Molecular Cloning, Expression, and Enzymatic ... - Semantic Scholar

5 downloads 0 Views 3MB Size Report
metabolism of arachidonic acid to a mixture of 5,6-,. 8,9-, 11,12-, and 14,15-epoxyeicosatrienoic acids as the only oxygenation products. The enantiofacial se-.
Vol. 268, No.18, Issue of June 25,pp. 1356.-13570,1993 Printed in U.S. A .

THEJOURNAL OF BIOLOGICAL CHEMISTRY Q 1993 by The American Society for Biochemistry and Molecular Biology, Inc.

Molecular Cloning, Expression, and Enzymatic Characterizationof the Rat Kidney CytochromeP-450 Arachidonic AcidEpoxygenase” (Received for publication, January 5, 1993, and in revised form, March 8, 1993)

Armando KararaS, Keiko Makita$, Harry R. Jacobson$, John Falckt, R. F. Peter Guengerichll, Raymond N. DuBois$ 11, and Jorge H. Capdevila$ll** From the Departments of $Medicine, llBiochemistry, and )ICell Biology, Vanderbilt University Medical School, Nashville, Tennessee 37232 and the §Department of Molecular Genetics, Southwestern Medical Center, Dallas, Teras75235

Studies utilizing purified P-450 isoforms demonstrated that A cDNA containing an open reading frame coding for the rat kidney cytochrome P-450 arachidonic acid the epoxygenase is highly stereospecific and that the asymepoxygenase was isolatedfromamaleratkidney metry of oxygen addition is P-450 isoform-specific (5). To cDNA library. Sequenceanalysis showed thatwith the date, arachidonic acid epoxygenase activity has been demonexception of11 nucleotides,this cDNA is identical with strated in numerous tissues, including kidney, liver, brain, the published sequence for ratliver cytochrome 2C23 pituitary,adrenal,and endothelium (6, 7and references and encodes a polypeptide 494 of amino acids. Nucleic therein). Thedemonstration of a role for P-450 in the in uiuo acidblothybridizationindicatedthatthe levels of metabolism of endogenous arachidonic acid established the expression of the corresponding mRNA are high in rat epoxygenase as amember of the arachidonic acid cascade and kidney and liver and are undetectable in brain and documented a novel endogenous role for this hemoprotein in heart. the generation of bioactive eicosanoids (6, 7 and references The cDNA coding regionwas cloned into a pCMV2 vector and expressed inCOS-1 cells. The recombinant therein). The potentbiological activities associated with several memicrosomal protein catalyzed the NADPH-dependent tabolites of the arachidonic acid epoxygenase havestimulated metabolism of arachidonic acid to a mixture of 5,6-, 8,9-, 11,12-, and 14,15-epoxyeicosatrienoicacids as interest in the role of P-450 in the control of cell and organ the only oxygenation products. The enantiofacial se- physiology (6,7 and references therein). Among the biological lectivity of the recombinant protein was nearly iden- activities attributed to the EETs or to their hydration products, the DHETs, those of significance to renal physiology tical with thatreportedforthekidneymicrosomal enzymeandgenerated 8(R),9(S)-, ll(R),l2(S)-,and include ( a ) changes in the permeability of the rabbit isolated 14(S),lS(R)with optical purities of 95, 85, and 75%, cortical collecting tubule to Na’ and K+ (8) and in the respectively. On the basis of mRNA abundance and thetransport of Na+ by the proximal tubule (9); ( b ) inhibition of close similarities between the regio- and stereochemi- vasopressin-stimulated water reabsorption (10, 11); and (c) cal selectivity of the recombinant and kidney microenantioselective renal vasoconstriction (12). Studies with the somal proteins, we concluded that cytochrome P-450 spontaneous hypertensive rat model indicated that P-4502C23 is the predominant enzyme isoform responsible derived arachidonic acid metabolites may play a role in the for arachidonic acid epoxidation in the rat kidney. pathophysiology of hypertension (6). This view has recently been strengthened by ( a ) the demonstration, in humans, of increased urinary excretion of DHETsduring pregnancyThe physiological significance of arachidonic acid as the induced hypertension (13); ( b ) the preferential expression of precursor for a variety of oxygenated metabolites is well the P-450 4A2 gene in spontaneous hypertensive rats (14); documented (Ref. 1 and references therein).Thus,it is and (c) the regulation of the renal epoxygenase by dietary salt through the action of regio- and stereospecific oxygenases loading (15). Although there have been extensive biochemical and structhat cellular mediators such as prostanoids and leukotrienes are formed (1and references therein). In 1981, several groups tural studies of the epoxygenase metabolites, little is as yet demonstrated a role for microsomal cytochrome P-450 in the known with regard to the molecular properties of the renal in uitro catalysis of arachidonic acid metabolism (2-4). The cytochrome P-450 proteinisoform(s) active in arachidonic P-4501 arachidonic acid epoxygenase catalyzes the NADPH- acid epoxidation. A member of the P-450 2C gene family was dependent formation of 5,6-, 8,9-, llJ2-, and 14,15-EETs. recently purified from rabbit kidney and shown to catalyze arachidonic acid epoxidation and, to a lesser extent, w / w - l * This work wassupported by United States Public Health Service oxidation (16). Inhibition studies utilizing a panel of polyGrants NIHDK 38226 (to J. R. F., H. R. J., and J. H. C.), NIHCA clonal antibodies to several rat liver P-450 isoforms indicated 44353 (to F. P. G.), NIHES 00267 (to R. N. D., J. H. C., and F. P. that the rat kidney epoxygenase belongs to the hemoprotein G.), and by Dialysis Clinics, Inc. (to J. H. C. and H. R. J.). The costs of publication of this article were defrayed in part by the payment of 2C gene family (15). Thus, it became apparent that further page charges. This article must therefore be hereby marked “aduer- progress in delineatingthe functional role of the epoxygenase tisement” in accordance with 18 U.S.C. Section 1734 solely to indicate enzyme system requires, on the one hand, a detailed knowlthis fact. edge of the molecular properties of the enzyme and, on the ** To whom correspondence should be addressed Medical Center otherhand, access to biospecific probes for studies of its North S-3223, Vanderbilt University Medical School, Nashville, T N regulation at the gene and/or protein level. We report here 37232. The abbreviations used are: P-450, cytochrome P-450; EETs, cis- the molecular cloning of the rat kidney P-450 epoxygenase; epoxyeicosatrienoic acids; DHETs, uic-dihydroxyeicosatrienoicacids; demonstrate that its nucleic acid sequence is nearly identical kb, kilobase; HPLC, high pressure liquid chromatography. to that reported for P-450 2C23 (17); and establish unequiv-

