Molecular Characterization of an Arachidonic Acid Epoxygenase in Rat Brain Astrocytes
Background and Purpose Brain parenchymal tissue metabolizes arachidonic acid (AA) via the cytochrome P450 (P450) epoxygenase to epoxyeicosatrienoic acids (EETs). EETs dilate cerebral arterioles and enhance K+ current in vascular smooth muscle cells from large cerebral arteries. Because of the close association between astrocytes and the cerebral microcirculation, we hypothesized that brain epoxygenase activity originates from astrocytes. This study was designed to identify and localize an AA epoxygenase in rat brain astrocytes. We also tested the effect of EETs on whole-cell K+ current in rat cerebral microvascular smooth muscle cells.
Methods A functional assay was used to demonstrate endogenous epoxygenase activity of intact astrocytes in culture. Oligonucleotide primers derived from the sequence of a known hepatic epoxygenase, P450 2C11, were used in reverse transcription/polymerase chain reaction of RNA isolated from cultured rat astrocytes. The appropriate size reverse transcription/polymerase chain reaction product was cloned into a plasmid vector and sequenced. A polyclonal peptide antibody was raised against P450 2C11 and used in Western blotting and immunocytochemical staining of cultured astrocytes. A voltage-clamp technique was used to test the effect of EETs on whole-cell K+ current recorded from rat cerebral microvascular muscle cells.
Results Based on elution time of known standards and inhibition by miconazole, an inhibitor of P450 AA epoxygenase, cultured astrocytes produce 11,12- and 14,15-EETs when incubated with AA. The sequence of a cDNA derived from RNA isolated from cultured rat astrocytes was 100% identical to P450 2C11. Immunoreactivity to glial fibrillary acidic protein, a marker for astrocytes, colocalized with 2C11 immunoreactivity in double immunochemical staining of cultured astrocytes. EETs enhanced outward K+ current in muscle cells from rat brain microvessels.
Conclusions Our results demonstrate that a P450 2C11 mRNA is expressed in astrocytes and may be responsible for astrocyte epoxygenase activity. Given the vasodilatory effect of EETs, our findings suggest a role for astrocytes in the control of cerebral microcirculation mediated by P450 2C11-catalyzed conversion of AA to EETs. The mechanism of EET-induced dilation of rat cerebral microvessels may involve activation of K+ channels.
The role of fatty acid metabolites as paracrine and autocrine mediators of organ blood flow is well documented.1 Membrane phospholipids of the brain are highly enriched in polyunsaturated fatty acids, especially arachidonate.2 AA is found in esterified form in cell membrane phospholipids, from which it can be liberated by a variety of physiological, pharmacological, and pathological stimuli.1 Free AA is subsequently metabolized by the cyclooxygenase, lipoxygenase, and the P450 enzymatic pathways to a number of biologically active metabolites. All three pathways for the metabolism of AA have been described in the brain.3 4 The role of cyclooxygenase and lipoxygenase enzymes in signal transduction and cell-cell interactions has been thoroughly studied. Recently, a number of studies have focused attention on the role of the P450 metabolites of AA in the regulation of vascular tone. The P450 enzymes in the presence of NADPH and molecular oxygen metabolize AA to a series of EETs and HETEs.5 6 7 8
Previous studies have demonstrated that EETs can be produced by microsomal fractions of the hypothalamus and anterior pituitary, slices of mouse brain and homogenates of cat brain,9 10 and cultured astrocytes. In addition, EETs were shown to be endogenous constituents of various areas of the brain.11 Ellis et al12 demonstrated that EETs are potent dilators of cat and rabbit cerebral arterioles in vivo. In addition, we reported that EET-induced dilation of cat cerebral arteries in vitro is attenuated by tetraethylammonium, a potassium channel blocker. We also demonstrated that EETs increase K+ channel activity in enzymatically dispersed cat large cerebral arterial muscle cells.
Despite the potential role played by P450 metabolites of AA in the control of cerebral microcirculation, little is known about the identity of P450 isoforms responsible for the metabolism of AA in the brain. P450 enzymes that metabolize AA to vasoactive metabolites belong to two broad categories: epoxygenases that catalyze the formation of EETs and ω-hydroxylases that produce HETEs.1 In the brain, epoxygenase activity has been detected in astrocytes.10 However, the P450 enzyme responsible for the formation of EETs in astrocytes has yet to be identified. P450 enzymes belong to a large gene superfamily that currently consists of 36 gene families and more than 200 isoforms.13 A complicating factor encountered in the identification of AA-metabolizing P450 isozymes is that many of the drug-metabolizing P450 enzymes are promiscuous and have been shown to catalyze the formation of EETs from AA, including 1A1, 1A2, 2B1, 2B2, 2B4, 2C2, 2C8, 2C9, 2C11, 2C23, and 2E2 isozymes of P450.1 14
Several isoforms of P450 have been identified in the brain. Two alternatively spliced isoforms of P450 2C6 mRNA were found to be expressed and temporally regulated in the brain of the female rat.15 16 17 The regional distribution of the ethanol-inducible P450 2E1–related and the phenytoin-inducible 2B1-related immunoreactivities were mapped in the central nervous system of the rat and mouse, respectively. With the use of RT/PCR, 2D, 2E1, and the P450 reductase were amplified from untreated rat brain, and 1A1, 2B1, and 2B2 were amplified from brains treated with various inducers of P450s.18 With the use of degenerate primers corresponding to conserved regions among members of the 2C subfamily, 2C7, 2C11, and 2C12 were amplified from the brain of female rats, and 2C6, 2C11, 2C12, and 2C23 were amplified from the olfactory lobes of male rats.19 20 21 P450 2D4 was isolated from a cDNA library prepared from the brain of the rat. The immunoreactivities to P450 1A1, 1A2, 2E1, 3A, and microsomal epoxide hydrolase in human brain were mapped with the use of antipeptide-derived antibodies. With the use of Western blot analysis and microsequencing of the N-terminal ends of the purified protein, 2C7, 2C11, 2E1, 4A3, 4A8, and a member of the 2D family were identified in the brain of ethanol-treated rats, while in the brain of control rats, only 2C and 4A were detectable in immunoblot analysis with anti-2A, -3A, -4A, -2C, and -2D antibodies.22
The purpose of this study was to identify and localize a P450 epoxygenase in rat brain astrocytes. The close proximity of astrocytes to cerebral microcirculation makes vasodilatory EETs an attractive paracrine system to couple cerebral blood flow to changes in local metabolic and neuronal activity in the brain. Thus, we identified a P450 epoxygenase expressed in astrocytes and tested the hypothesis that the brain epoxygenase activity originates from astrocytes. In addition, we examined the effect of exogenously applied EETs on whole-cell K+ current in vascular smooth muscle cells isolated from rat cerebral microvessels.
