Prostaglandin G/H synthase-2 is a major contributor of brain prostaglandins in the newborn.

In order to understand the molecular basis of the elevated cerebral prostaglandin levels in the newborn, we compared the expression of the mRNAs and proteins of prostaglandin G/H synthases (PGHS), PGHS-1 and PGHS-2, in various regions of the brain and the microvasculature of newborn (1-2-day-old) and juvenile (4-7-week-old) pigs and also measured the relative contribution of PGHS-2 to cerebral prostaglandin synthesis both in vivo and in vitro by using a novel inhibitor of PGHS-2, NS-398. Ribonuclease protection assays using total RNA isolated from various regions of the porcine brain revealed that, unlike PGHS-1 mRNA, PGHS-2 mRNA was abundantly expressed in the cortex and the microvasculature of the newborn compared with those of the juvenile animal. PGHS-2 immunoreactive protein comprised the majority of total PGHS enzyme in neonatal cerebral microvasculature due to a 2-3-fold lower expression of immunoreactive PGHS-1 protein. Inhibition of PGHS-2 by NS-398 decreased the rate of prostaglandin synthesis by purified cerebral microvessels of the newborn by approximately 65% and of juvenile pigs by 30%. The decrease in brain tissue prostaglandin concentrations following intravenous administration of NS-398 was greater in newborn pigs (≥90%) than in the juvenile animals (≤30%). Furthermore, NS-398 substantially reduced the net in vivo cerebrovascular production of prostaglandins in newborn pigs. Taken together, these results indicate that PGHS-2 is the predominant form of prostaglandin G/H synthase in the newborn brain and cerebral microvasculature and the main contributor to the brain prostaglandin levels in the newborn animal.

Prostaglandins act as modulators in several neurological (1,2) and cerebral hemodynamic functions (3,4). During the perinatal period the concentrations of prostaglandins in blood and brain in the newborn are higher than those in the normal adult (5,6). These higher levels of cerebral prostanoids in the newborn significantly affect cerebral blood flow autoregulation as well as cause down-regulation of prostanoid receptor expression and receptor function in brain (7,8). However, the cause of increased prostaglandin levels and the relative contributions of the two prostaglandin G/H synthases (PGHS) 1 to prostanoid synthesis in the neonatal brain are not yet known.
Of the two prostaglandin G/H synthases so far described, PGHS-1 (EC 1.14.99.1) is constitutively expressed in all tissues, albeit to varying degrees (for reviews see Refs. 9 -11). The other isozyme, PGHS-2, shares significant homology with PGHS-1 in amino acid sequence (11) and exhibits similar enzymatic properties (12)(13)(14) but differs in its pharmacological properties (14 -16), mRNA size (11), chromosomal location (17), and gene organization (18,19). Moreover, PGHS-2 can be rapidly induced in various tissues by diverse stimuli such as mitogenic agents, growth factors (20,21), hormones (22,23), inflammatory agents (24), synaptic activity (25), and muscle stretch/relaxation (26). Elevation of PGHS-2 expression in inflammation (10,11) and suppression of PGHS-2 gene activity by dexamethasone and other corticosteroids both in tissue cultures (20) and in vivo (27) suggest that PGHS-2 may have a role in inflammatory response. Nonetheless, low but varying levels of PGHS-2 expression have been found in all tissues (28) by using the reverse transcription-polymerase chain reaction technique. Thus, despite its induction in inflammation and mitogenesis, the role of PGHS-2 under normal physiological conditions of increased prostaglandin synthesis in the brain, as seen in the perinatal period (5,6), is still a matter of conjecture. This is of particular interest given that expression of the other isozyme, PGHS-1, has been shown to be low in the newborn and increases to reach maximum levels in the adult (29). We hypothesized that the elevated prostaglandin G/H synthase activity in the newborn brain could be due to increased expression of PGHS-2. For this purpose, we analyzed the expression of porcine PGHS-1 and PGHS-2 mRNAs by ribonuclease protection assays and analyzed the expression of PGHS-1 and PGHS-2 proteins by immunochemical methods in the newborn and the juvenile animals. We also examined the relative contribution of PGHS-2 to the cerebral production of prostaglandins both in vivo and in vitro by using a PGHS-2 inhibitor, NS-398. MA). Aprotinin and leupeptin were from Boehringer Mannheim Canada (Montreal, Canada). Soybean trypsin inhibitor (type II-S), phenylmethylsulfonyl fluoride, arachidonic acid, ibuprofen, indomethacin, phorbol 13-myristate, dimethyl sulfoxide, and ␤-mercaptoethanol were from Sigma. NS-398 was from Biomol (Plymouth Meeting, CA). Ficoll-400, ribonuclease A, oligo(dT)-cellulose, and T 7 sequencing kit were from Pharmacia Biotech Inc. pGEM-3 plasmid vector and in vitro transcription kit were from Promega (Madison, WI). Protein assay and electrophoretic reagents were purchased from Bio-Rad. Taq polymerase, deoxynucleotides, guanidinium isothiocyanate, T 4 DNA ligase, T 4 polynucleotide kinase, immunoprecipitin, M-MLV reverse transcriptase, random hexamers, and restriction enzymes were purchased from Life Technologies, Inc. All other chemicals were of analytical reagent grade and were purchased from either Sigma or ICN Biochemicals.
