Contribution of CYP4A8 to the Formation of 20-Hydroxyeicosatetraenoic Acid from Arachidonic Acid in Rat Kidney

Drug Metabol. Pharmacokin. 17 (2): 109–116 (2002).

Regular Article Contribution of CYP4A8 to the Formation of 20-Hydroxyeicosatetraenoic Acid from Arachidonic Acid in Rat Kidney Yoshitaka YAMAGUCHI1, Shirou KIRITA1, Hiroshi HASEGAWA1, Junko AOYAMA1, Susumu IMAOKA2, Sachiko MINAMIYAMA3, Yoshihiko FUNAE2, Takahiko BABA1 and Takashi MATSUBARA1 1Department

of ADME and Toxicology for Screening, Developmental Research Laboratories, Shionogi & Co., Ltd., Toyonaka, Osaka, Japan Departments of 2Chemical Biology and of 3Biochemistry, Osaka City University Medical School, Abeno-ku, Osaka, Japan Summary: 20-Hydroxyeicosatetraenoic acid (20-HETE) has been shown to be an arachidonic acid metabolite of the cytochrome P450 (CYP) enzymes belonging to the CYP4A subfamily and is a predominant regulator of renal vascular tone and tubular ion reabsorption in rat kidney. CYP4A8 is one of the CYP4A enzymes expressed in rat kidney, but its contribution to 20-HETE formation has not been assessed. In order to clarify that the role of CYP4A8, we have developed bacterial expression systems for the expression of recombinant CYP4A8 (rCYP4A8). We also produced an antibody against rCYP4A8 which was used for immunoinhibition and immunohistochemical studies. In a reconstituted system, rCYP4A8 su‹ciently catalyzed 20-HETE formation as well as prostaglandin A1 v-hydroxylation, a marker activity for CYP4A8. In addition, anti-rCYP4A8 sera signiˆcantly inhibited prostaglandin A1 vhydroxylation and strongly inhibited arachidonic acid v-hydroxylation in rat kidney microsomes. These observations suggested for the ˆrst time that CYP4A8 also contributed to 20-HETE formation in rat kidney. Furthermore, immunohistochemstry suggested that CYP4A8 is present in preglomerular arteries, where 20-HETE has been established to be a vasoconstrictor.

Key words: cytochrome P450; CYP4A8; 20-HETE; kidney; immunoinhibition; immunohistochemistry can be eŠectively suppressed by inhibitors of the CYP4A enzymes,2,9,10) indicating that CYP4A enzymes are involved in the formation of 20-HETE. Four CYP4A enzymes (CYP4A1, 4A2, 4A3 and 4A8) have been isolated from rats and are present in rat kidney. CYP4A2 is the major constitutive CYP enzyme in rat kidney (approximately 70z of total CYP in rat kidney microsomes) and is expressed at signiˆcantly higher levels in spontaneous hypertensive rats than in normal rats.11) 20-HETE formation is also higher in spontaneous hypertensive rats when compared to normal;12–14) these observations have lead to the theory that CYP4A2 may play a signiˆcant role in renal 20-HETE formation and the regulation of vascular tone in rats. However, this idea is inconsistent with results from previous reports. For example, Kroetz et al. demonstrated that 20-HETE formation by kidney microsomes and CYP4A2 expression levels increased in an age-depen-

Introduction 20-hydroxy-5,8,11,14-eicosatetraenoic acid (20HETE) is a major metabolite of arachidonic acid (AA) in male rat kidney and has been shown to play an important role in the regulation of renal vascular tone and tubular ion reabsorption. 20-HETE inhibits the opening of a large conductance potassium channel in vascular smooth muscle cells, resulting in vasoconstriction and the regulation of arteriole vascular tone.1,2) In addition, both the proximal tubular Na+, K+-ATPase and the Na+-K+-2Cl„ cotransporter in the medullary thick ascending limb of kidney are inhibited by 20-HETE.3–5) In salt-sensitive Dahl rats, reduction of 20-HETE formation causes an increase in Cl„ reabsorption.6) The formation of 20-HETE is catalyzed by cytochrome P450 (CYP) enzymes belonging to the CYP4 gene subfamily7,8) and its physiological actions

Received; November 30, 2001, Accepted; January 28, 2002 To whom correspondence should be addressed : Yoshitaka YAMAGUCHI, Department of ADME and Toxicology for Screening, Developmental Research Laboratories, Shionogi & Co., Ltd., 3-1-1 Futaba-cho, Toyonaka, Osaka 561-0825, Japan. Tel. +81-6-6331-8342; Fax. +81-6-63318900; E-mail: yoshitaka.yamaguchi＠shionogi.co.jp

109

110

Yoshitaka YAMAGUCHI, et al.