Materials and Methods
Astrocytes were cultured from cerebral cortices and hippocampi of 2- to 3-day-old Sprague-Dawley rat brains under aseptic conditions. Primary cultures were prepared as previously described23 with slight modifications. Briefly, brain tissue was dissected free of meninges, cut into small pieces, and transferred to a sterile dish containing 20 U/mL papain (Worthington Biochemical Corp) and cysteine (Sigma Chemical Co) dissolved in Earle’s balanced salt solution (GIBCO BRL). Tissue pieces were incubated at 37°C for 40 minutes with gentle agitation and then washed three times in the feeding medium, which contained DMEM (GIBCO BRL) with 10% fetal bovine serum (ICN Biomedical) and 1% penicillin-streptomycin solution (Sigma). The tissue was then dissociated by triturating with a flame-narrowed Pasteur pipette. The cell suspension was diluted with feeding medium and seeded into 75-cm2 culture flasks (Costar) at an initial density of 2×105 cells per square centimeter. Cells were incubated at 37°C in a 95%/5% mixture of atmospheric air and CO2. The medium was changed after 2 days and subsequently twice a week. For all experiments reported here we used confluent monolayers (10 to 14 days old) of primary cultures of hippocampal astrocytes. In preliminary experiments, epoxygenase activity and P450 2C11 mRNA expression were compared between cortical and hippocampal astrocytes and between astrocytes isolated from male and female pups. Since no apparent regional or sex differences were detected in epoxygenase activity or 2C11 mRNA expression, subsequent cultures were prepared only from the hippocampus with no attempt made to determine the sex of newborn rats.
Astrocytes were identified by indirect immunocytochemistry with the use of monoclonal anti-GFAP23 mouse IgG (Boehringer Mannheim) as the primary antibody and FITC-conjugated goat anti-mouse IgG (Boehringer Mannheim) as the secondary antibody. Approximately 3×104 cells in 1.5 mL medium were seeded on 22×22-mm coverslips (VWR) in a 35×10-mm Petri dish (Costar) and incubated at 37°C in 95% air/5% CO2 for 24 to 48 hours. Coverslips were then rinsed with DPBS containing Ca2+ and Mg2+ (GIBCO BRL), and cells were fixed in cold methanol for 5 minutes at −20°C. Cells were permeabilized in 0.25% Triton X-100 in DPBS (Sigma) for 2 minutes at room temperature, blocked with 3% normal goat serum in DPBS for 30 minutes at room temperature, and incubated with 100 μL of 5 μg/mL (1:4 dilution in 0.1% normal goat serum) primary antibody for 1 hour at room temperature and, finally, with 100 μL of 2.68 μg/mL (1:500 dilution in 0.5% bovine serum albumin/DPBS) of secondary antibody for 1 hour at room temperature in the dark. Coverslips were then washed with DPBS and mounted on 25×75-mm slides (VWR) with a drop of Gel-Mount (Biomedical Corp).
For double immunostaining experiments, coverslips were simultaneously incubated with both mouse anti-GFAP (Boehringer Mannheim) and rabbit anti-2C11 (raised in our laboratory; see below) primary antisera at 1:100 dilution for 2 hours at room temperature, followed by incubation with 1:40 dilution of the corresponding secondary antibody. An FITC-labeled goat anti-rabbit IgG and a rhodamine-labeled anti-mouse IgG were used as the secondary antibodies. We treated control sections the same way except for eliminating the primary antibody or, in the case of 2C11-immunostaining, substituting it with a preimmune rabbit serum. Slides were visualized with a Nikon Diaphot inverted microscope equipped for epifluorescence. Rhodamine and FITC fluorescent signals emitted from the same field were visualized, imported into a computer, and overlaid on top of each other with the use of Image-1 software (version 4.1, Universal Imaging Corp).
