Leukotriene A synthase activity of purified mouse skin arachidonate 8-lipoxygenase expressed in Escherichia coli

Biochimica et Biophysica Acta 1438 (1999) 131^139

Leukotriene A synthase activity of puri¢ed mouse skin arachidonate 8-lipoxygenase expressed in Escherichia coli Na Qiao, Yoshitaka Takahashi, Hiroyuki Takamatsu, Tanihiro Yoshimoto * Department of Pharmacology, Kanazawa University School of Medicine, Kanazawa 920-8640, Japan Received 15 December 1998; received in revised form 16 February 1999; accepted 19 February 1999

Abstract Mouse skin 8-lipoxygenase was expressed in COS-7 cells by transient transfection of its cDNA in pEF-BOS carrying an elongation factor-1K promoter. When crude extract of the transfected COS-7 cells was incubated with arachidonic acid, 8hydroxy-5,9,11,14-eicosatetraenoic acid was produced as assessed by reverse- and straight-phase high performance liquid chromatographies. The recombinant enzyme also reacted on K-linolenic and docosahexaenoic acids at almost the same rate as that with arachidonic acid. Eicosapentaenoic and Q-linolenic acids were also oxygenated at 43% and 56% reaction rates of arachidonic acid, respectively. In contrast, linoleic acid was a poor substrate for this enzyme. The 8-lipoxygenase reaction with these fatty acids proceeded almost linearly for 40 min. The 8-lipoxygenase was also expressed in an Escherichia coli system using pQE-32 carrying six histidine residues at N-terminal of the enzyme. The expressed enzyme was purified over 380-fold giving a specific activity of approximately 0.2 Wmol/45 min per mg protein by nickel^nitrilotriacetate affinity chromatography. The enzymatic properties of the purified 8-lipoxygenase were essentially the same as those of the enzyme expressed in COS-7 cells. When the purified 8-lipoxygenase was incubated with 5-hydroperoxy-6,8,11,14-eicosatetraenoic acid, two epimers of 6-trans-leukotriene B4 , degradation products of unstable leukotriene A4 , were observed upon high performance liquid chromatography. Thus, the 8-lipoxygenase catalyzed synthesis of leukotriene A4 from 5-hydroperoxy fatty acid. Reaction rate of the leukotriene A synthase was approximately 7% of arachidonate 8-lipoxygenation. In contrast to the linear time course of 8-lipoxygenase reaction with arachidonic acid, leukotriene A synthase activity leveled off within 10 min, indicating suicide inactivation. ß 1999 Elsevier Science B.V. All rights reserved. Keywords: Arachidonic acid; 8-Lipoxygenase; Leukotriene A synthase; Eukaryotic expression; Prokaryotic expression; (Mouse)

1. Introduction Lipoxygenase is a dioxygenase which incorporates a molecular oxygen regiospeci¢cally and stereospeAbbreviations: HPETE, hydroperoxyeicosatetraenoic acid; HETE, hydroxyeicosatetraenoic acid; LT, leukotriene; PCR, polymerase chain reaction; HPLC, high-performance liquid chromatography; NTA, nitrilotriacetate * Corresponding author. Fax: +81-76-234-4227; E-mail: [email protected]

ci¢cally into unsaturated fatty acids such as arachidonic and linoleic acids. In mammalian cells, 5-, 8-, 12- and 15-lipoxygenases have so far been identi¢ed according to the oxygenation site in arachidonic acid, and their properties were characterized [1^4]. The primary products of these lipoxygenases from arachidonic acid are 5-, 8-, 12- and 15-hydroperoxyeicosatetraenoic acids (HPETEs). In intact cells these hydroperoxy derivatives are readily reduced to hydroxyeicosatetraenoic acids (HETEs) by glutathione peroxidase or other equivalent enzymes. In

1388-1981 / 99 / $ ^ see front matter ß 1999 Elsevier Science B.V. All rights reserved. PII: S 1 3 8 8 - 1 9 8 1 ( 9 9 ) 0 0 0 3 5 - 9