Animals-Newborn (1-2-day-old) and juvenile (4 -7-week-old) pigs were purchased from Fermes Ménard Inc., L'Ange Gardien, Quebec and used according to a protocol of the Animal Care Committee of Ste-Justine Hospital, Montréal.
Purification of Porcine Cerebral Microvasculature-Newborn and juvenile pigs were anesthetized with 2% halothane and killed by intracardiac injection of pentobarbital (120 mg/kg). The brains were perfused with heparinized saline (200 -250 ml for newborn and 1 liter for juvenile pigs) to eliminate blood elements. The brain was removed and immediately kept in ice-cold Krebs buffer (120 mM NaCl, 4.5 mM KCl, 2.5 mM CaCl 2 , 1 mM MgCl 2 , 27 mM NaHCO 3 , 1 mM KH 2 PO 4 , 0.01 mM sodium edetate, and 10 mM glucose). Brain tissue was homogenized gently (5-6 strokes) in ice-cold phosphate-buffered saline containing 20% Ficoll-400 using a glass homogenizer with loose fitting glass pestle. The homogenate was centrifuged at 20,000 ϫ g for 20 min at 4°C. The pellet containing the microvessels was washed 3 or 4 times with 20 volumes of ice-cold phosphate-buffered saline to eliminate Ficoll. The resultant microvessel preparations were assessed for purity by light microscopy and ␥-glutamyl transpeptidase activity (8,30). The microvessels were used immediately for either total RNA isolation or prostaglandin G/H synthase assays.
Isolation of cDNA Probes for Porcine PGHS-1 and PGHS-2-Total and poly(A) ϩ RNA from porcine ileum and cerebral microvascular smooth muscle cell cultures that were stimulated with phorbol 13myristate (100 ng/ml) were obtained as described previously (31). Two micrograms of poly(A) ϩ RNA was reverse transcribed using 400 units of Moloney murine leukemia virus reverse transcriptase and 10 g/ml random hexamers, in a 50-l reaction containing 50 mM Tris-HCl, pH 8.3, 75 mM KCl, 3 mM MgCl 2 , 10 mM dithiothreitol, and 0.5 mM each of dCTP, dGTP, dATP, and dTTP, for 1 h at 42°C. An aliquot of the cDNA (equivalent to 1 g of poly(A) ϩ RNA) was amplified using 1.  (32,33), human (34,35), and rat (36) cDNAs. The reaction products, an 0.82kilobase DNA fragment for porcine PGHS-1 and 0.8-kilobase fragment DNA for porcine PGHS-2, were phosphorylated by T 4 kinase and cloned in pGEM-3 vector by blunt end ligation using T 4 DNA ligase. Multiple plasmid clones were sequenced using the T7 sequencing kit to determine the authentic sequence of porcine PGHS-1 and PGHS-2 partial cDNAs.