Assay of AA Metabolism
In preliminary experiments, P450 activity was assayed in microsomal extracts from cultured astrocytes according to a protocol published elsewhere and modified to cultured cells.9 Briefly, confluent 10- to 14-day-old primary cultures of astrocytes were washed three times in DMEM and incubated overnight with 0.45 μCi of [14C]AA (57.0 mCi/mmol; DuPont) added to 14 mL of 0.05% fatty acid-free bovine serum albumin in DMEM (0.6 μmol/L final concentration of [14C]AA). The medium was removed, and cells were washed three times with DMEM, scraped off the flask, and homogenized in ice-cold PBS. In some experiments miconazole (20 μmol/L), an inhibitor of P450 epoxygenase,1 was added to the medium, and incubation was continued for 30 minutes before homogenization. The lipid component of the cell homogenate was extracted twice with 3 mL of ethyl acetate, back-extracted with 1 mL of distilled water, evaporated to dryness under nitrogen, and reconstituted in 0.5 mL of ethanol. AA products were separated with the use of a 2.1-mm×25-cm C18 reverse-phase HPLC column (Supelco LC-18) and a linear solvent gradient ranging from 30 acetonitrile/70 water/1 acetic acid (vol/vol/vol) to 100 acetonitrile/1 acetic acid (vol/vol) over 50 minutes at a flow rate of 0.5 mL/min. AA metabolites were monitored with an on-line radioactive flow detector (Beckman System Gold, model 171).
Confluent monolayers (10 to 14 days old) of primary cultures of astrocytes were used to extract total RNA with TRIzol reagent (GIBCO BRL). Culture medium was discarded, and cells were washed three times with ice-cold DPBS (GIBCO BRL). Cells were lysed in the culture dish by the addition of TRIzol reagent, and the cell lysate was passed several times through a syringe needle for further homogenization. Chloroform was used for phase separation, and RNA was precipitated from the aqueous phase by isopropanol. The RNA pellet was washed with 75% ethanol and redissolved in RNase-free water. Total RNA (5 μg) was reverse transcribed with the use of cloned murine reverse transcriptase (First-Strand cDNA Synthesis Kit, Pharmacia) with 1 mmol/L oligo(dT)18 primer. The reaction was incubated at 37°C for 1 hour and terminated by heating to 100°C for 5 minutes. Sense and antisense primers (1 μmol/L) derived from hepatic P450 2C11 oligonucleotide sequence24 were added to the RT reaction mix for PCR amplification of cDNA with the use of Taq DNA polymerase (2.5 U) (Pharmacia). Full-length P450 2C11 cDNA was amplified with the use of the following primers: 5′-CATGGATCCAGTCCTAGTCC-3′ as the sense, and 5′-TCCCTTGTCCTTATCTTGAT-3′ as the antisense primer. Primers were annealed at melting temperatures [Tm=4°C (G/C)+2°C (A/T)] and PCRs allowed to proceed for 30 cycles. For each RT/PCR, two control reactions were performed simultaneously: (1) PCR of untranscribed RNA, to exclude amplification of contaminating genomic DNA, and (2) PCR negative control with no DNA, to exclude contamination of reagents.
Cloning and Sequencing
PCR products were purified from a 1% low-melting agarose gel (FMC) with the use of a QIA quick gel extraction kit (Qiagen) and ligated into a linearized pCR II plasmid vector (TA Cloning Kit, Invitrogen) with T4 DNA ligase. Competent Escherichia coli cells (One Shot cells, Invitrogen) were transformed by electroporation (Bio-Rad), and cells containing the recombinant plasmid were selected for by growing on X-Gal/isopropylthiogalactose agar plates containing ampicillin (50 μg/mL). β-Galactosidase-negative colonies were color-identified and screened by PCR with the use of universal M13 forward and reverse primers, which flank the multiple cloning site in pCR II. Restriction digest of mini-prep DNA (QIAprep, Qiagen) from PCR-positive colonies was performed to confirm the presence of an insert.
Cloned cDNA was sequenced from the double-stranded plasmid by the dideoxynucleotide method25 with the use of Sequenase DNA polymerase version 2.0 (USB) with both vector and insert primers. Briefly, 5 μg of double-stranded plasmid DNA purified from a large-scale overnight bacterial culture with the use of Qiagen-tip 2500 (Qiagen) was alkaline-denatured by NaOH and salt-precipitated with ethanol. The sequencing primer was annealed to the denatured template, and the sequencing reaction was allowed to proceed for 5 minutes at room temperature in the presence of Sequenase (USB) and 10 μCi/μL [α-35S]dATP (DuPont). The reaction was allowed to proceed further at 37°C for another 15 minutes in the presence of dideoxynucleotides.
Sequencing reactions were heat-denatured, loaded on a denaturing 6% polyacrylamide/urea gel (Sequagel, National Diagnostics), and electrophoresed (S2 Electrophoresis apparatus, GIBCO) for 3 hours at 80 W. Gels were then dried and exposed to x-ray films (Eastman Kodak Co) for 18 hours at room temperature. Sequences obtained from a single set of reactions averaged between 300 and 400 base pairs in length. Sequences were obtained progressively by the use of nested primer sites from the already determined sequences. Obtained sequences were deduced from three separate clones and both strands of the insert. Sequences were read and recorded at least twice by two different people and compared with published sequences in GenBank (National Center for Biotechnology Information, National Library of Medicine, National Institutes of Health) with the use of BLAST software (National Center for Biotechnology Information).