BBAMCB 55424 8-4-99

132

N. Qiao et al. / Biochimica et Biophysica Acta 1438 (1999) 131^139

the 5-lipoxygenase pathway, 5-HPETE is converted to leukotriene (LT) A4 by the 5-lipoxygenase enzyme [5,6]. LTA4 is further metabolized to either LTB4 by LTA hydrolase or LTC4 by LTC synthase, which play a pivotal role in allergy and immune reactions [7,8]. On the other hand, there are no known metabolites which show a potent biological activity in other lipoxygenase pathways [1^4]. 8-Lipoxygenase activity was ¢rst detected in mouse skin treated with phorbol ester [9,10]. Recently, cDNA for the enzyme encoding 677 amino acids (V76 kDa) was cloned [11,12], and expressed in a transient mammalian cell system [11]. Reaction of mouse skin 8-lipoxygenase was initiated by removal of the pro-10 DR-hydrogen from arachidonic acid with concomitant antrafacial 8S-oxygenation [13]. As described above, the 5-lipoxygenase catalyzes both 5S-lipoxygenation to yield 5S-HPETE and its conversion to LTA4 . The latter LTA synthase reaction is initiated by abstraction of pro-10 DR-hydrogen [14], which has the same con¢guration as that of arachidonic acid to be removed by 8-lipoxygenase. These observations raise a question as to whether the 8-lipoxygenase transforms 5-HPETE to LTA4 by removing pro-10 DR-hydrogen. We clearly demonstrate, for the ¢rst time, that the puri¢ed 8-lipoxygenase expressed in Escherichia coli could catalyze this reaction. Interesting and novel features of the mouse skin 8-lipoxygenase are also presented in this study. 2. Materials and methods 2.1. Materials [1-14 C]Arachidonic acid (2.1 GBq/mmol) was purchased from Amersham International (Bucks, UK), arachidonic acid from Nu-Chek Prep (Elysian, USA), eicosapentaenoic and docosahexaenoic acids from Cayman Chemical Co. (Ann Arbor, USA), Kand Q-linolenic acids from Doosan Serdary Research Lab (Englewood Cli¡s, USA), and linoleic acid from Wako (Osaka, Japan). Ex Taq DNA polymerase was obtained from Takara Syuzo (Kyoto, Japan), glutathione peroxidase, DEAE-dextran, chloroquine, isopropyl L-D-thiogalactoside and 11,14-eicosadienoic acid from Sigma (St. Louis, USA), TA cloning kit

including plasmid pCR 2.1 from Invitrogen (Carlbad, USA), pQE-32 and nickel^nitrilotriacetate (NTA) agarose from Qiagen (Chatsworth, USA), precoated silica-gel plates for thin-layer chromatography from Merck (Darmstadt, Germany), mouse skin cDNA library and Pfu DNA polymerase from Stratagene (La Jolla, USA), restriction enzymes from Toyobo (Osaka, Japan), and dRhodamine terminator cycle sequencing kit from PE Applied Biosystems (Foster City, USA). An expression vector, pEF-BOS, having a powerful elongation factor-1K promoter [15,16], was kindly provided by Dr. S. Nagata (Osaka Bioscience Institute). AA861 was a gift from Takeda Chemical Co. (Osaka, Japan). COS-7 cells were obtained from Institute for Fermentation (Osaka, Japan), and rat basophilic leukemia (RBL-2H3) cells from JCRB cell bank of National Institute of Health Sciences (Tokyo, Japan). Oligonucleotides for polymerase chain reaction (PCR) were synthesized by Hokkaido System Science (Sapporo, Japan). 2.2. PCR cloning of mouse skin 8-lipoxygenase 8-Lipoxygenase cDNA was ampli¢ed by PCR from mouse skin cDNA library using Ex Taq DNA polymerase. The primers were designed according to the reported sequence [11] as follows: upstream, 5P-ATGGCGAAATGCAGGGTGAGAGTATCC; downstream, 5P-GATGTTAGATGGAGACACTGTTCTCAATG, producing a 2034-bp product covering entire coding sequence. The PCR was carried out for 30 cycles under the following conditions: denaturation at 94³C for 20 s, annealing and extension at 72³C for 1 min, followed by one cycle at 72³C for 3 min. Ampli¢ed DNA of the expected size was subcloned into pCR 2.1 plasmid using TA cloning kit. The sequence was con¢rmed using an ABI DNA sequencer PRISM 310 (PE Applied Biosystems, Foster City, USA) with dRhodamine terminator cycle sequencing kit. 2.3. Expression of 8-lipoxygenase in COS-7 and E. coli A full-length cDNA encoding mouse skin 8-lipoxygenase was excised from the recombinant pCR 2.1, attached by the XbaI sites at both ends, and ligated to pEF-BOS which had been digested by XbaI. The