13565

13566

Molecular Cloning

Renal of Arachidonic a Acid Epoxygenase

ocally that a single recombinant protein can catalyze, albeit with different efficiencies, the highly asymmetric epoxidation of the arachidonic acid 8,9-, 11,12-, and 14,15-0lefins. MATERIALS ANDMETHODS

Isolation of Rat Kidney Poly(A)+ RNA, Synthesis, and Screening of the cDNA Libraly-Whole kidneys were removed from anesthetized (Nembutal, 50 mg/kg body weight) adult male Sprague-Dawley rats (300-350 g body weight), immediately frozen in liquid NP,and stored at -80 'C. Within the next 48 h, poly(A)+ RNA was prepared by the guanidinium isothiocyanate/oligo(dT)-Sepharose method using an mRNA extraction kit from Pharmacia LKB Biotechnology Inc. An oligo(dT)-primedUni-Zap cDNA library was synthesized using a Lambda-Zap cDNA synthesis kit from Stratagene (La Jolla, CA) following the manufacturer's instructions. For screening, the phage were plated, employing XL-1-Blue ~ s c h e r ~ h coli i a as host, at a density of lo6plaque-forming uni~/plate.A total of approximately 2 X 10' clones were screened with a 1-kb fragment probe containing the 3' sequence of rat liver cytochrome P-450 2Cll cDNA (18), labeled by nick translation with [32P]dATP.Nucleic acid hybridizations were done at 65 "C in 0.9 M NaCI, 90 mM sodium citrate (pH 7.0) containing heat-denatured salmon sperm DNA (0.1 pglml). Duplicate positive clones were plaque-purified, the phage DNA isolated and rescued into pBluescript SR(+) (Stratagene). DNA Sequencing and Nucleic Acid Blot Hybridization-Plasmid DNAs were replicated in DH5a competent Escherichia coli (In Vitrogen, San Diego, CA) grownin Luria Broth containing ampicillin (100 pg/ml). For plasmid purification, the cells were lysed in the presence of0.2 N NaOH containing 1% sodium dodecyl sulfate andthe pBluescript plasmid purified using Qiagen Columns (QiagenInc. Chatsworth, CA)following the manufacturer's instructions. The pBluescript cDNA insert (1.9 kb) was sequenced by the dideoxy chain terminat~onmethod (19) using Sequenase (U. S. Biochemical Corp.). 25 oligonuclotides (15-20-mer, each) spanning thelength of the cDNA were utilized as primers for the sense and/or antisense cDNA strands. Nucleotide sequences were analyzed utilizing the IntelliGenetics Suite Molecular Biology Software andthe GenBank On-line Service (GenBank/IntelliGenetics, Inc. Mountain View, CA). Samples of poly(A)+RNA (2-5 pg), isolated from rat liver, kidney, heart, and brain, were denatured and electrophoresed in agarose gels (1.3%)containing 0.2 M formaldehyde as described (20). After transfer to a nitrocellulose membrane, the blots were hybridized with either a 1-kb probe containing the 3' sequence of P-450 2Cll (18), a 1.3-kb fragment containing a region of the 3' sequence of P-450 2C7 (21), the cloned P-450 2C23, or alternatively, a P-450 2C23-specific 51mer oligonucleotide (5'-ccttcccattattgggaatctactgcaactgaacctcaaggacatcctgca-3'; residues 118-168 of the RD3 cDNA clone). Doublestrand probes were labeled by nick translation using E. coli polymerase Iand [ c x - ~ ~ P J ~ A Single-strand TP. probes were end labeled using T4 polynucleotide kinase and [y3'P]ATP. After exposure, the blots were rehybridized to a cDNA probe encoding rat liver tubulin. Transfection and Expression in COS-1 Cells-For transfection, the cDNA fragment was released from the pBluescript SK(+) phagemid and subcloned into pBluescript I1 (Stratagene), afteradding an EcoRI site at the 3' end. This cDNA was released from the pBluescript I1 plasmid by digestion with KpnI and XbaI and cloned into thepCMV2 vector (a gift from Dr. Michael Waterman, Vanderbilt University) as described (22). The presence of an XbaI site in the 3'-untranslated region of the cloned cDNA yielded a 1,680-basepair insert containing the open reading frame for a polypeptide of 494 amino acids. After propagation, the pCMV2 plasmid DNA was purified by centrifugation in CsCI. For transfection, 20 pg of plasmid DNA and 40-50 pg of Lipofectin (GIBCO), each in 250 pl of OPTI MEM (GIBCO/BRL) were mixed and used within the next 10min. COS-1 cells were grown to 60-70% confluence (100-mm dishes) in Dulbecco's modified Eagle's medium (Sigma) containing 10% (v/v) fetal calf serum,washed twice with serum-free medium, and then 2.5 ml of OPTI MEM and 0.5 ml of the DNA/Lipofecting mixture were added. After 5 hat 37 "C, 7ml of Dulbecco's modified Eagle's medium containing 10% fetal calf serum was added and the incubation continued for a total of 72 h with medium changes every 24 h. Cells were washed with serum-free medium, scraped from the plates into 10 mM Tris-C1 (pH 7.5) containing 0.5 M sucrose, 1 mM EDTA, 1 mM phenylmethylsulfonyl fluoride, leupeptin (0.1 pg/ml), and aprotinin (0.04units/ml) (22) and homogenized utilizing a glass/Teflon homogenizer. Microsomal fractions were prepared by differential centrifugation exactly as described (22). Microsomal fractions were isolated