Synthesis of 32 P-Labeled Antisense RNA Probes and Ribonuclease Protection Assays-32 P-Labeled antisense RNA probes for PGHS-1 and PGHS-2 were prepared using an in vitro transcription kit (Promega). The specific activities of both PGHS-1 and PGHS-2 cRNA probes were approximately equal. Total RNA was extracted from cortex, cerebellum, medulla oblongata, hippocampus, thalamus, periventricular area, retina, choroid, lung, and cerebral microvasculature of newborn and juvenile pigs as described above. Aliquots of the total RNAs were resolved by formaldehyde-agarose gel electrophoresis and stained with ethidium bromide to assess the integrity and also to verify quantitation by spectrophotometry. The RNase protection assays were conducted following a published protocol (37) with minor modifications. Briefly, 50 -100 g of total RNA was incubated overnight at 50°C with 5 ϫ 10 5 cpm of either PGHS-1 or PGHS-2 probes in 40 l of hybridization buffer (80% deionized formamide, 40 mM PIPES, pH 6.8, 1 mM EDTA, and 0.4 M NaCl). The RNA hybrids were digested in 400 l of digestion buffer (20 mM Tris-HCl, pH 7.5, 1 mM EDTA, and 0.3 M NaCl) containing ribonuclease A (40 g/ml) for 30 min. The digestion products were purified by phenol/chloroform extraction and ethanol precipitation. The protected RNA fragments were resolved on urea, 6% polyacrylamide gels and autoradiographed using Kodak-XO-matic film. In addition, the bands were visualized and quantified by PhosphorImager (Molecular Dynamics).
Immunoprecipitation and Western Blotting of PGHS-1 and PGHS-2-Purified microvessels from newborn and juvenile brains were homogenized in ice-cold buffer (20 mM Tris-HCl, pH 8.0, 1 mM EDTA, 150 mM NaCl, 1% Nonidet P-40, and 10 g/ml each of leupeptin, aprotinin, soybean trypsin inhibitor, 0.2 mM phenylmethylsulfonyl fluoride) with an Omni 2000 tissue grinder (Omni International, Waterbury, CT) for 15-20 s. The homogenates were centrifuged at 12,000 ϫ g for 10 min at 4°C. Total protein concentration of the supernatant was determined by using the Bradford dye-binding assay (Ref. 38; Bio-Rad protein assay reagent) and bovine serum albumin as the standard. All operations, unless specified, were conducted at 4°C. An aliquot of the total protein (2 mg) was preadsorbed with 50 l of immunoprecipitin for 30 min, followed by centrifugation at 12,000 ϫ g for 10 min to remove the immunoprecipitin. The supernatants were incubated with PGHS-1-or PGHS-2-specific antibodies for 1.5 h with gentle agitation, and the immune complexes were collected by incubation with 50 l of immunoprecipitin for 30 min followed by centrifugation. The selectivity of the antibodies for PGHS-1 and PGHS-2 was assured by the supplier (Merck-Frosst) and tested by us both in immunoprecipitation and Western blotting. The immune precipitates were washed with wash buffer (20 mM Tris-HCl, pH 8.0, 150 mM NaCl, 1 mM EDTA) three times, denatured in SDS-sample buffer (125 mM Tris-HCl, pH 6.8, 2% (w/v) SDS, 10% (v/v) glycerol, 0.1 mg/ml Bromphenol Blue) for 15 min at room temperature, and centrifuged at 12,000 ϫ g for 15 min to remove the immunoprecipitin. After adding ␤-mercaptoethanol to a final volume of 10%, the supernatants were boiled for 5 min before loading on SDSpolyacrylamide gels.
The proteins were electrophoretically transferred to nitrocellulose membranes, and the nonspecific binding sites on the membranes were blocked with buffer A (20 mM Tris-HCl, pH 7.5, 150 mM NaCl, 1 mM EDTA, 0.1% (v/v) Tween 20) containing 3% skim milk for 1 h. The membranes were briefly rinsed with buffer B (20 mM Tris-HCl, pH 7.5, 1 mM EDTA, 150 mM NaCl) and incubated for 1 h in buffer A containing 1% skim milk and PGHS-1-or PGHS-2-specific polyclonal rabbit antibodies (1:7500). The membranes were washed six times (5 min each) with buffer B. A second incubation with horseradish peroxidase-conjugated anti-rabbit IgG antibodies (Amersham Corp.) in buffer A containing 1% skim milk for 1 h, followed by several washes of the membranes was conducted as described above. Finally, the immunoreactive bands were visualized by using the enhanced chemiluminescence kit (Amersham Corp.) as instructed by the manufacturer.