Southern Blot Analysis
To verify the identity of appropriate size RT/PCR products, a Southern blot hybridization was performed with a 2C11-specific oligonucleotide probe. RT/PCR products were resolved on a 1% DNA agarose gel (GIBCO BRL) together with a HindIII digest of λ-phage DNA (Pharmacia), which served as a molecular weight marker and as a negative control for hybridization. The agarose gel was depurinated in 0.25 mol/L HCl solution, alkaline-denatured, and neutralized before transfer to a Nytran Plus membrane (Schleicher & Schuell) with the use of a vacuum blotter (Bio-Rad). After transfer, the membrane was air-dried and DNA was immobilized by UV cross-linking (Stratalinker, Stratagene). The membrane was prehybridized for 3 hours in 6× SSPE/1% SDS/10× Denhardt’s solution25 with 20 μg/mL tRNA (Sigma) and 50 μg/mL denatured heterologous DNA (salmon sperm DNA, Sigma) at 42°C. Membranes were hybridized overnight at TH=Tm−5°C in 6× SSPE/1% SDS with a radiolabeled 2C11-specific oligonucleotide probe having the following sequence: 5′-GACCTTGTCCCCACAAAC-3′. The oligonucleotide was end-labeled with [γ-32P]ATP (DuPont) with the use of T4 polynucleotide kinase (Promega). Blots were washed in 6× SSPE/1% SDS three times for 5 to 10 minutes each. The final wash was performed in 1× SSPE/1% SDS for 1 to 3 minutes at the appropriate TH before exposure to an x-ray film (Kodak) for 6 to 8 hours at −80°C.
P450 2C11 antipeptide antibody was raised in rabbits by immunizing the animals with the following synthetic peptides derived from both the amino and carboxy terminals of P450 2C11, respectively: DIGQSIKKFSKV and QRADSLSSHL. Sequences were designed with the use of GCG software (Genetic Computer Group). Peptides were synthesized and HPLC purified in the Nucleic Acid Facility at the Medical College of Wisconsin. Amino acid composition of the peptides was confirmed with mass spectrometry and amino acid analysis. Peptides were conjugated with Ovalbumin (Pierce Chemical Co) with the use of 1-ethyl-3-(3-dimethylaminopropyl)-carbodiimide hydrochloride as the cross-linking reagent. Conjugation was performed at room temperature for 2 hours at a peptide-protein ratio of 1:1. Conjugated peptides were column-purified and identified in 0.8 mL fractions by spectrophotometry at a wavelength of 280 nm; 125 μg of each peptide was injected subcutaneously in 500 μL 0.9% NaCl together with 2 mL Freund’s adjuvant (GIBCO BRL). Before injections, both animals were bled for collection of preimmune sera. After 4 weeks, a booster injection was given the same way. Blood was withdrawn 7 days later from the marginal ear vein and allowed to clot at 37°C for 2 hours. Clotted whole blood was centrifuged and serum was decanted and stored at −80°C. Antibody titers were checked by dot blot assay with the use of the free peptide. When titers were low, animals were given another boosting injection. Antibody specificity was verified by Western blot analysis of male and female rat liver microsomes26 (also see “Discussion”) and by a comparison of the immunoreactivity of the conjugated peptides with that of the carrier proteins (Ovalbumin and bovine serum albumin).
Western Blot Analysis
Confluent monolayers (10 to 14) of primary cultures of astrocytes were washed three times with DPBS and scraped for preparation of microsomes by differential centrifugation as previously described.5 Briefly, cells were washed three times with ice-cold DPBS, scraped, and homogenized in a 10 mmol/L potassium phosphate buffer (pH 7.7) containing 250 mmol/L sucrose, 1 mmol/L EDTA, and 10 mmol/L MgCl2. The homogenate was centrifuged for 10 minutes at 3500g, 30 minutes at 9000g, and 60 minutes at 100 000g to obtain the microsomal pellet. Microsomal proteins (20 μg) were separated on a 7.5% SDS–polyacrylamide gel and transferred electrophoretically to a 0.2-μm supported nitrocellulose membrane (Bio-Rad) in a transfer buffer (25 mmol/L Tris, 192 mmol/L glycine, pH 8.3) containing 20% vol/vol methanol at room temperature for 1 hour at 100 V. In some experiments, 20 μg protein from the supernatant of the final centrifugation step was also analyzed for P450 2C11 immunoreactivity. The membrane was blocked in a TBS-T buffer (50 mmol/L Tris, 0.2 mol/L NaCl, 0.08% Tween-20, pH 7.5) containing 5% nonfat dry milk at 4°C overnight. The membrane was then incubated with P450 2C11 antiserum (1:3000 dilution in TBS-T containing 2% nonfat dry milk) for 2 hours at room temperature. The membrane was washed three times with TBS-T and incubated with the secondary antibody. Horseradish peroxidase–conjugated goat anti-rabbit IgG (Bio-Rad) was used as the secondary antibody. Incubations were performed at room temperature for 30 minutes at a 1:1000 dilution of the secondary antibody in TBS-T containing 2% nonfat dry milk. Avidin–horseradish peroxidase IgG (Bio-Rad) was included at 1:3000 dilution to visualize biotinylated marker proteins. Membranes were washed three times with TBS-T and detected by enhanced chemiluminescence (Amersham).