dent fashion in spontaneous hypertensive rats, but they were not synchronized at each other.15) Another report showed that although CYP4A2 is expressed at a low level in female rat kidney,16,17) 20-HETE formation in both male- and female-derived kidney microsomes was similar.18) Additionally, pretreatment with antisence oligonucleotides of CYP4A2 does not signiˆcantly reduce 20-HETE formation in proximal tubules or microvessels of rat kidney.19) Collectively, these observations suggest that another CYP4A enzyme may be contributing to 20-HETE formation in rat kidney. In the other members of the CYP4A enzyme family, CYP4A1 and CYP4A3 are expressed at very low levels in rat kidney.11,20–24) In contrast, CYP4A8 is constitutively expressed in rat kidney25,26) and its expression level is about 10-fold less than that of CYP4A2.27) Although its 20-HETE formation activity is much higher than that of CYP4A2,28) it has been not assessed whether CYP4A8 is actually contributing to 20-HETE formation in rat kidney. In order to clarify that, we performed immunoinhibition and immunohistochemical studies and obtained the results suggesting that CYP4A8 also contributes to 20-HETE formation in rat kidney and likely in proglomerular arteries. Materials and Methods

Materials [1-14C]-AA (1.89 GBq W mmol) was obtained from DuPont-New England Nuclear (Boston, MA). Prostaglandin A1 (PGA1 ), lauric acid (LA), AA, dilauroylphosphatidylcholine (DLPC) and dithiothreitol (DTT) were purchased from Wako Pure Chemicals (Osaka, Japan). 20-Hydroxy-PGE1 was a kind gift from Dr. Masami Tsuboshima, Ono Pharmaceuticals, Osaka, Japan. Phosphatidylserine (PS), phosphatidylcholine (PC), 20-HETE, and 12-hydroxylauric acid were from Sigma Chemical, Co. (St. Louis, MO). Emulgen 911 was supplied from Kao Co., Ltd. (Tokyo, Japan). Bacterial Expression and Puriˆcation of CYP4A8 CYP4A8 cDNA (1527-bp) was cloned from a rat kidney cDNA (Stratagene, La Jolla, CA) using PCR with the speciˆc primers designed from a previously reported sequence.25) Two diŠerent expression systems were developed for recombinant CYP4A8 (rCYP4A8) using pCW vector, a kind gift from Professor F. P. Guengerich (Vanderbilt University, Nashville, TN). The ˆrst expression system of rCYP4A8 required base modiˆcations at the N-terminal sequence identical to those used in the expression of rCYP3A4 in Escherichia coli ( E. coli ).29) These modiˆcations were done by PCR mutagenesis in two stages. The forward PCR primer for the ˆrst stage was 5?-TT-

ATTAGCAGTTTTTCTCACTGTGCTCCTGCTG-3? and that for the second was 5?-CGGCATATGGCTCTGTTATTAGCAGTTTTTCTC-3?. The reverse primer in both rounds of PCR was 5?-CAGCCGCACAAGCTTAGATTATTGGAGCTT-3?. The rCYP4A8 PCR product was inserted into the pCW vector according to the method of Gillam et al.29). The second expression system involved modiˆcations to the one above to include a (His)6 -tag at the C-terminus of rCYP4A8. DH5aF?IQ E. coli competent cells (Gibco-BRL, Gaithersburg, MD) were transformed with pCWrCYP4A8 or pCW-(His)6 -rCYP4A8 plasmids and these recombinant enzymes were expressed as previously described.29) E. coli cells were harvested and preparation of the spheroplast and membranes was done by slight modiˆcation to a method previously described.29) rCYP4A8 was principally puriˆed according to the published methods.29) In a case of (His)6-rCYP4A8, bacterial membranes were solubilized and mixed with TALON metal a‹nity resin (CLONTECH Lab. Inc., Palo Alto, CA). (His)6rCYP4A8 was eluted with 100 mM imidazole and further puriˆed with a KB Type-S column. The purity of the isolated enzymes was judged by SDS-PAGE using an acrylamide concentration of 7.5z (w W v).30) Proteins were visualized with Coomassie Brilliant Blue R250.

Animals and Tissue Preparation Male and female Sprague-Dawley rats were obtained from Japan Clea Laboratories (Tokyo, Japan) and sacriˆced at 8 weeks of age for tissue preparation. Both kidneys were homogenized with 50 mM tris-HCl buŠer (pH 7.4) containing 20z (w W v) sucrose and 0.1 mM EDTA, and microsomal fractions were obtained by diŠerential centrifugation as previously reported.26) For immunohistochemistry, rat tissues were perfused with 0.1 M phosphate buŠer (pH 7.2) containing 4z (w W v) paraformaldehyde and 0.5z (w W v) picric acid. Kidneys were removed and ˆxed in the above solution for 2 h at room temperature. After washing and overnight incubation in Holt's gum sucrose solution, 10 mm frozen tissue sections were prepared on poly-L-Lysine coated slides and stored at „409C prior to immunochemical staining. Preparation of Antibodies Against rCYP4A8 Antisera against rCYP4A8 were prepared according to the method of Kaminsky et al.31) Anti-IgG fractions were partially puriˆed with Hi-trap protein G column (Pharmacia LKB Biotech AB, Uppsala, Sweden). Antiserum against CYP4A2 was that as previously reported.11,32) Cross-reactivity of this antibody fraction was deter-

Contribution of CYP4A8 to 20-Hydroxyeicosatetraenoic Acid Formation

mined with CYP4A1, CYP4A2, CYP4A3 and CYP4A8 puriˆed from rat liver or kidney microsomes by immunoblot analysis.