Prostaglandin Synthesis by Purified Cerebral Microvessels-Purified cerebral microvessels from newborn and juvenile pigs were homogenized in ice-cold buffer (50 mM Tris-HCl, pH 7.5, 1 mM EDTA, 0.2 mM phenylmethylsulfonyl fluoride, and 10 g/ml each of leupeptin, aprotinin, soybean trypsin inhibitor) using an Omni-2000 tissue grinder for 20 -30 s. An aliquot of the homogenate (400 -600 g of protein/ml) was incubated with 50 M arachidonic acid for 10 min at 25°C in the absence or presence of the following PGHS inhibitors: indomethacin (0.1 mM), ibuprofen (1 mM), and NS-398 (0.1 mM). The homogenate was preincubated with vehicle or PGHS inhibitors for 20 min at 25°C prior to the addition of arachidonic acid. Ibuprofen and indomethacin have been shown to inhibit both enzymes at the concentrations used here (14), and 0.1 mM NS-398 markedly blocks PGHS-2 activity but has little effect on PGHS-1 (20). At the end of the incubation period, the samples were boiled for 2 min and centrifuged to remove the flocculate. PGE 2 , PGF 2␣ , and 6-keto-PGF 1␣ concentrations in the supernatants were determined by radioimmunoassay as described previously (39).
Tissue Prostaglandin Levels and Cerebrovascular Prostaglandin Synthesis in Vivo-Newborn and juvenile pigs were anesthetized with 1.5% halothane. Tracheostomy was performed, and the animals were ventilated with air using a Harvard animal respirator. Catheterization of the right femoral artery for measurement of blood pressure (Gould TA240), blood gases, and arterial blood prostaglandins, and of the right femoral vein for drug injections were performed as described before (4). After the surgery, halothane was discontinued, and the animals were sedated with ␣-chloralose (50 mg/kg bolus followed by an infusion of 10 mg/kg/h) and paralyzed with pancuronium (0.1 mg/kg intravenously). The animals were allowed to stabilize for 2 h and then were injected with either saline or NS-398 (10 mg/kg intravenously). Forty-five min later, the animals were killed by intracardiac injection of pentobarbital (120 mg/kg), and the brain and eyes were immediately removed and frozen in liquid nitrogen. Brain cortex and retinal prostaglandin concentrations were determined by radioimmunoassay after extraction on octadecylsilyl silica columns (40).
In a separate study designed to measure cerebrovascular production of prostaglandins, newborn pigs were anesthetized as above, and catheters were placed in the left ventricle via the right subclavian artery and in the sagittal sinus. After the animals were allowed to stabilize for 2 h, blood samples were collected from the femoral artery and sagittal sinus, before and 45 min after injection of NS-398 (10 mg/kg intravenously) for prostaglandin determination (41). Cerebral blood flow was measured using radiolabeled microspheres as described in detail previously (4,41). The net cerebrovascular production of prostaglandins was calculated as total cerebral blood flow ϫ difference in prostaglandin concentrations of sagittal sinus and arterial blood samples and expressed as ng/min/100 g of brain tissue. Similar studies were not done on juvenile pigs because of surgical difficulties in catheterization of sagittal sinus.
Statistical Analysis-Data were analyzed by paired or unpaired Student's t test, and a p Ͻ 0.05 was assumed to denote significance. Throughout this paper data are presented as means Ϯ S.E.