from nontransfected cells or from cells transfected with the pCMV2 vector either lacking the coding insert or containing an antisense cDNA insert. Microsomal Incubations and Product C~r~~erizution-Microsomal fractions were suspended (1 mg of proteinlml) in 50 mM Tris-C1 buffer (pH 7.5) containing 150 mM KCl, 10 mM MgCI,, 8 mM sodium isocitrate and isocitrate dehydrogenase (0.5 IU/ml) After 1 min at 30 "C, [1-"Clarachidonic acid (25-50 pCi/pmol, 100 p~ final concentration) was added, followed by NADPH (1mM final concentration). Aliquots of the reaction mixture were withdrawn at different time periods, the organic soluble products extracted into ethyl ether and analyzed by TLC (SO': hexane/ether/CHZC12/CH3COOH,52:l:O.l). Prior to TLC analysis, excess [l-'4C]arachidonic acid was removed by chromatography in a Si02 column (0.4 X 1 cm) equilibrated with hexane/CH&OOH (201). Organic products, dissolved in equilibration solvent, were loaded onto thecolumn, and afterelution of excess arachidonic acid with the same solvent, oxygenated metabolites were eluted with ethyl ether/~exane/CH3COOH (416:l). Products were detected and identified using synthetic [l-"CJEETsand [ l - W ] DHETs by exposing the plates to x-ray film (Eastman Kodak). For regiochemistry studies, reaction products were first purified by reversed phase HPLC (23) and then resolved by normal phase HPLC exactly as described (23). For chiral analysis the EETswere collected from the HPLC eluant, derivatized, purified, and resolved into the corresponding antipodes by chiral phase HPLC as described previously in (24). The resolved [l-"C]EET antipodes were dried under argon and quantified by liquid scintillation. RESULTSANDDISCUSSION