Determination of PGA1 and Fatty Acid v-Hydroxylation Activities NADPH-CYP reductase and cytochrome b5 were puriˆed from rat liver microsomes as previously described.33,34) Recombinant CYP4A8 (50 pmol) was sequentially mixed with NADPH-CYP reductase (0.1 mmol cytochrome c reduced W min), phospholipids (30 mg), and cytochrome b5 (50 pmol). Either these reconstituted systems or rat renal microsomes (0.1 mg protein) were incubated at 379C for 10–15 min in 0.5 ml potassium phosphate buŠer (0.1 M, pH 7.4) containing PGA1 (100 mM), [1-14C]-AA (5 or 30 mM) or LA (200 mM) and 1 mM NADPH. The v-hydroxylated metabolites were measured by the methods outlined below. PGA1 and LA v-hydroxylation activities were determined as described previously.35,28) AA v-hydroxylation (20-HETE formation) activities were determined according to the method of Yokotani et al.36) with some modiˆcations. Formed 20-HETE was quantiˆed by measuring the ratio of 20-HETE radioactivity against total radioactivity with BAS-2000 radioluminography (Fuji Photo Film Co., Ltd., Tokyo, Japan). For immunoinhibition studies, rat kidney microsomes were preincubated with anti-rCYP4A8 immunoglobulin G (IgG) fractions for 30 min at room temperature and residual catalytic activities were measured. Immunohistochemistry Frozen tissue sections were incubated for 30 min in v) H2 O2 in methanol to destroy endogenous 0.3z (v W peroxidase activity. Sections were washed with 0.01 M phosphate buŠer (pH 7.4) containing 0.9z (w W v) NaCl and incubated overnight at 49 C with a 1:2000 dilution of preimmune, anti-rCYP4A8 or anti-CYP4A2 rabbit sera. The primary antibody was detected by using Vectastain Elite ABC kit with horseradish peroxidase and peroxidase substrate solution (Vector Laboratory). Other Assays CYP concentrations were measured spectrophotometrically according to the method of Omura and Sato.37) Protein concentrations were determined using a bicinchoninic acid protein assay kit (Pierce Chemical Co., Rockford, IL), using bovine serum albumin as the standard. Results

Expression and puriˆcation of rCYP4A8 and (His)6linked rCYP4A8 from E. coli.

111

The harvested bacterial cells expressing rCYP4A8 showed a typical CO-reduced diŠerence spectrum with the maximum absorbance at 452 nm (data not shown). Expression levels were approximately 200 nmol CYP per 1 liter of culture medium. Following puriˆcation, the recovery of rCYP4A8 was 1.9z from whole cells; the comparative CYP content was 2.53 nmol CYP W mg protein. Some contaminated proteins were detected by SDS-PAGE, but the partially puriˆed rCYP4A8 was judged to be acceptable for evaluating metabolic activities in a reconstituted system because no endogenous CYP enzymes were present in E. coli. (His)6-rCYP4A8 was highly puriˆed using metal a‹nity chromatography, but its CYP content was lower than that of rCYP4A8 (1.82 nmol CYP W mg protein). Final preparations of rCYP4A8 and (His)6-rCYP4A8 exhibited a characteristic and identical reduced COdiŠerence spectra, with maximum absorbance at 452 nm and very little absorbance at 420 nm, indicating the presence of little or no inactive enzyme (data not shown). PGA1 and AA v-hydroxylation by rCYP4A8 in a reconstituted system. As previously reported, CYP4A8 possesses PGA1 v-hydroxylation activity; which is not catalyzed by CYP4A1, CYP4A2 and CYP4A3.7,26) Therefore, PGA1 v-hydroxylation activity was measured using the partially puriˆed rCYP4A8. The PGA1 v-hydroxylated metabolite was formed by rCYP4A8 in the reconstituted system and its formation was signiˆcantly enhanced by min W nmol CYP vs. the addition of b5 [0.25 nmol W 2.73 nmol W min W nmol CYP (DLPC:PS＝1:1)]. Furthermore, various combinations of phospholipids produced a large range of PGA1 v-hydroxylation activities as following; 0.13 (DLPC:PC＝1:1), 0.52 (PC), 0.78 (DLPC:PC:PS＝1:1:1) and 3.72 (PC:PS＝1:1) nmol W min W nmol CYP. PS was likely to be an important component for the high activity of CYP4A8. 20-HETE formation activity by rCYP4A8 was also measured in the reconstituted system. Reconstituted rCYP4A8 formed 20-HETE from AA even at a physiologically relevant concentration (AA＝5 mM), and the activity was strongly enhanced by the addition of b5, similar to that seen with PGA1 v-hydroxylation activity. The turnover rate of 20-HETE formation by min W nmol CYP in rCYP4A8 was 0.78 and 2.97 nmol W the absence of b5 and 2.55 and 13.8 nmol W min W nmol CYP in the presence of b5 at 5 and 30 mM of AA, respectively. EŠects of anti-rCYP4A8 on LA v-hydroxylation and 20-HETE formation (AA v-hydroxylation) by rat kidney microsomes. In order to assess the contribution of CYP4A8 to 20HETE formation in rat kidney microsomes, an immunoinhibition study was performed using antibodies