Isolation of Vascular Muscle Cells From Cerebral Microvessels
Cerebral microvessels were isolated according to a previously published protocol.5 Briefly, adult (10 to 12 weeks) Sprague-Dawley rats (Harlan) were anesthetized with sodium pentobarbital (65 mg/kg body wt IP; Anpro Pharmaceutical). The cerebral microcirculation was filled with a 5% suspension of iron oxide (particle size=10 μm; Aldrich Chemical Co) in saline via the internal carotid arteries. The brain was then removed and passed through a series of wire-mesh sieves. Tissue mixture retained on top of the screen was digested at 37°C for 20 minutes in an enzyme solution containing 16.6 U/mL collagenase type II (Worthington Biochemical), 4 mmol/L dithiothreitol (Sigma), and 1 mmol/L soybean trypsin inhibitor (Sigma) in a low-Ca2+ PSS (composition described below). The tissue mixture was repeatedly suspended in ice-cold PSS and agitated against a magnetic plate. Cerebral arterioles that were filled with iron oxide migrated to the magnetic plate and were visualized under a dissecting microscope.
Isolated arterioles were minced and placed in a low-Ca2+ PSS containing (mmol/L): NaCl 134, KCl 5.4, MgSO4 1.2, KH2PO4 0.24, CaCl2 0.05, glucose 11, and HEPES 10 (pH adjusted to 7.4 with NaOH). Vessel fragments were transferred to a vial containing 88.5 U/mL collagenase type II (Worthington), 2 mmol/L dithiothreitol (Sigma), and 1 mmol/L trypsin inhibitor (Sigma) in low-Ca2+ PSS. The vial was placed in a water-jacketed beaker on a microstirrer, and the tissue was stirred (12 rpm) at 37°C for a total of 1 hour in the enzyme solution. In 10-minute intervals, the supernatant was replaced by a fresh enzyme solution and checked for the appearance of dispersed cells. Pieces of vessel were disrupted mechanically by forcing them repeatedly through a Pasteur pipette. Cell suspension was transferred to a test tube containing PSS and placed on ice. Patch-clamp experiments were initiated immediately.
Patch-Clamp Analysis of Whole-Cell K+ Currents
Dispersed vascular smooth muscle cells were placed in a perfusion chamber on the stage of an inverted microscope (Olympus IMT-2). Patch pipettes with tip resistance of 3 to 10 MΩ were fabricated from borosilicate glass pulled on a two-stage micropipette puller (PC-84) and heat-polished under a microscope (MF-83 heat polisher, Narishige). At ×500, a three-way hydraulic micromanipulator (Narishige) was used to position patch pipettes on the membranes of single vascular smooth muscle cells. High-resistance seals (>1 GΩ) were attained between patch pipettes and cell membranes. Suction was applied until there was a large increase in capacitive current, indicating rupture of membrane patches. Membrane potential was clamped and currents measured with a List EPC-7 patch-clamp amplifier. The amplifier output was low-pass filtered at 1 kHz with an eight-pole Bessel filter (Frequency Devices). Current signals were digitized at a sampling rate of 3 kHz and stored in a computer hard disk (Citus 386SX). Pulse protocols were controlled and data analyzed with the use of pClamp software (pClamp version 5.5, Axon Instruments). Pipette solution contained (in mmol/L): KCl 75, potassium glutamate 70, MgCl2 1, CaCl2 1.8, EGTA 5, dipotassium ATP 2, HEPES 10, with a final pH adjusted to 7.2 with KOH. The external solution for whole-cell recording contained (in mmol/L): NaCl 140, KCl 4.7, MgCl2 1.1, CaCl2 1.8, glucose 11, HEPES 10, with pH adjusted to 7.4 with NaOH. In some experiments, tetraethylammonium (1 mmol/L, Sigma) was added to the external solution to determine its effect on the macroscopic K+ current. The external solution was continuously superfused over the cells by gravitational flow.
Formation of EETs by Cultured Astrocytes
Primary cultures of GFAP-positive rat brain astrocytes were labeled with [14C]AA and assayed for their ability to convert AA to EETs. As can be seen in the representative reverse-phase HPLC chromatogram in Fig 1⇓, intact astrocytes cultured from neonatal rat brain converted AA predominantly to products that have retention times identical to authentic [14C]EETs standards. Based on elution time, the major isomers of EETs produced by astrocytes were 11,12- and 14,15-EETs (Fig 1A⇓ and 1C⇓). Preincubation of astrocytes with 20 μmol/L miconazole for 30 minutes blocked the appearance of these products (Fig 1B⇓). On the basis of specific activity of AA, the concentration of EETs produced by astrocytes was estimated to be 2.2±1.1 nmol/L (n=5).
Cloning and Sequencing of P450 2C11 cDNA From Cultured Astrocytes
RT/PCR of mRNA from cultured astrocytes with oligonucleotide primers flanking the full-length coding cDNA of P450 2C11 resulted in the amplification of a 1.7-kb band (Fig 2A⇓). An internal 2C11-specific oligonucleotide probe hybridized with this band in a Southern blot analysis of RT/PCR products (Fig 2B⇓). The band was purified from an agarose gel and cloned into a plasmid vector. Sequencing RT/PCR clones from cultured astrocytes revealed a 100% identity to the sequence of hepatic P450 2C11 cDNA. Fig 3⇓ illustrates the sequencing strategy and the locations of PCR and sequencing primers.