~so~ation and Charac~erizationof the Rat Kidney Arachidonic Acid Epxygenase c ~ ~ A - I n h i b i t i o nstudies utilizing polyclonal antibodies against several rat liver P-450 isoforms showed that anti-cytochrome P-450 2Cll was a selective and potent inhibitor of kidney epoxygenase (15). These results suggested that the rat kidney arachidonate epoxygenase belongs to the P-450 2C gene family (25 and references therein). Furthermore, comparisons of the regio- and stereoselectivity of therat kidney microsomal epoxygenase with those of purified P-450 2Cll (5) indicate that although these proteins exhibited common immunochemical determinants, they are nonetheless catalytically different proteins with distinct active site geometries ( 5 ) .The high enantiofacial selectivity of the rat kidney microsomal enzyme as well as the inability of typical P-450 inducers to alter either its activity, regioselectivity, and/or s~reoselectivityindicated that, different from the liver, the molecular heterogeneity of the renal epoxygenase is limited (5). Based on the immunological evidence, the rat kidney cDNA library was screened with a 1-kb nucleotide fragment containing the 3' sequence of rat liver P-450 2Cll (18). Approximately 100 positive clones were identified of which 40clones, selected at random, were plaque purified and rescued into pBluescript. Insert sizes were determined by agarose electrophoresis after EcoRI and XhoI digestion of the purified plasmids. Initial sequence analysis, utilizing T3 and T7 primers, demonstrated that ( a ) the majority of the positive clones (27 out of 40, 68%) shared sequence identity with rat liver P-450 2C23 (17); ( b )12 clones contained non-cytochrome P-450 inserts; and (c) the remaining two clones contained nucleotide sequences homologous to P-450 2E1 (26). One clone (RD3) contained part of the published 5' end and the entire coding sequence for P-450 2C23, and it was therefore selected for further study. Sequence analysis demonstrated that RD3 contained an insert of 1,924 nucleotides with 99% sequence identity to the published sequence for rat liver P-450 2C23 (17). An open reading frame, coding for a polypeptide of 494 amino acids, was flanked by initiation and termination codons (ATG and TAA) positioned between nucleotides 6and 1488, respectively. Compared with the cDNA for P-450 2C23 (171, the 3' end of RD3 was 230 nucleotides shorter and did not contain

Molecular Cloning of a Renal Arachidonic Acid Epoxygenase a poly(A) tail. Additionally, RD3 lacked a nucleotide ( G ) a t position 111 and had an extra base (C) at position 168 (17). The deduced amino acid sequence for RD3 contained a putative heme-binding FSLGKEAGVGESLAR peptide with the underlined conserved residues and t h e invariant cysteine heme-ligand at position 439 (27 and references therein). As a consequence of the base deletion and insertion mentioned above, between nucleotides 110 and 167 the coding sequences for RD3 and P-450 2C23 were different (17). Only the RD3 region codes for a peptide that is conserved among most of the members of this P-450 gene family. A comparison of the RD3 nucleotide and/or amino acid sequences with those of other members of the P-450 2C subfamily indicated that with the exception of P-450 2C23, the extentof similarity was limited. Thus, humanP-450 2C18 (21) exhibited the second highest degree of sequence identity with RD3 (67 and 61% for nucleotide and amino acid sequence, respectively). Moreover, the differences in nucleotide sequence between RD3 and other members of the 2C gene subfamily were randomly distributed along the entire length of the cDNA, indicating that thegene for P-450 2C23 or RD3 diverged from the members of the 2C subfamily at an early stage. Importantly, many of the differences in the aminoacid sequence between P-450 2C23 or RD3 and other members of this subfamily represent conservative changes, i.e. replacements with residues with overall similar chemical properties. Expression and Enzymatic Characterization of the Recombinant Arachidonic Acid Epoxygenase-The use of COS-1 cells for the transient expression of cloned P-450s is well documented (22). COS-1 cells have an endogenous NADPH-cytochrome P-450 reductase, fully capable of effective redox coupling to several expressed P-450s (22). Additionally, expression requires minimalmanipulations of the cDNA, and the cells are easily grown and maintained. To characterize the catalytic properties of the protein molecule coded by RD3, COS-1 cells were transfected with a pCMV2 plasmid containing either asense oranantisense 1,680-base pairinsert spanning theRD3 5’ end andopen readingframe. Microsomal fractions were then isolated from the cultured cells and incubated with [l-’4C]arachidonic acid inthe presence of NADPH and an NADPH-regenerating system (23). As shown in Fig. 1, microsomes isolated from cells containing the antisense RD3 plasmid did not metabolize arachidonic acid to a significant extent. On the other hand, microsomal fractions isolated from cells transfected with the sense RD3 plasmid metabolized the fatty acid to products with chromatographic properties similar to those of synthetic EETs and DHETs. Furthermore, as shown in Fig. 2, the catalysis of EET and DHET formation was incubation time-dependent. It is apparent from the datain Figs. 1and 2 that either,once formed, the EETs underwent extensive chemical hydration or that COS-1 microsomal fractions contained anactive epoxide hydrolase activity, responsible for their rapid hydration to the corresponding DHETs (28). Inasmuch as DHET formation must be preceded by EET formation, we concluded that RD3 coded for a renal arachidonic acid epoxygenase. The translated, primarystructure of the kidney arachidonate epoxygenase was nearly identical to that of the protein coded by a cDNA isolated previously from a rat liver cDNA library and designated P-450 2C23 (17). To the best of our knowledge, this is the first time that the P-450 2C23 gene product has been expressed and a catalytic activity for the recombinant protein determined. The analysis of product regio- and stereochemical properties provided important information concerning the P-450 isoform heterogeneity of the rat liver arachidonic acid epox-