112

Yoshitaka YAMAGUCHI, et al. Table 1.

EŠect of anti-rCYP4A8 IgG on LA v-hydroxylation and 20-HETE formation by male and female rat kidney microsomes.

Condition Control Preimmune Anti-rCYP4A8

LA v-hydroxylation activities (nmol W min W mg proein, n＝4)

20-HETE formation activities (pmol W min W mg proein, n＝3)

Male

Female

Male

Female

1.00±0.05 (100) 0.97±0.04 (97) 0.63±0.06* (62)

1.09±0.14 (100) 0.99±0.13 (92) 0.50±0.14* (46)

498±29 (100) 529±16 (106) 182±4* (37)

602±58 (100) 628±56 (104) 203±22* (34)

Kidney microsomes were pre-incubated for 30 min at room temperature with 0.25 mg IgG protein per 0.1 mg microsomal protein and then incubated with 200 mM of LA (A) or 5 mM of AA (B). Activity values represent the mean±S.D. from 4 or 3 animals. The relative values against control condition are shown in the parentheses. *Signiˆcantly diŠerent from control group ( pº0.05, Dunnett's t-test).

prepared against the recombinant (His)6-rCYP4A8. Prior to activity measurements, cross-reactivity of the antibody preparation against four native CYP4A enzymes (CYP4A1, 4A2, 4A3 and 4A8) was checked by immunoblot analysis (Fig. 1). The rCYP4A8 antibody recognized native CYP4A8 puriˆed from rat kidney and also had strong cross-reactivity with CYP4A1. In contrast, there was little cross-reactivity with CYP4A2 and none with CYP4A3. Although cross-reactivity of the anti-rCYP4A8 with CYP4A2 was quite low, there still remained a slight possibility that the antibody would also inhibit CYP4A2 activities. Therefore, we determined the eŠect of this antiserum on the both PGA1 v-hydroxylation activity, a CYP4A8 speciˆc marker, and LA v-hydroxylation activity, a general marker for CYP4A enzymes. First, the immunoinhibition of PGA1 v-hydroxylation in kidney microsomes from male and female rats was examined. After preincubating anti-rCYP4A8 IgG with kidney microsomal protein, PGA1 v-hydroxylation activity was signiˆcantly inhibited; its maximum eŠects were 77z and 86z inhibition in male and female rats, respectively. Control activities were higher in min W mg profemale than in male (male, 113±22 pmol W tein; female, 194±35 pmol W min W mg protein), indicating that the CYP4A8 content was approximately 2-fold higher in female rats than in male ones. In rat kidney, CYP4A2 is approximately 70z of the total CYP in male, but at low level in females.16,17) CYP4A2 is the predominant CYP enzyme responsible for renal LA v-hydroxylation in male rats, but not in females. Therefore, we compared the eŠects of antirCYP4A8 IgG on LA v-hydroxylation activities in male and female rat kidney microsomes in order to estimate the inhibitory eŠects on CYP4A2 (Table 1). There was no diŠerence in control activities between male and female rats, but the inhibitory eŠect of anti-rCYP4A8 IgGs (0.25 mg IgG protein per 1 mg microsomal protein) was lower in male rats (38z) than in females (54z). Finally, the immunoinhibition of 20-HETE formation in rat kidney microsomes by anti-rCYP4A8 IgG

Fig. 1. Immunoblot analysis of puriˆed rat CYP4A enzymes and rat kidney microsomes with anti-rCYP4A8 serum. Puriˆed CYP4A enzymes (lane 1, CYP4A1; lane 2, CYP4A2; lane 3, CYP4A3; lane 4, CYP4A8; 0.5 pmol CYP W well) and kidney microsomal protein from male rats (lane 5; 5 mg microsomal protein W well) were transferred to nitrocellulose membrane after SDS-PAGE and immunostained using anti-rCYP4A8 serum.