BBAMCB 55424 8-4-99

N. Qiao et al. / Biochimica et Biophysica Acta 1438 (1999) 131^139

resultant plasmid, pBOS-8-LOX, was introduced into COS-7 cells (5 Wg/100-mm plate) by DEAE^dextran/ chloroquine method [17]. After 48 h, cells were harvested from 20 plates, and sonicated on ice for 5 s in 500 Wl of 50 mM Tris^HCl bu¡er at pH 7.4 containing 1 mM EDTA. The sonicated cells were centrifuged at 11 000Ug for 10 min, and the supernatant was used as a crude enzyme preparation. Human platelet 12-lipoxygenase [18^20] and porcine leukocyte 12-lipoxygenase [21] were also expressed in COS-7 cells by the same method. The cells transfected with parental pEF-BOS were used as `mocktransfected cells'. Alternatively, the 8-lipoxygenase containing six histidine residues at the N-terminal was expressed in E. coli. The full-length cDNA of 8-lipoxygenase attached by HindIII sites at both ends was ligated to pQE-32 at a HindIII site. The N-terminal region of 8-lipoxygenase cDNA was ampli¢ed by PCR with Pfu DNA polymerase using primers as follows: upstream, 5P-GGGGTACCCCATGGCGAAATGCAGGGTGAGAGTATCC (KpnI site and start codon as underlined); downstream, 5P-TGCAGAGAACTGAGGCTTTCATTGAGG (nucleotides at 909^936). Ampli¢ed DNA was digested with KpnI and AccI, and the fragment was inserted into the above recombinant pQE-32 from which the KpnI^ AccI region had been removed. The pQE-8-LOX was introduced into competent E. coli by standard calcium chloride method. The transformed E. coli was grown at 25³C for 15 h, and further cultured for 18 h in the presence of 0.1 mM isopropyl L-Dthiogalactoside. E. coli was harvested by centrifugation, suspended in 50 mM potassium phosphate bu¡er at pH 7.4, and sonicated ¢ve times each for 60 s on ice. The sonicated E. coli was centrifuged at 11 000Ug for 10 min and then at 105 000Ug for 60 min. A high-speed supernatant was fractionated with ammonium sulfate (20^40% saturation), and subjected to nickel-NTA column chromatography according to the manufacturer's instructions. The active fractions were concentrated by a Dia£o PM-30 membrane, and dialyzed against 50 mM potassium phosphate bu¡er at pH 7.4 containing 300 mM NaCl and 20% glycerol. Approximately 4.3 mg of the puri¢ed 8-lipoxygenase was obtained from 12-l culture. The crude and puri¢ed enzyme preparations could be stored at 370³C without appreciable loss of enzyme

133

activity for at least 6 months. The protein concentration was determined by the method of Lowry et al. using bovine serum albumin as a standard [22]. 2.4. Preparation of HPETE and HETE 12-HETE was synthesized by incubation of porcine leukocyte cytosol with 25 WM arachidonic acid in 50 mM Tris^HCl bu¡er at pH 7.4 at 30³C for 10 min [23]. Products were reduced with 1 unit/ml glutathione peroxidase and 5 mM glutathione, extracted with diethyl ether, separated by silica-gel thin-layer chromatography using a solvent system of diethyl ether/petroleum ether/acetic acid (85:15: 0.1, by volume) at 310³C. 12-HETE was extracted from the silica-gel plate, and dissolved in ethanol. 8HETE was prepared by oxidation of arachidonic acid with copper chloride and hydrogen peroxide [24], and reduced products were puri¢ed by reverseand straight-phase high-performance liquid chromatographies (HPLC). 5-HPETE was prepared using rat basophilic leukemia cells [25]. Cytosol fraction of the cells was incubated with 50 WM arachidonic acid in 50 mM Tris^HCl bu¡er at pH 7.4 containing 2 mM CaCl2 and 2 mM ATP at 24³C for 5 min. 5-HPETE was puri¢ed by thin-layer chromatography as described for 12-HETE. 15-Hydroxy-11,13-eicosadienoic acid as an internal standard for HPLC was prepared by incubation of 11,14-eicosadienoic acid with soybean lipoxygenase (type I) [26], and reduced product was puri¢ed by thin-layer chromatography. The concentration of various H(P)ETEs was determined by UV absorbance. LTA4 was prepared by saponi¢cation of LTA4 methyl ester [27]. 2.5. Enzyme reaction The crude or puri¢ed 8-lipoxygenase was incubated with 25 WM [1-14 C]arachidonic acid (0.1 WCi/ 5 nmol/5 Wl ethanol) at 37³C in 200 Wl of 50 mM Tris^HCl bu¡er at pH 7.4 with vigorous shaking. Products extracted with diethyl ether were separated by silica-gel thin-layer chromatography as described above. Radioactivity on the plate was quanti¢ed by a BAS 1000 imaging analyzer (Fujix, Tokyo, Japan). Reactions with 25 WM unlabeled arachidonic, linoleic, K-linolenic, Q-linolenic, eicosapentaenoic and docosahexaenoic acids were carried out in 200 Wl reac-