Tissue Distribution of PGHS-1 and PGHS-2-
The cDNA probes of PGHS-1 (0.82 kilobase) and PGHS-2 (0.8 kilobase) isolated by reverse transcription-polymerase chain reaction of the mRNA comprised approximately 45% of the reading frames of both cDNAs. Sequence analysis (data not shown) revealed that both porcine enzymes shared significant amino acid identity (85-90%) with their respective murine (32,34), human (34,35), and rat (36) enzymes; in addition, some functionally and catalytically important amino acids as well as amino acid sequences identified in porcine PGHS-1 and PGHS-2 cDNAs were totally conserved. Radiolabeled cRNA probes generated from these cDNAs were used in RNase protection assays of total RNAs isolated from various tissues of the newborn and juvenile pigs. As shown in Fig. 1A and B, PGHS-1 mRNA was expressed in all tissues at varying levels. Non-neural tissues such as the choroid and lungs of both newborn and juvenile animals (lanes 16 -19) expressed PGHS-1 mRNA abundantly. Among tissues of neural origin, high expression of PGHS-1 was seen in the medulla oblongata (lanes 6 and 7). Densitometric analyses of the radioactive bands revealed that there was 1.5-2.0-fold higher expression of PGHS-1 in brain tissue of the juvenile than in that of the newborn animal (Fig. 1B), especially in the cortex (lanes 2 and 3), cerebellum (lanes 4 and 5), medulla oblongata (lanes 6 and 7), hippocampus (lanes 10 and 11), and thalamus (lanes 12 and 13).
In contrast to PGHS-1 mRNA, PGHS-2 mRNA in both neural and non-neural tissues was barely detectable; a significant exception was the brain cortex and, to a lesser extent, the choroid of the newborn (Fig. 1, A and B, lanes 3 and 17,  respectively). Newborn brain cortex contained a 3-fold greater expression of PGHS-2 mRNA compared with juvenile cortex (Fig. 1, A and B, lanes 2 and 3). The abundance of PGHS-2 mRNA was variable in different regions of the brain and in other tissues; where PGHS-2 mRNA was detectable, expression of PGHS-2 was higher in the newborn than in juvenile animals.

PGHS-1 and PGHS-2 Protein Expression in Cerebral
Microvasculature-To demonstrate if the ontogenic differences in the mRNA levels of PGHS-1 and PGHS-2 in newborn and juvenile cerebral microvasculature were reflected in the protein levels, immunoprecipitations of PGHS-1 and PGHS-2 proteins with specific antibodies followed by Western blotting were conducted (Fig. 3). There was a 3-4-fold lesser expression of immunoreactive PGHS-1 in newborn than in juvenile cerebral microvasculature, whereas expression of PGHS-2 was comparable in newborn and juvenile tissues.
Ontogenic Differences in Prostanoid Synthesis by Purified Cerebral Microvasculature-The rates of prostaglandin synthesis by homogenates of cerebral microvessels from newborn and juvenile pigs in the presence of 50 M arachidonic acid were compared. Prostaglandins were synthesized at apparent V max for at least 10 min in this protocol. The basal production of PGE 2 , PGF 2␣ , and 6-keto-PGF 1␣ (in the absence of PGHS inhibitors) was 2-3-fold greater in preparations from the newborn than from juvenile animals (data not shown). Ibuprofen and indomethacin caused a Ͼ80% decrease in prostanoid synthesis by cerebral microvessels from both newborn and juvenile pigs (Fig. 4). NS-398 decreased prostanoid production by cerebral microvessels from the newborn (60 -67%), but only caused a small reduction in prostanoid synthesis by cerebral microvessels from juvenile pigs (23-35%).
Effects of NS-398 on Prostaglandin Synthesis in Vivo-Intravenous administration of NS-398 did not alter arterial pH, pO 2 , pCO 2 , blood pressure, and heart rate of the animals. PGE 2 and  Lanes 2,4,6,8,10,12,14,16, and 18 contain samples from the juvenile tissues, and lanes 3, 5, 7, 9, 11, 13, 15, 17, and 19 contain samples from the newborn tissues. Total RNA (50 g) isolated from various regions of the brain and other tissues was subjected to RNase protection analysis as described under "Experimental Procedures." Autoradiographic exposure was for 7 days. PGF 2␣ concentrations were higher in the brain cortex and retina of the newborn than of juvenile pigs (Fig. 5). NS-398 reduced prostaglandin concentrations by Ն90% in the brain cortex of the newborn (Fig. 5), comparable with what was observed previously using ibuprofen and indomethacin (42); in contrast, NS-398, decreased brain prostaglandin levels in the juvenile animals only by 26 -30%. In retina where mRNA for PGHS-2 was not detectable (Fig. 1, lanes 16 and 17), NS-398 caused a 21-33% reduction in PGE 2 and PGF 2␣ concentrations in tissues from newborn and juvenile animals.