was examined. Anti-rCYP4A8 IgG (0.25 mg IgG protein per 1 mg microsomal protein) inhibited 63z and 66z of total 20-HETE formation by kidney microsomes from male and female rats, respectively (Table 1), and the inhibitory eŠects were much higher than the eŠects on LA v-hydroxylation in both sexes. The control activities were slightly higher in kidney microsomes from female rats than that from male rats similar to PGA1 v-hydroxylation, consistent with prominent expression of CYP4A8 in female rats. Immunohistochemical localization of CYP4A8 in rat kidney. To compare the distribution of CYP4A2 and CYP4A8 in rat kidney, frozen tissue segments from male rat kidney were immunochemically stained. The anti-CYP4A2 serum used in this study has no crossreactivity with CYP4A8 but does cross-react with CYP4A3 as previously reported;11,32) the selectivity of the anti-rCYP4A8 serum was as mentioned above. As shown in Fig. 2A, 2B and 2C, positive immunostaining with anti-rCYP4A8 serum was detected from the outer stripe of the outer medulla to the cortex, but was not detected in the inner stripe of the outer medulla or the in-

Contribution of CYP4A8 to 20-Hydroxyeicosatetraenoic Acid Formation

113

Fig. 2. Immunohistochemical localization of CYP4A8 and CYP4A2 in male rat kidney. Frozen tissue segments of whole kidney from male rats were immunostained with preimmune (A, negative control), anti-rCYP4A8 (B), and antiCYP4A2 (C) sera and a positive immunostaining was shown in brown color (A-C; bar＝1 mm). A cortex area in (B) and (C) was partially magniˆed (D, anti-CYP4A8 and E, anti-CYP4A2) and including the following regions; proximal straight tubule (PST), proximal convoluted tubule (PCT), cortical collecting duct (CCD), and glomerulus (Gm) sections (D and E; bar＝0.1 mm). Additional magniˆed pictures from those of whole kidney showed preglomerular arteries in outer stripe of outer medulla immunostained with preimmune (F, negative control), anti-rCYP4A8 (G), and anti-CYP4A2 (H) sera, respectively (F-H; bar＝20 mm).

114

Yoshitaka YAMAGUCHI, et al.

ner medulla. These immunohistochemistry observations are consistent with the results of Stromstedt et al. who did in situ hybridization using a speciˆc oligonucleotide probe for CYP4A8.25) Magniˆcation of the strongly stained region showed that rCYP4A8 immunoreactive proteins were localized in the proximal straight tubules and were weakly distributed in proximal convoluted tubules (Fig. 2D). In contrast, there was no detectable immunostaining in the distal convoluted tubules, cortical collecting ducts or glomerulus. Magniˆcation of the outer stripe of the outer medulla clearly showed immunostaining with anti-rCYP4A8 in the preglomerular arteries, although the extent of staining was lower than in other positive regions (Fig. 2G). There was no immunostaining with anti-rCYP4A8 serum in aŠerent arterioles or arcuate arteries (data not shown). The regions that showed positive staining for CYP4A2 were similar to those for CYP4A8, but were slightly expanded to the cortex (Fig. 2C) due to the fact that the immunostaining was distributed more strongly in the proximal convoluted tubules (Fig. 2E). Unlike anti-rCYP4A8 staining, anti-CYP4A2 serum produced no immunostaining in the preglomerular arteries (Fig. 2F and 2H); however, like with anti-rCYP4A8, the aŠerent arterioles and arcuate arteries were not stained by antiCYP4A2 (data not shown). In female rats, the distribution of immunostained proteins was similar to that in male rats both with the anti-CYP4A2 and anti-rCYP4A8 sera (data not shown). The staining density with anti-CYP4A2 sera was much lower in female rats than in males, but no signiˆcant diŠerence was observed in immunodensity with antirCYP4A8 between males and females. Discussion Bacterially expressed rCYP4A8 extensively catalyzed 20-HETE formation, as has been shown with puriˆed enzyme from rat kidney microsomes.26,28) Our present study demonstrated that the turnover rate of reconstituted CYP4A8 was strongly aŠected by the presence of b5 and phospholipid composition. Recently, Hoch et al. succeeded in the expression of (His)6-linked rCYP4A8 enzyme in E. coli,38) but formation of 20-HETE was ¿1 nmol W min W nmol CYP even at a high substrate concentration (AA＝100 mM), much lower than formation rates here. Additionally, Nguyen et al. reported that rCYP4A8 expressed in Sf9 cell membranes could not form 20-HETE.39) These discrepant results might be due to the use of suboptimal b5 and phospholipid concentrations in these studies. The antibody prepared against rCYP4A8 partially inhibited LA v-hydroxylation as shown in Table 1. The antibody has little cross-reactivity against CYP4A2 (Fig. 1) and has a low inhibitory eŠect on CYP4A2 catalyzed reactions. This was supported by the diŠerent in-