ARA

13567

+

4-

EETs -D

4-

DHETs -D

4-

I

I

I

I

I

A

B

C

D

E

FIG. 1. TLC analysis of the metabolites generated by incubates containing the microsomal recombinant epoxygenase and arachidonic acid. Microsomal fractions, isolated from COS-1 cells transfected with the pCMV2 vector containing either the sense (lane D )or the antisense ( l a n e C) cDNA coding for the rat kidney epoxygenase, were incubated with [I-“Clarachidonic acid and NADPH for 30 min a t 30 “C. Reaction products were extracted into ethyl ether and, after removal of excess arachidonic acid, resolved by TLC as detailed under “Materials and Methods.” For detection, the TLC plates were exposed to x-ray film for 24-48h. A mixture of synthetic [1-“Clarachidonic acid, 8,9-, 11,12-, and 14,15-[1-14C]EET, and of 8,9-, 11,12-, and 14,15-[1-’4C]DHETwas added to lanesA and E; lane B contained the [ l-14C]fattyacid used as substrate. (RFvalues for selected metabolites were 0.225, 0.103, and 0.058 for EETs, 19and 20-hydroxyeicosatetraenoicacids, and DHETs, respectively.)

“ I

4- ARA

“ I

+EETs

“ I

4-

I

0

I

15

DHETs

I

30

TIME (mid

FIG. 2. Time course of EET and DHET formation by incubates containing the microsomal recombinant epoxygenase and arachidonic acid. Microsomal fractions isolated from COS-1 cells transfected with the pCMV2 vector containing the RD3 clone were incubated with [1-“Clarachidonic acid and NADPH. Samples of the reaction mixture were taken immediately after the addition of NADPH, as well as 15 and 30 min after. The organic soluble products were extracted into ethyl ether and analyzed by TLC as described under “Materials and Methods.” The chromatographic properties of 1-“C-labeled synthetic EETs,DHETs,and arachidonic acid are shown.

ygenase (5). Studies utilizing purified hemoproteins and/or microsomal fractions, isolated from animals treated with selected P-450 inducers, demonstrated the coexistence in liver of several protein isoforms capable of epoxidizing arachidonate with distinct regio- and/or enantiofacial selectivities (5). T o ascertainthe role of the recombinant enzyme in the catalysis of arachidonic acid epoxidation by rat kidney, mi-

13568

~ o ~ cC u l ~ o rof ~a Renal ~ ~A r ~ ~ i ~ o nAcid i c~poxygen~e

crosomal fractions isolated from transfected COS-1 cells were incubated with radiolabeled arachidonic acid, NADPH, and 4-fluorochalcone oxide(10 PM,final concentration), an inhibitor of epoxide hydrolase (29). After extraction, the reaction products were resolved by reversed HPLC as described (23). The data in Fig. 3 confirm that the RL)3 cDNA clone coded for an arachidonate epoxygenase. Thus, microsomal fractions prepared from COS-1 cells containing recombinant rat kidney P-450 2C23 metabolized arachidonic acid to generate EETs and DHETsas their only oxygenationproducts (Fig. 3).As it is the case with the kidney microsomal enzyme (Is),11,12EET was the predominant epoxygenase metabolite generated by the recombinant cytochrome P-450 2C23 epoxygenase (54% of the total), followed by 8,9-and 14,15-EET (27 and 14% of the total, respectively). The recombinant enzyme generated small quantities of the labile 5,6-EET regioisomer as revealed by the presence of radioactive material eluting with an HPLC retention time similar to thatof authentic 5,6DHET. Importantly, the enantiofacial selectivity of the recombinant epoxygenase closelymatches that reported for the rat kidney microsomal enzyme (15) and generated 8(R),9(S)-, ll(R),12(S)- and 14(S)15(R)-EETs with optical purities of 95,85, and 75%, respectively (Table I). We concluded, therefore, that the RD3 cDNA codes for the predominant P-450 epoxygenase of rat kidney and that therenal hemoprotein is nearly identical to liver P-450 2C23 (17). A corollary of the studies with the recombinant enzyme was the demonstration that a single protein catalyst mediates, albeit with different efficiencies, the asymmetric epoxidation at the fatty acid 8,9-,11,12- and 14,15-0lefins,Since with the recombinant protein the a s y m m e t ~of the epoxygenase is under the control of a single protein catalyst, we propose that the P-450 2C23 active site molecular coordinates responsible for heme-fatty acid spatial orientation allowfor a limited degree of substrate lateral displacement but, on the other hand, are remarkably rigid in defining the olefin enantiotopic face exposedto theheme-bound active oxygen (5). E x ~ r e s s ~Levels n of the C y t o ~ h P-450 ~ o ~ 2C23 Arachi-