Production of the Antibody and Western Blot of Cultured Astrocytes
A polyclonal antipeptide antibody against P450 2C11 was raised in rabbits. The specificity of the antiserum to P450 2C11 was verified by Western blot analysis of male and female rat liver microsomes (see “Discussion”). Fig 4A⇓ demonstrates reactivity of the antiserum with a 55-kD protein band in male but not female rat liver microsomes. P450 2C11 is a 55-kD protein that is expressed in male but not in female rat liver.26 Microsomal proteins extracted from primary cultures of astrocytes were probed with 2C11 antibody. Fig 4B⇓ represents a Western blot of cultured astrocytes in which 2C11 antiserum is used, demonstrating reactivity of 2C11 antibody with a 55-kD band in the microsomal fraction of cultured astrocytes (Fig 4B⇓, lane 2). Lane 3 in Fig 4B⇓ demonstrates reactivity of 2C11 antibody with a similar band in the cytosolic fraction (see “Discussion”). The polyclonal antiserum also reacts with another 75-kD protein band in astrocytes. This 75-kD band is present in all tissues when polyclonal antibodies to a variety of P450 isoforms are used and possibly represents P450 reductase.
Colocalization of GFAP and 2C11 Immunoreactivities in Astrocytes
Cultured astrocytes were stained for both GFAP, an astrocyte-specific protein, and P450 2C11. Fig 5⇓ represents a double immunostaining of cultured astrocytes, demonstrating colocalization of GFAP and 2C11 immunoreactivities in astrocytes. Control stainings of cultured astrocytes with rabbit preimmune serum were negative for 2C11-like immunoreactivity.
Effect of 14,15-EET on Macroscopic K+ Current in Rat Cerebral Microvascular Muscle Cells
Vascular smooth muscle cells were isolated from rat cerebral microvessels, and whole-cell K+ current was recorded with the use of the patch-clamp technique. Depolarizing pulses between −70 and +80 mV elicited a voltage-dependent outward K+ current that was almost completely blocked by 1 mmol/L tetraethylammonium, a potassium channel blocker (data not shown). External application of 100 nmol/L 14,15-EET to the bath solution enhanced whole-cell K+ current, as demonstrated by the upward shift in the current-voltage relationship (Fig 6⇓). The effect was especially prominent over the positive range of membrane potentials where basal K+ current is high.
The purpose of the present study was to identify a P450 AA epoxygenase in astrocytes. We (1) found that a P450 2C11 mRNA is expressed in rat brain astrocytes, (2) provided further support for previous reports indicating that primary cultures of astrocytes metabolize AA to EETs, and (3) showed that EETs can have a direct action on rat cerebral microvascular smooth muscle by enhancing outward K+ current.
Previous reports have demonstrated that EETs are endogenous constituents of the brain11 and that preparations of the brain can metabolize AA to EETs.6 7 8 We have also reported that cortical microsomes from cat brain convert AA to metabolites with reverse-phase HPLC retention times identical to 8,9-, 11,12-, and 14,15-EETs.9 In the present study we show that intact cultured astrocytes from rats can convert radiolabeled AA to metabolites having retention times identical to authentic EET standards. Based on the elution time, the major isomers of EETs produced by astrocytes appear to be 11,12- and 14,15-EETs. This is in agreement with the findings of Amruthesh et al,10 who demonstrated that 14,15-EET was the major P450 epoxide of AA produced by astrocytes. Further proof for the identity of these compounds is gained by the fact that miconazole, an inhibitor of P450 epoxygenase,1 blocks the appearance of the peak tentatively identified as EETs.
The biological activity of the major regioisomers of EETs, namely, 5,6-, 8,9-, 11,12-, and 14,15-EET, is one of dilation of precontracted or pressurized arteries in vitro and superfused cranial vessels in vivo.9 12 The sensitivity of EETs tends to increase as smaller and smaller vessels are involved, with an initial sensitivity of <1×10−9 mol/L in renal arterioles.27 The fact that astrocytic foot processes are in close apposition to cerebral microvessels makes the release of a vasoactive metabolite from astrocytes a potentially important factor in controlling nutritive cerebral blood flow.
In the present study we assayed for cell-associated metabolites of AA. However, in a subsequent study the culture medium was assayed for the presence of AA metabolites. Our preliminary findings (N.J.A. et al, unpublished data) indicate that EETs are released by astrocytes into the extracellular medium. A previous report10 has shown that astrocytes contain epoxide hydrolase activity, which metabolizes EETs to their vicinal diols, DHETs. In a subsequent study by the same research group,28 the authors reported that the metabolism of 14,15-EET to 14,15-DHET is a protein kinase C-regulated process. Under the more physiological assay conditions used in the present study, in which the metabolism of AA is assayed for in intact rather than homogenized astrocytes, we predominantly detect EETs. Infrequently, we detect smaller peaks corresponding to the elution time characteristic of DHETs (eg, the small peaks between 18 and 20 minutes in Fig 1A⇑). The discrepancy between our findings and those of Amruthesh et al10 28 may be due to differences in the assay conditions (ie, the use of intact versus homogenized cells). It is possible that the epoxide hydrolase has a better access to EETs in homogenized compared with intact cells. Another possibility is that the activity of the epoxide hydrolase in intact cells is regulated so that fewer DHETs are formed.