hibitory eŠects between male and female rats (Table 1) which was consistent with the fact that CYP4A2 is a male-speciˆc CYP isoform and mainly responsible for LA v-hydroxylation in male rat kidney microsomes.11,16,27,24) Therefore, the ability of anti-rCYP4A8 to inhibit LA v-hydroxylation activity was found to be due to inhibition of CYP4A8, and not due to the inhibition of CYP4A2. Accordingly, the prominent inhibition of 20-HETE formation with anti-rCYP4A8 IgG conˆrmed that CYP4A8 is contributing to 20-HETE formation in rat kidney. As the content of CYP4A8 was about 10-fold less than that of CYP4A2 in kidney microsomes from male rats,26,11) the turnover rate of CYP4A8 must be much higher than that of CYP4A2. The turnover rate of rCYP4A8 obtained here was 13.8 nmol W min W nmol CYP at 30 mM AA and is approximately 10-fold higher than rates obtained with puriˆed CYP4A2 enzyme or recombinant CYP4A2 enzymes (0.9–2.0 nmol W min W nmol CYP at 30–100 mM of AA).28,38–40) The relative contribution of CYP4A2 and CYP4A8 to 20-HETE formation, estimated from their turnover rates and relative abundances, closely correlates with the results observed in the immunoinhibition study here (Table 1). Moreover, as previously reported by Ma et al.,18) 20HETE formation activities in kidney microsomes were higher in females compared with males (Table 1), an indication that CYP4A8 was capable of the formation of physiologically relevant quantities of 20-HETE. As the antibody preparation raised against rCYP4A8 displayed high cross-reactivity with CYP4A1, there still remains the possibility of participation of CYP4A1 in renal 20HETE formation, however CYP4A1 expression levels are very low in rat kidney.15,20,23,24) Collectively, these observations suggest that CYP4A8 also plays a signiˆcant role in 20-HETE formation in rat kidney. When rat kidney segments were immunostained with anti-rCYP4A8 or antiCYP4A2 sera, there was distinct localization of stain in the proximal tubules, but not in the glomerulus or other regions (Fig. 2). These stained areas match reported renal distributions for the mRNAs for these enzymes as well as the distribution of 20HETE formation.14,25,41) In addtion, positive immunostaining in the preglomerular arteries was observed only with antiserum raised against rCYP4A8, not with anti-CYP4A2 (Fig. 2G and 2H), strongly suggesting that CYP4A8 is directly involved in 20-HETE formation in this region. On the other hand, because the antiserum raised against rCYP4A8 displays strong cross-reactivity with CYP4A1 (Fig. 1), this positive immunostaining might include CYP4A1 molecules. Speciˆc oligonucleotide probes have demonstrated that CYP4A1 expression in rat kidney is very low,15,20,23) but pretreatment with an antisense oligonucleotide to CYP4A1 reduces both renal 20-HETE formation and blood pressure in rats.19) Although CYP4A1 might be

Contribution of CYP4A8 to 20-Hydroxyeicosatetraenoic Acid Formation

localized at preglomerular arteries, its low expression level in rat kidney supports the idea that CYP4A8 is one of the predominant enzymes responsible for 20-HETE formation and the regulation of renal vasoconstriction via inhibition of the Ca+-activated potassium channels. Recently, Gebremedhin et al. reported that both CYP4A1 and CYP4A8 were expressed at a similar level in rat cerebral arterioles.42) Deˆnitive clariˆcation of the relative contribution of the CYP4A enzymes to 20HETE formation in preglomerular arteries requires further investigation. In this study, we expressed rCYP4A8 in E. coli and prepared an antibody against this enzyme. Experimental evidence showed that CYP4A8 is one of the major contributors to 20-HETE formation in rat kidney, in spite of its low content in comparison to CYP4A2. The high formation rate of 20-HETE by CYP4A8 might be su‹cient to produce enough 20-HETE to inhibit a Ca+activated potassium channel in the preglomerular arteries. Acknowledgments: We are grateful to Professor F. P. Guengerich (Vanderbilt University, Nashville, TN) providing pCW vector for the expression of rCYP4A8. Our thanks are also given to Mr. Mitsuru Notoya (Discovery Research Laboratories, Shionogi & Co., Ltd.) for his technical supporting and suggesting in the immunohistochemical study.

8) 9)

10)

11)

12)

13)

14)

15)

References 1)

2)

3)

4)

5)

6)

7)

