusside. These data indicate that the two major signal transduction systems (cAMP mediated and GMP mediated) coexist in vascular smooth muscle in newborn and adult animals and have similar sensitivity to the agonists used regardless of age.

Relative contributions of prostanoids and NO to control of cerebral circulation may differ among species (15). In those animals studied, including humans, contributions of each endothelium-derived vasodilator to cerebral blood flow regulation appear to be different in neonates and older individuals. In adult rats, both PG and NO contribute to hypercapnia-induced vasodilation, a major physiologically important cerebral vascular response (24). In juvenile pigs, both NO and prostanoids contribute to vasodilation of cerebral microvessels in response to hypercapnia and acetylcholine, whereas in newborn piglets a role for NO cannot be demonstrated (1, 31, 32). Our data demonstrate that endothelial NO production increases with age due to an increased NO synthase expression and activity in cerebral microvascular endothelial cells. Developmental upregulation of NO production by vascular endothelium may explain the observation that in juvenile pigs NO contributes to cerebral vasodilation responses to hypercapnia and acetylcholine, whereas in newborn pigs prostanoids but not NO contribute to hypercapnia-induced dilation, and acetylcholine causes constriction (1, 31, 32). Dilator prostanoids appear to be important in cerebral circulation of newborn babies. By comparison to newborn piglets, premature human neonates appear to be extremely sensitive to the cyclooxygenase inhibition. COX inhibitors, especially indomethacin, at very low doses are widely used in clinical practice in preterm human babies with patent ductus arteriosus. It has been shown that indomethacin causes a significant reduction in blood flow to the brain and inhibits vasodilation to hypercapnia, suggesting contribution of dilator prostanoids to the cerebral blood flow regulation in human preterms (15). Experimental studies of age-dependent mechanisms regulating cerebral circulation are important in understanding pathophysiological mechanisms underlying such serious medical problems as intracranial hemorrhages, asphyxia complications, and stroke in patients.

In summary, the results of the present study demonstrate developmental changes in the production of endothelium-derived vasorelaxant factors in the cerebral microcirculation of newborn and adult pigs. The expression of eNOS and the production of NO are upregulated in adult animals, whereas the expression of endothelial cyclooxygenase, the functional role of COX-1 and COX-2 isoforms, and the production of endothelium-derived dilator prostanoids remain unaltered on maturation. These changes may account for the increased contribution of NO-mediated influences in the cerebral microcirculation of pigs on maturation; prostanoid-mediated endothelial influences remain important in both newborn and older pigs.