donic Acid Epoxygenase Gene-The relative abundance of the P-450 2C23 epoxygenase mRNA in the rat kidney was evaluated by nucleic acid blot hybridization (20). Poly(A)+RNA was isolated from a male rat kidney, electrophoresed, and hybridized to several P-450 2C family probes, cloned fromour cDNA library, and to a P-450 2C23 specific oligonucleotide (for sequence details see “Materials and Methods”). The labeled cDNA probes utilized contained sequences for P450 2Cll (RD1, 1 kb), 2C7 (RD2, 1.3 kb), and 2C23 (RD3, 1.9 kb). The proposal that the predominant arachidonic acid epoxygenase isoform in rat kidney is P-450 2C23 is further substantiated by the data in Fig. 4. As shown in this figure the organ steady-state concentrations of P-450 2C23 mRNA, measured by hybridization with the RD3 cDNA or with a P450 2C23-specificnucleotide,were markedly higher than those for P-450 2Cll (RD1) and 2C7 (RD2). Previous studies have demonstra~d that in addition to P-450 2C23, several other purified P-450 isoformscan actively catalyze the epoxidation of arachidonic acid (5-7, 16). Additionally, no information is as yet available concerning the catalytic activity of P-450 2C23 toward substrates otherthan arachidonic acid. The levels of expression of the P-450 2C23 arachidonic acid epoxygenase in different organs were studied in poly(A)+ mRNA samples extracted from whole liver, kidneys, heart, and brainobtained from a single male rat. Blot hybridizati~n using the RD3 cDNA clone demonstrated that although the levels of expression for liver and kidney were similar, the mRNAwas undetectabIe in brain and heart (Fig. 4). The members of the P-450 2C gene family show extensive sequence homology, and therefore cross-hybridization is a common problem (30). The P-450 2C23-specific 51-base oligonucleotide was designed to minimize cross-hybridization between members of this gene subfamily. Blot hybridization with the 51-base oligonucleotideindicated the predominant expression of the mRNA coding for P-450 2C23 in the kidneys (Fig. 4). Compared with liver, the kidney is functionally and anatomically a highly heterogeneous organ. Several correlations between enzymatic strati~cationand the anatomical andfor COS-1 Cells Microsomal Fraction

FIG. 3. Reversedphase HPLC resolution of thearachidonicacid metabolites generated by microsomal fractions isolated from COS-1 cells expressing the recombinantrat Microsomal kidney epoxygenase. fractions were incubated at 30 “C in the presence of [lJ4C]arachidonic acid (50 mCi/mmol, 100 pM), 4-fluorochalcone oxide (10 PM), and NADPH (1 mM) for 20 min. Products were extracted into tion, by after chromatography in a 5ethyl resolved ether and, solvent evapora-

+

I a

pm

k* z

X

0

Dynamax Microsorb GIscolumn (4.6 250 mm, Rainin Inst. Co.) exactly as described (23). Reaction products were

detected and quantified with an on-line Radiomatic Flow-one 6 Detector (Radiomatic Instrument Co.).Shown are the radiochromatograms derived from a total of 1 mg of microsomal protein isolated from control, nontransfected cells (bottom) or cells transfected with the epoxygenase cDNA (top).The retention times for 11,12-DHET ( A ) , 5,6-DHET ( B ) ,1 4 , S E E T ( C ) , 11,lZ-EET (D), and 8,9-EET ( E ) are indicated in thefigure.

pCMV2-2C23

h

O

1.

I

J\

I I

A

-

1

0

j, 0

6

I Ii

C DE

pCMV2-2C23

~

i,

t” ,

-;’”.,.’.*; 10

~.

20 RETENTION (mid

30

40

Molecular Cloning of a Renal Arachidonic Acid Epoxygenase TABLE I

4A1

Regio and stereoselectivityof the recombinant rat kidney arachidonic acid epoxygennse Microsomal fractions, isolated from COS-1 cells transfected with the pCMV2 plasmid containing the cloned RD3 cDNA, were incubated with [I-"Clarachidonic acid in the presence of NADPH and 4fluorochalcone oxide, an epoxide hydrolase inhibitor. After 20 min a t 30"C, the reaction products were extracted, resolved by reversed phase HPLC, and derivatized as described under"Materials and Methods." The optical antipodesof derivatized 8,9-, 11,12-,and 14,15EET were resolved bv chiral uhase HPLC asdescribed (24). Regioisomers

% of Total

% R,S

%S,R

8,9-EET 11,12-EET 14,15-EET

27

95 85 25

5 15

RD 1

54

75 19 RD2

5 1 -mer

RD3

RD1

~-~~

" " I

I I

c

I I

c

s

s

13569 51 -mer

RD3 . - ."