Ma, Y.-H., Gebremedhin, D., Schwartzman, M. L., Falck, J. R., Clark, J. E., Masters, B. S., Harder, D. R. and Roman, R. J.: Altered renal P-450 metabolism of arachidonic acid in Dahl salt-sensitive rats. Circ. Res., 72: 126–136 (1993). Zou, A.-P., Fleming, J. T., Falck, J. R., Jacobs, E. R., Gebremendhin, D., Harder, D. R., and Roman, R. J.: 20-HETE is an endogenous inhibitor of the large-conductance Ca 2+-activated K+ channel inrenal arterioles. Am. J. Physiol., 270: R228–R237 (1996). Schwartzman, M., Ferrreri, N. R., Caroll, M. A., Songu-Mize, E. and McGiŠ, J. C.: Renal cytochrome P450-related arachidonate metabolite inhibits (Na++K +)ATPase. Nature, 314: 620–622 (1985). Escalante, B., Erlij, D., Falck ,J. R. and McGiŠ, J. C.: EŠect of cytochrome P450 arachidonate metabolites on ion transport in rabbit kidney loop of Henle. Science, 251: 799–802 (1991). Escalante, B., Erlij, D., Falck, J. R. and McGiŠ, J. C.: Cytochrome P-450 arachidonate metabolites aŠect ion ‰uxes in rabbit meddulary thich ascending limb. Am. J. Physiol., 266: C1775–C1782 (1994). Roman, R. J., Alonso-Gaicia, M. and Wilson, T. W.: Renal P450 metabolites of arachidonic acid and the development of hypertension on Dahl salt-sensitive rats. Am. J. Hypertens., 10: 635–675 (1997). Schwartzman, M. L. and McGiŠ, J. C.: Renal

16)

17)

18)

19)

20)

21)

22)

115

cytochrome P450. J. Lipid Mediators Cell Signalling, 12: 229–242 (1995). Simpson, A. B. C. M.: The Cytochrome P450 4 (CYP4) Family. Gen. Pharmac., 28: 351–359 (1997). Zou, A.-P., Imig, J. D., Kaldunski, M., Ortiz de Montellano, P. R., Sui, Z. and Roman, R. J.: Inhibition of renal vascular 20-HETE production impairs autoregulation of renal blood ‰ow. Am. J. Physiol., 266: F275–F282 (1994). Su, P., Kaushal, K. M. and Kroetz, D. L.: Inhibition of renal arachidonic acid v-hydroxylase activity with ABT reduces blood pressure in the SHR. Am. J. Physiol., 275: R426–R438 (1998). Imaoka, S. and Funae, Y.: Hepatic and renal cytochrome P-450s in spontaneously hypertensive rats. Biochim. Biophys. Acta, 1074: 209–213 (1991). Sacerdoti, D., Abrahan, N. G., McGriŠ, J. C. and Schwarzman, M. L.: Renal cytochrome P-450-dependent metabolism of arachidonic acid in spontaneously hypertensive rats. Biochem. Pharmacol., 37: 521–527 (1988). Omata, K., Abraham, N. G., Escalante, B. and Schwartzman, M. L.: Age-related changes in renal cytochrome P-450 arachidonic acid metabolism in spontaneously hypertensive rats. Am. J. Physiol., 262: F8–F16 (1992). Omata, K., Abraham, N. G. and Schwartzman, M. L.: Renal cytochrome P-450-arachidonic acid metabolism: localization and hormonal regulation in SHR. Am. J. Physiol., 262: F591–F599, (1992). Kroetz, D. L., Huse, L. M., Thuresson, A. and Grillo, M. P.: Developmentally regulated expression of the CYP4A genes in the spontaneously hypertensive rat kidney. Mol. Pharmacol., 52: 362–372 (1997). Sundseth, S. S. and Waxman, D. J.: Sex-dependent expression and cloˆbrate inducibility of cytochrome P450 4A fatty acid v-hydroxylase. J. Biol. Chem., 267: 3915–3921, (1992). Helvig, C., Dishman, E. and Capdevilla, J. H.: Molecular, enzymatic, and regulatory characterization of rat kidney cytochrome P450 4A2 and 4A3. Biochemistry, 37: 12546–12558 (1998). Ma, Y.-H., Schwartzman, M. L. and Roman, R. J.: Altered renal P-450 metabolism of arachidonic acid in Dahl salt-sensitive rats. Am. J. Physiol., 267: R579–R589 (1994). Wang, M.-H., Guan, H., Nguyen, X., Zand, B. A., Nasjletti, A. and Schwartzman, M. L.: Contribution of cytochrome P-450 4A1 and 4A2 to vascular 20-hydroxyeicosatetraenoic acid synthesis in rat kidneys. Am. J. Physiol., 276: F246–F253 (1999). Kimura, S., Hanioka, N., Matsunaga, E. and Gonzalez, F. J.: The rat cloˆbrate-inducible CYP4A gene subfamily I. DNA, 8: 503–516 (1989). Kimura, S., Hardwick, J. P., Kozak, C. A. and Gonzalez, F. J.: The rat cloˆbrate-inducible CYP4A gene subfamily II. DNA, 8: 517–525 (1989). Schwartzman, M. L, da Silva, J.-L., Lin F., Nishimura, M. and Abraham, N. G.: Cytochrome P450 4A expres-

116

23)

24)

25)

26)

27)

28)

29)

30)

31)

32)

33)

Yoshitaka YAMAGUCHI, et al.