To assess the effect of NS-398 on cerebrovascular prostaglandins in the newborn animal, their concentrations in blood samples from arterial and sagittal sinus blood were determined before and after the injection of NS-398 to newborn pigs, and in vivo cerebrovascular prostaglandin production was calculated (4). Net cerebrovascular production of prostaglandins was reduced by Ͼ65% in response to NS-398 treatment (Fig. 6); this decrease was unrelated to cerebral blood flow, which actually increased by 33-45% after NS-398 treatment.

DISCUSSION
Several studies have reported that prostaglandin levels are elevated in the neonatal blood and brain during the perinatal period (5,6,8). However, the reasons for this increase in cerebral prostaglandins are not known. We tested the hypothesis that high prostaglandin levels in brain during neonatal period may be due to increased PGHS-2 activity and provided two main lines of evidence in support of this hypothesis. First, brain cortex and microvasculature of the newborn expressed more PGHS-2 mRNA than the juvenile animal, whereas PGHS-1 mRNA was more abundant in the juvenile than in cerebral cortex and microvessels of the newborn. Moreover, PGHS-2 comprised the majority of immunoreactive PGHS proteins in the newborn brain cortex. Second, NS-398, a relatively specific PGHS-2 inhibitor, produced a much greater decrease in prostaglandin synthesis in the newborn compared with the juvenile pig.
RNase protection assays revealed that PGHS-1 mRNA was ubiquitously expressed but that its abundance differed within various regions of the brains of the newborn and juvenile pigs. The hindbrain and midbrain contained highest expression of PGHS-1. Similar observations have been made by others using different techniques such as Northern analysis and in situhybridization (29,43). However, the expression of PGHS-1 in brain is considerably lower than that in peripheral tissues such as the choroid and lungs, in accordance with other data (28,29). In addition to its diverse tissue expression, PGHS-1 mRNA increased with age in brain (29) and cerebral vasculature (Fig.  2); in other tissues such as fetal cotyledon and amnion, PGHS-1 expression does not increase with gestational age (44,45). The ontogenic increase in PGHS-1 mRNA in cerebral vasculature is associated with a corresponding 3-fold increase in immunoreactive PGHS-1 in brain microvessels of the juvenile animal.
In contrast to PGHS-1, PGHS-2 mRNA was not readily detectable in various regions of the porcine brain. However, the highest PGHS-2 expression was observed in brain cortex and microvasculature of newborn animals. PGHS-2 mRNA levels decreased with age, and although immunoreactive PGHS-2 did not exhibit similar ontogenic changes, PGHS-1 protein and mRNA were markedly less in the newborn, thus disclosing the relative abundance of PGHS-2 in newborn brain and microvessels.
Further support for the suggestion that increased PGHS-2 expression accounts for the high prostaglandin G/H synthase activity and, in turn, high prostaglandin levels in newborn comes from studies using PGHS inhibitors. NS-398 is more than 100-fold more potent in inhibiting PGHS-2 than PGHS-1 . Arrows point to PGHS-1 and PGHS-2 polypeptides (70 kDa) that were immunoprecipitated with specific antibodies from the detergent-lysates of cerebral microvasculature (2 mg of protein) isolated from newborn and juvenile pig brains. Following electrophoretic transfer of the proteins to nitrocellulose membranes, PGHS-1 and PGHS-2 polypeptides were detected by Western blotting and visualized by enhanced chemiluminescence (Amersham Corp.). (46). Furthermore, inhibition of PGHS-1 by NS-398 could be reversed by excess substrate, arachidonic acid, whereas timedependent inactivation of PGHS-2 by NS-398 renders PGHS-2 refractory to substrate-mediated relief of inhibition (47). In this context, pretreatment of microvessels with NS-398 may have enabled us to differentiate PGHS-1 and PGHS-2 in their contribution toward prostanoid synthesis by cerebral microvasculature.