' " I

I I

c

s

I I

c

s

FIG.5. Effectof excess dietarysalt in the levels ofrat kidney P-450mRNAs. 3-5 pg of poly(A)+ RNA, isolated from the kidneys of control (C) and of rats fed a 2% NaCl solution as drinking water for 10 days (S), was electrophoresed and blot hybridized to each of the following 32P-labeledrat kidney probes: the RD1 cDNA clone, the RD3 cDNA clone, a 51-mer oligonucleotide specific for P4502C23 and, to a cDNA clone containing the full-length coding sequence of rat P-450 4A1(25,27 andreferences therein). To increase detection sensitivity, the individual blots were exposed to the x-ray film for different time periods. After exposure to x-ray film (51-mer, 12 h; RD3 and 4A1, 24 h; RD1, 72 h), the blots were stripped and probed with [ 3 2 P ] t ~ b ~cDNA. lin Densitometric analysis showed that the percent relative abundance of the tubulin mRNA in control and salt-treated blots was similar.

excess dietary salt is not P-450 2C23. Moreover, the data also indicate that salt induces a P-450 isoform either absent or present a t very low levels in the kidneys of control animals HT ER KO LV HT ER KO LV HT BR KD LV HT ER KO LV FIG.4. Nucleic acid blot hybridization analysis of male rat (15). mRNAs. 5 pgof poly(A)' RNA, isolated from the kidneys (KD), In conclusion, we report here the molecular cloning, the liver (LV), heart (HZ'), and brain ( B R ) of a male Sprague-Dawley initial structural characterization, and the expression of the rat (320 g, body weight), was submitted to electrophoresis and blot major form of rat kidney cytochrome P-450 active in the hybridized either to a 51-mer oligonucleotide specific for cytochrome highly asymmetric epoxidation of arachidonic acid. AddiP-450 2C23, to theRD3 cDNA clone, or totwo cDNA clones, isolated tionally, we demonstrate unequivocally that a single protein from a male rat kidney cDNA library,containing sequences for can catalyze epoxidation at the fatty acid 5,6-, 8,9-, 11,12-, cytochromes P-450 2Cll (RD1) or 2C7 (RD2). After exposure to xray film (RD1, RD2, or RD3 probe, 2 h; 51-mer, 12 h), theblots were and 14,15- olefins. These results indicatethat asa member of stripped and probed with [32P]tubulincDNA. Densitometric analysis the renal arachidonic acid metabolic cascade, the cytochrome showed that the percent relative abundance of the tubulin mRNA P-450 2C23 epoxygenase plays a central role in the biosynwith respect to the liver sample (100%) was 260, 530, and 126% for thesis of the biologically active EETs and, consequently, may kidney, brain, and heart, respectively. have important role(s) in renal function.