Of particular interest is the demonstration of the ability of astrocytes in culture to form EETs at a substrate concentration comparable to that found in the normal brain (0.6 μmol/L of [14C]AA).8 This finding suggests that brain astrocytes may synthesize EETs endogenously. In the present study we also show that astrocytes in culture produce EETs at an estimated concentration of 2.2±1.1 nmol/L, a concentration of EETs that exceeds the 1 nmol/L threshold we have previously shown to be needed to produce a significant dilation in small renal arteries of the rat.27 Given that some of the labeled products are lost during the processes of extraction and separation, the actual concentration of EETs produced by astrocytes may be higher. Moreover, assuming that EETs are produced in vivo at the same rate, their local effective concentration may be higher due to the close proximity of astrocytic end-feet to cerebral microvessels. These findings are in agreement with the findings of Amruthesh et al,8 who have demonstrated formation of EETs by mouse brain slices at a substrate concentration of 5 μmol/L. The latter authors have estimated the concentration of the bioactive product in a peak identified as 5,6-EET to be 2.3 μg/mL of cerebrospinal fluid. The authors have also shown that the isolated peak dilated rabbit pial arterioles in vivo.
These findings suggest that endogenous formation of EETs may be sufficient to affect the brain microvasculature. Recently, we examined the contribution of brain P450 AA epoxygenase in the control of blood flow in the cerebral microcirculation of the rat in vivo (N.J.A. et al, unpublished data). In the latter study, administration of miconazole, an inhibitor of P450 epoxygenase,1 into the subdural space of the rat for 30 minutes at a dose sufficient to inhibit formation of EETs by cultured astrocytes decreased baseline laser-Doppler cerebral blood flow by 29.7±7.3%. Furthermore, in a more recent study (N.J.A. et al, unpublished data), administration of miconazole into the subdural space of the rat and intracerebroventricular infusion of P450 2C11 antisense oligodeoxynuculeotide attenuated the response of cerebral blood flow to glutamate. These findings suggest that brain P450 AA epoxygenase may contribute importantly to the regulation of cerebral microcirculation in the rat.
Despite evidence for endogenous formation of P450 eicosanoids and the ability of EETs to regulate cerebrovascular tone,1 little is known regarding specific P450 isoforms responsible for the formation of EETs in cerebral tissue. We used RT/PCR to detect the expression of a P450 AA epoxygenase in astrocytes of rat brain. The identity of the amplified product was verified by a Southern blot analysis of RT/PCR products with the use of an internal P450 2C11–specific oligonucleotide probe. The probe strongly hybridized to a 1.7-kb band. This cDNA band was purified, cloned, and sequenced. The sequence of the cloned cDNA was 100% homologous to a P450 2C11 full-length coding cDNA previously isolated from rat liver.24 A weaker hybridization was observed with a smaller size band on the Southern blot. We have not pursued the nature of this band. Since the sequence of the oligonucleotide probe was unique for P450 2C11, under the stringent conditions used in the Southern blot analysis of RT/PCR products, a cross-hybridization with a related P450 isoform is unlikely. Consequently, the second band could be either an as yet unidentified P450 isoform that shares homology with P450 2C11 over the region containing the sequence of the probe, or an alternatively spliced isoform of P450 2C11 that retains the exon containing the sequence of the probe.
P450 2C11 is a male-specific steroid hydroxylase in rat liver.24 26 Capdevilla et al29 have also demonstrated AA epoxygenase activity associated with a P450 2C11 enzyme isolated from the liver of rats. Our present study provides the first demonstration that brain P450 2C11 is expressed in astrocytes. Previous reports in the literature have demonstrated that P450 2C11 is expressed in intact whole brain. Using 2C11-specific primers in RT/PCR, Zaphiropoulos and Wood19 detected expression of P450 2C11 mRNA in both male and female rat brains. Furthermore, Warner and Gustafsson22 demonstrated that P450 2C11 protein is expressed in intact brain of the rat by Western blot analysis using a 2C-specific antibody and by microsequencing of the N-terminal ends of the isolated protein.
To demonstrate expression of P450 2C11 protein in astrocytes, we used a polyclonal antibody raised against synthetic peptide sequences in the P450 2C11 enzyme. The specificity of the antibody to P450 2C11 protein was determined by size migration and also by cross-reacting with the enzyme, which is expressed only in the liver of male rats.26 In this regard, the astrocyte cultures used in this study were prepared from both male and female rat pups with no apparent difference in P450 2C11 gene expression (see “Materials and Methods” and also refer to Zaphiropoulos and Wood19 ), suggesting that astrocytic 2C11 promoter or promoter elements regulate the expression of P450 2C11 gene differently in the liver of the same animal. Western blot analysis of cultured astrocytes with the use of a 2C11 antibody reveals that a P450 2C11–immunoreactive protein is expressed in the microsomal fraction of cultured astrocytes. The presence of a 2C11-immunoreactive band in the cytosolic fraction may be due to contaminating microsomal protein. Alternatively, this finding may suggest that a P450 2C11– or a 2C11-immunoreactive protein is expressed in the cytosol of cultured astrocytes. Double immunostaining of cultured astrocytes with antibodies to GFAP and P450 2C11 further demonstrates the expression of P450 2C11 protein in astrocytes.
At this point, we cannot say that 2C11 is the only P450 epoxygenase gene present in astrocytes or that it is even the major epoxygenase responsible for formation of EETs in these cells. However, in preliminary experiments we used degenerate PCR primers designed to amplify any member of the 2C family of P450 genes under low stringency conditions, and only one band was amplified. Future work is needed to determine whether any other P450 epoxygenase isoforms are expressed in astrocytes given recent demonstrations regarding the action of EETs as effectors of cerebrovascular tone.9 12
We have previously demonstrated that EETs dilate large cerebral arteries from cats by increasing the activity of the large-conductance Ca2+-activated K+ channel.9 In the present study external application of 14,15-EET enhanced outward K+ current in freshly dispersed muscle cells from rat cerebral arterioles. The action of 14,15-EET to enhance whole-cell K+ current in cerebral microvascular muscle cells demonstrates its ability to modify cerebral vascular function in the rat.