NS-398 decreased prostaglandin levels and synthesis by Ͻ35% in juvenile tissues and by Ͼ60% in those of the newborn. The differential effects of NS-398 in newborn animals compared with the juvenile pigs were consistently observed in prostaglandin synthesis by isolated cerebral vasculature, in prostaglandin levels in brain cortex, and in in vivo cerebrovascular prostaglandin production. Moreover, unlike the brain prostaglandins, NS-398 minimally reduced prostanoids in the newborn retina (Fig. 5) in which PGHS-2 mRNA could not be detected (Fig. 1). Thus, the preponderance of PGHS-2 mRNA and protein in newborn cortex and microvasculature, minimal expression of immunoreactive PGHS-1 in the newborn, and pronounced inhibition of prostanoid synthesis both in vivo and in vitro by NS-398 in newborn tissues compared with those of juvenile animals, taken together, indicate that PGHS-2 is a major contributor to prostaglandin synthesis in newborn brain and cerebral microvasculature.
The factors responsible for the increased expression of PGHS-2 in newborn cerebral cortex and microvasculature are not known but may include estrogens which increase late in gestation and induce a gradual increase in PGHS activity (48,49). Indeed, high PGHS-2 expression and concomitant increase in prostanoid synthesis in late gestation and at term was observed in fetal placental tissues (45). Besides being transcriptionally regulated by various stimuli, PGHS-2 expression is also controlled by factors affecting mRNA stability (50) and its translatability (51). The mechanisms governing the increased expression of PGHS-2 in neonatal brain remain to be elucidated.
The function of the elevated prostaglandin concentrations in the brain is so far not known. Because prostaglandins have been shown to exhibit neuroprotective properties (52), It is likely that their increased levels during the perinatal period may provide protection to the fetal brain toward the end of parturition when oxygen tension is markedly reduced (53) and the risk of hypoxic brain injury increases. Being a rapidly FIG. 5. Prostaglandin concentrations in the brain cortex and retina of newborn and juvenile pigs treated with saline or NS-398. Brain cortex and retina from saline-and NS-398-treated animals were homogenized, the prostaglandins were extracted on octadecylsilyl silica columns, and their concentrations were determined by radioimmunoassay as described under "Experimental Procedures." Results are the mean Ϯ S.E. of four experiments. *, p Ͻ 0.05; **, p Ͻ 0.01, salinetreated versus NS-398-treated. Open bars, saline; filled bars, NS-398.
FIG. 6. In vivo cerebrovascular production of prostaglandins in newborn pigs treated with NS-398. The net cerebrovascular production of prostaglandins in vivo was determined before and 45 min after administration of NS-398 and calculated as the product of total cerebral blood flow times the difference in prostaglandin concentrations of sagittal sinus blood and arterial blood, which was expressed as ng/min/100 g of brain tissue; cerebral blood flow was measured using the microsphere technique and prostaglandins by radioimmunoassay,  4. Prostanoid synthesis in vitro by purified cerebral microvasculature from newborn and adult pigs. Results, calculated as means Ϯ S.E. of quadruplicate experiments, are expressed as percentages of the basal production of prostaglandins (in the absence of PGHS inhibitors) in tissues from the corresponding age group; basal synthesis of PGE 2 , PGF 2␣ , and 6-keto-PGF 1␣ was 2-3-fold greater in tissues from newborns than from adults. The PGHS inhibitors, ibuprofen (1 mM), indomethacin (0.1 mM), and NS-398 (0.1 mM), were preincubated with the homogenates of cerebral microvessels for 20 min at 25°C, and the reaction was initiated by adding 50 M arachidonic acid. The reaction rate was linear with time for 10 min. The samples were boiled for 2 min and clarified by centrifugation. Prostaglandin concentrations in the supernatants were determined by radioimmunoassay. *, p Ͻ 0.05, newborn versus juvenile. Open bars, juvenile; filled bars, newborn. inducible enzyme, PGHS-2 would be suited for such a temporary but important role in perinatal life.