I I I I

I l l 1

I l l 1

I

I

I

I

REFERENCES

functional segmentation of the nephron are known (31 and 1. Needleman. P.. Turk. J.. Jakschik. B. A.. Morrison. A. R. & Lefkowith. J. references therein). The data in Fig. 4, in conjunction with B. (1986fAnnu. Rev. Biochem. 55,691102 2. Capdevila, J., Chacos, N., Werringloer, J., Prough, R. A. & Estabrook, R. the reported segment specific biosynthesis of some P-450 W. (1981) Proc. Natl. Acad. Sci. U.S. A. 78.5362-5366 isoforms in the rat and rabbit nephron (32-34), suggest the 3. Morrison, A. R. & Pascoe, N. (1981) Proc. Natl. Acad. Sei. U.S A . 78, 7375-7378 potential for localized high level expression of the P-450 2C23 4. Oliw, E. H. & Oates, J. A. (1981) Biochim. Biophys. Acta 666,327-340 arachidonate epoxygenase. 5. Capdevila, J. H., Karara, A., Waxman, D. J., Martin, M. V., Falck, J. R. & Guengerich, F. P. (1990) J. Brol. Chem. 265, 10865-10871 Recently, it was demonstrated that a high dietarysalt 6. McGiff, J. C. (1991) Annu. Reu. Phrmacof. Toxrcol. 31,339-369 intake induces a kidney P-450 isoform active in the in uiuo 7. Ca_p_d_evila,J. H., Falck, J. R. & Estabrook, R. W. (1992) FASEB J. 6,731idb and in vitro epoxidation of arachidonic acid (15). Moreover, H.R., Corona, S., Capdevila, J., Chacos, N., Manna, S., Womack, the salt-inducedhemoprotein(s)appeared immunologically 8. Jacobson, A. & Falck, J. R. (1984) in Prostaglandins and Membrane Ion Transport (Braauet. P.. Garav. R. P.. Frolich.. J. C... and Nicosla.. S... eds) .. DD. 311different from that present in control rat kidney (15). The 318, h e n Press, New York effect of dietary salt onthe kidney levels of P-450 expression 9. Romero, M.F., Madhun, Z., Hopfer, U. & Douglas, J. G. (1990) Adu. Prostaglandin Thromboxane Leukotriene Res. 21, 205-208 was studied by blot hybridization utilizing poly(A)+ RNA Hirt, D. L.,Capdevila, J. H., Falck, J. R., Breyer, M. D. &Jacobson, H. R. 10. samples extractedfrom the kidneys of control and salt-treated (1989) J. Cfin.Invest. 84, 1805-1812 rats (15) and full-length cDNA probes coding for P-450s 2C23 11. Schlondorf, D., Petty, E., Oates, J. A., Jacoby, M. & Levine, S. D. (1987) Am. J. Physiol. 253, F464-F470 (RD3) and 4A1 (a gift from Dr. Gordon Gibson, University of 12. Kato, T.,Takahashi, K., Capdevila, J., Karara, A., Falck, J. R., Jacobson, H. R. & Badr, K. F. (1991) Am. J. Phystol. 261, F578-F586 Surrey, Guilford, Surrey) (25, 27 and references therein), the F., Lawson, J. A., Fitzgerald, D. J. & Fitzgerald, G. A. (1990) Proc. RD1 (2Cll) cDNA clone, as well as, the 51-mer 2C23-specific 13. Catella, Natf.Acad. Sci. U.S. A. 87,5893-5897 oligonucleotide described above. As shown in Fig. 5, high 14. Iwai, N.& Inagami, T. (1990) Hypertension 17,161-169 J. H., Wei, S., Yan, J., Karara, A., Jacobson, H. R., Falck, J. R., dietary salt intakeresulted in a marked reduction in the levels 15. Capdevila, Guengerich, F. P. & DuBois, R. N. (1992) J. Biol. Chem. 267, 2172021726 of a kidney P450 mRNA that hybridized to the 2Cll probe R. M., Laethem, C. L. & Koop, D. R. (1992) J.Biol. Chem. 267, (RD1). This marked reduction in the organ levels of 2 C l l 16. Laethem, 5552-5559 17. E. A., Groenewegen, W. A., Gloger, I. S., Piggott, J. R., Suckling, K. Cook, mRNA was evident within the first 12 h of salt treatment, E. & Wolf, R. C. (1990) Nucleic Acids Res. 18,7156 and it was maintained for a t least 12 days (not shown). On 18. Shimada. T., Misono. K. S. & Guengerich, F. P. (1986) J. Biol. Chem. 261, 909-921 the other hand, salt treatment did not significantly alter the 19. Sanger, F., Nicklen, S. & Coulson. A. R. (1977) Proc. Natf. Acad. Sci. steady-state levels of kidney cytochromes P450 2C23 or 4A1 1 1. . S. . A.. ?.A- ,.SARR-.SAfi'l .- - .- . mRNAs. The resultsin Fig. 5 as well as published data 20. DuBois, R. N., McLane, M., Ryder, K., Lau, L. F. & Nathans, D. (1990) J. Eiof. Chem. 265,19185-19191 obtained utilizing immunological techniques (15) are consist- 21. Romkes, M., Faletto, M. B., Blaisdell, J. A., Raucy, J. L. & Goldstein, J. A. ent with the proposal that the P-450 isoform induced by (1991) Biochemistry 30,3247-3255 ~~

~

Y

1

13570

Mobculur Cloning of a Renal A r ~ ~ ~ i dAcid o n ~Epoxygenase c

22. Clark, B. J. i% Waterman, M. R. (1991) Methods Enzymoi. 206,100-108 23. Capdevila, J. H.,Falck, J. R., Dishman, E. & Karara, A. (1990) Methods EnzymoZ. 187,385394 24. Hammonds, T.D.,Blair, A. I., Faick, J. R. i% Capdevila, J. H.(1989) Anal. B b ~ k m 182,300-303 . 25. Nebert, D. W. & Gondez, F. J. (1987)Annu. Rev. Biochem. 56,945-993 26. Khani, S. C., Porter, T . D. & Coon, M.J. (1988) Biochem. Biophys. Res. Commun 150,10-17 27. Gonzalez, F. J. (1989) P h a r ~ o L Rev. 40,243-288 28. Chacos, N., Capdevila, J., Falck, J. R, Manna, S., Martin-Wixtrom, C,,

Gill S. S HammockB. D. & Estabrook, R. W. (1983) Arch. Bimkm. Eiobhys. 223,639-6d8 29. Prestwich, G. D. Kuo, J. W., Park, S. K., Loury,D. N. & Hammock, B. D. (1985) Arch dkhem. Biophys. 2 4 2 , l l - 1 5 30. Waxman D. J. (1991) Methods Enzymol. 206,249-267 31. Morris02 A. R. (19%) Am. J. Med. 80, 3-11 32. Dees, J. H., Masters, B. S. S., Muller-Eberhard,U. &Johnson, E. F. (1982) Cancer Res. 42,1423-1432 33. Endou, H.(1983) J p a J. PhamueoL 33,423-433 34. Omata, K., Abraham, N. G. & Schwartzman, M. L. (1992) Am. J. Physiol. 262, F591-F599