In summary, we used a functional assay to demonstrate the presence of an intrinsic AA epoxygenase in intact astrocytes. We also used RT/PCR to identify a P450 AA epoxygenase gene expressed in astrocytes. Having identified the expression of P450 2C11 mRNA in astrocytes, we used an immunologic approach to localize P450 2C11 protein in cultured astrocytes. Finally, we demonstrated enhancement of macroscopic K+ current in cerebral microvascular muscle cells by 14,15-EET. Our results suggest that P450 2C11 expressed in perivascular astrocytes may be responsible for brain epoxygenase activity. Given the vasodilatory effect of EETs on cerebral microvessels, these findings suggest a role for astrocytes and P450 2C11 in the control of cerebral microcirculation. The vasodilatory effect of EETs appears to involve activation of K+ current in cerebrovascular smooth muscle cell membranes. This effect may lead to membrane hyperpolarization and vascular smooth muscle cell relaxation and may be responsible for EET-induced dilation of cerebral microvessels.
Selected Abbreviations and Acronyms
|DMEM||=||Dulbecco’s modified Eagle’s medium|
|DPBS||=||Dulbecco’s phosphate-buffered saline|
|GFAP||=||glial fibrillary acidic protein|
|HPLC||=||high-performance liquid chromatography|
|PSS||=||physiological salt solution|
|RT/PCR||=||reverse transcription/polymerase chain reaction|
|SDS||=||sodium dodecyl sulfate|
- Received September 8, 1995.
- Revision received December 13, 1995.
- Accepted December 21, 1995.
- Copyright © 1996 by American Heart Association
Bazan NG, Allan G, Rodriguez de Turco EB. Role of phospholipase A2 and membrane-derived lipid second messengers in membrane function and transcriptional activation of genes: implications in cerebral ischemia and neuronal excitability. In: Kogure K, Hossmann K-A, Siesjö BK, eds. Progress in Brain Research. New York, NY: Elsevier Science Publishers BV; 1993;96:247-257.
Harder DR, Gebremedhin D, Narayanan J, Jefcoat C, Falck JR, Roman RJ. Formation and action of a P-450 4A metabolite of arachidonic acid in cat cerebral microvessels. Am J Physiol. 1994;266(Heart Circ Physiol 35):H2098-H2107.
Gebremedhin D, Ma Y-H, Falck JR, Roman RJ, Van Rollins M, Harder DR. Mechanism of action of cerebral epoxyeicosatrienoic acids on cerebral arterial smooth muscle. Am J Physiol. 1992;263:H519-H525.
Ellis EF, Police RJ, Yancey L, McKinney JS, Amruthesh SC. Dilation of cerebral arterioles by cytochrome P-450 metabolites of arachidonic acid. Am J Physiol. 1990;259(Heart Circ Physiol 28):H1171-H1177.
Daikh BE, Lasker JM, Raucy JL, Koop DR. Regio- and stereoselective epoxidation of arachidonic acid by human cytochromes 2C8 and 2C9. J Pharmacol Exp Ther.. 1994;271:1427-1433.
Kimura H, Sogawa K, Sakai Y, Fuji-Kuriyama Y. Alternative splicing mechanism in a cytochrome P-450 (P-450PB-1) gene generates the two mRNAs coding for proteins of different functions. J Biol Chem. 1989;264:2338-2342.
Farin FM, Omiecinski CJ. Regiospecific expression of cytochrome P-450s and microsomal epoxide hydrolase in human brain tissue. J Toxicol Environ Health. 1993;49:317-335.
Warner M, Gustafsson J-Å. Effect of ethanol on cytochrome P450 in the rat brain. Proc Natl Acad Sci U S A. 1994;91:1019-1023.
Hansson E, Ronnback L. Primary cultures of astroglia and neurons from different brain regions. In: Shahar A, de Vellis J, Vernadakis A, Haber B, eds. A Dissection and Tissue Culture Manual of the Nervous System. New York, NY: Alan R. Liss Inc; 1989:92-104.
Yoshioka H, Morohashi K, Sogawa K, Miyata T, Kawajiri K, Hirose T, Inayama S, Fuji-Kuriyama Y, Omura T. Structural analysis and specific expression of microsomal cytochrome P-450 (M-1) mRNA in male rat livers. J Biol Chem. 1987;262:1706-1711.
Sambrook J, Fritsch EF, Maniatis T. Molecular Cloning: A Laboratory Manual. 2nd ed. New York, NY: Cold Spring Harbor Laboratory Press; 1989.
Sundseth SS, Alberta JA, Waxman DJ. Sex-specific, growth hormone-regulated transcription of the cytochrome P450 2C11 and 2C12 genes. J Biol Chem. 1992;267:3907-3914.
Zou A-P, Fleming JT, Falck JR, Jacobs ER, Gebremedhin D, Harder DR, Roman RJ. 20-HGTG is an endogenous inhibitor of the large-conductance Ca2+–activated K+ channel in renal arterioles. Am J Physiol.. 1996;270:R228-R237.
Capdevila JH, Karara A, Waxman DJ, Martin MV, Falck JR, Guenguerich FP. Cytochrome P450 enzyme-specific control of the regio- and enantiofacial selectivity of the microsomal arachidonic acid epoxygenase. J Biol Chem. 1990;265:10865-10871.