sion and arachidonic acid omega-hydroxylation in the kidney of the spontaneously hypersensitive rat. Nephron, 73: 652–663 (1996). Ito, O., Alonso-Galicia, M., Hopp, A. K. and Roman, R. J.: Localization of cytochrome P-450 4A isoforms along the rat nephron. Am. J. Physiol., 274: F395–F404 (1998). Nakamura, M., Imaoka, S., Miura, K., Tanaka, E., Misawa, S. and Funae, Y.: Induction of cytochrome P450 isozymes in rat renal microsomes by cyclosporin A. Biochem. Pharmacol., 48: 1743–1746 (1994). Str äomstedt, M., Hayashi S., Zaphirophoulus, P. G. and Gustafsson, J. A.: Cloning and characterization of a novel member of the cytochrome P450 subfamily IVA in rat prostate. DNA, 9: 567–577 (1990). Imaoka, S., Nagashima, K. and Funae, Y.: Characterization of three cytochrome P450s puriˆed from renal microsomes of untreated male rats and comparison with human renal cytochrome P450. Arch. Biochem. Biophys., 276: 473–480 (1990). Imaoka, S., Nakamura, M, Ishizaki, T, Shimojo, N., Ohishi, N., Fujii, S. and Funae, Y.: Regulation of renal cytochrome P450s by thyroid hormone in diabetic rats. Biochem. Pharmacol., 46: 2197–2200 (1993). Imaoka, S., Tanaka, S. and Funae, Y.: v- and (v-1)hydroxylation of lauric acid and arachidonic acid by rat renal cytochrome P-450. Biochem. Int., 18: 731–740 (1989). Gillam, E. M. J., Baba, T., Kim, B.-R., Ohmori, S. and Guengerich, F. P.: Expression of modiˆed human cytochrome P450 3A4 in Escherichia coli and puriˆcation and reconstitution of the enzyme. Arch. Biochem. Biophys., 305: 123–131 (1993). Laemmli, U. K.: Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature, 227: 680–685 (1970). Kaminsky, L. S., Fasco, M. J. and Guengerich, F. P.: Production and application of antibodies to rat liver cytochrome P450. Methods Enzymol., 74: 262–272 (1981). Imaoka, S., Shimojo, H. and Funae, Y.: Induction of renal cytochrome P-450 in hepatic microsomes of diabetic rats. Biochem. Biophys. Res. Commun., 152: 680–687 (1988). Imaoka, S., Terano, Y. and Funae, Y: Expression of

34)

35)

36)

37)

38)

39)

40)

41)

42)

four phenobarbital-inducible cytochrome P-450s in liver, kidney, and lung of rats. J. Biochem. (Tokyo), 105: 939–945 (1989). Funae, Y and Imaoka, S.: Puriˆcation and characterization of liver microsomal cytochrome P-450 from untreated male rats. Biochim. Biophys. Acta, 926: 349–358 (1987). Kusunose, E., Kaku, M., Ichihara, K., Yamamoto, S., Yano, I. and Kusunose, M.: Hydroxylation of prostaglandin A1 by the microsomes of rabbit intestinal mucosa. J. Biochem. (Tokyo), 95: 1733–1739 (1985). Yokotani, N., Kusunose, E, Sogaya, K., Kawashima, H., Kinosaki, M., Kusunose, M., Fujii-Kuriyama, Y.: cDNA cloning and expression of the mRNA for cytochrome P-450kd which shows a fatty acid v-hydroxylating activity. Eur. J. Biochem., 196: 531–536 (1991). Omura, T. and Sato, R.: The carbon monooxide-binding pigment of liver microsomes. J. Biol. Chem., 239: 2370–2378 (1964). Hoch, U., Zhang, Z., Kroetz, D. L. and Ortiz de Montellano, P. R.: Structural determination of the substrate speciˆcities and regioselectivities of the rat and human fatty acid v-hydroxylases. Arch. Biochem. Biophys., 373: 63–71 (2000). Nguyen, X., Wang, M.-H., Reddy, K. M., Falck, J. R. and Schwartzman, M. L.: Kinetics proˆle of the rat CYP4A isoforms: arachidonic acid metabolism and isoform speciˆc inhibitors. Am. J. Physiol., 276: R1691–R1700 (1999). Wang, M. H., Stec, D. E., Balazy, M., Mastyugin, V., Yang, C. S., Roman, R. J. and Schwartzman, M. L.: Cloning, sequencing, and cDNA-directed expression of the rat renal CYP4A2: arachidonic acid v-hydroxylation and 11, 12-epoxygenation by CYP4A2 protein. Arch. Biochem. Biophys., 336: 240–250 (1996). Hardwick, J. P.: CYP4A subfamily: functional analysis by immunohistochemistry and in situ hybridyzation. Methods Enzymol., 206: 273–283 (1991). Gebremedhin, D., Lange, A. R., Lowry, T. F., Taheri, M. R., Birks, E. K., Hudetz, A. G., Narayanan, J., Falck, J. R., Okamoto, H., Roman, R. J., Nithipatikom, K., Campbell, W. B. and Harder, D. R.: Production of 20-HETE and its role in autoregulation of cerebral blood ‰ow. Circ. Res., 87: 60–65 (2000).