BBAMCB 55424 8-4-99

134

N. Qiao et al. / Biochimica et Biophysica Acta 1438 (1999) 131^139

Fig. 1. Expression vectors, pBOS-8-LOX for COS-7 (A) and pQE-8-LOX for E. coli (B). The coding sequence of 8-lipoxygenase cDNA was ligated to pEF-BOS or pQE-32 vector. Note that pQE-8-LOX contains six histidine residues at the N-terminal of 8-lipoxygenase.

tion mixture containing 50 mM Tris^HCl bu¡er (pH 7.4) at 37³C for 45 min. Reaction of the puri¢ed 8lipoxygenase with 25 WM 5-HPETE was performed at 37³C for 5 min. Products were reduced with 1 unit/ ml glutathione peroxidase and 5 mM glutathione at room temperature for 5 min. This step was included to minimize the degradation of unreacted 5-HPETE which might interfere with subsequent analysis. After addition of 0.5 nmol of 15-hydroxy-11,13-eicosadienoic acid as an internal standard, products were extracted with diethyl ether, evaporated, dissolved in methanol, and stored at 320³C until HPLC analysis.

droxy acids. Straight-phase HPLC was performed for the separation of 8-HETE and 12-HETE using a Nova-Pak silica column (5 Wm particle, 3.9U150 mm) with a solvent system of n-hexane/2-propanol/ acetic acid (100:0.5:0.1, by volume) at a £ow rate of 1 ml/min. The compounds were quanti¢ed with molecular coe¤cients: O = 27 000 at 235 nm for conjugated diene and O = 50 000 at 270 nm for conjugated triene [26].

2.6. HPLC analyses

3.1. cDNA cloning and expression of mouse skin 8-lipoxygenase

Reverse-phase HPLC was carried out using TSK ODS-120T column (5 Wm particle, 4.6U250 mm) with a solvent system of methanol/water/acetic acid (80:20:0.01, by volume) at a £ow rate of 1 ml/min, and absorbance at 235 nm for a conjugated diene was continuously monitored. Product pro¢le of each fatty acid was compared with that of the mock-transfected cells, and peaks observed only in the reaction with the 8-lipoxygenase-transfected cells were quanti¢ed by comparing with internal standard. For product analysis from 5-HPETE, a solvent system of methanol/water/acetic acid (75:25:0.01, by volume) was used for the better resolution of dihy-

3. Results

PCR was carried out with primers speci¢c for 8lipoxygenase and mouse skin cDNA library, and an aliquot was analyzed by agarose gel electrophoresis. An approximately 2-kb band was excised and subcloned into pCR 2.1 plasmid. The DNA sequence was identical to that of mouse skin 8-lipoxygenase as reported previously [11]. The 8-lipoxygenase cDNA was ligated to pEF-BOS, and the pBOS-8-LOX expression vector (Fig. 1A) was introduced into COS-7 cells. The crude enzyme preparation was incubated with arachidonic acid. As shown in Fig. 2A, a major product (peak I) was detected by monitoring 235 nm

BBAMCB 55424 8-4-99

N. Qiao et al. / Biochimica et Biophysica Acta 1438 (1999) 131^139

135

upon reverse-phase HPLC. Because it is di¤cult to separate 8-HETE and 12-HETE on reverse-phase HPLC, the peak I was collected, and subjected to straight-phase HPLC analysis (Fig. 2A, inset). A single peak cochromatographed with authentic 8-HETE was detected, and there was no peak of 12-HETE. Speci¢c activity of the crude enzyme was calculated to be 5.7 nmol/45 min per mg protein (Table 1). To obtain the su¤cient amount of enzyme for characterization in details, an E. coli expression system was employed. The 8-lipoxygenase cDNA was inserted to pQE-32 carrying six histidines at 5P-upstream region of the enzyme (Fig. 1B). The sequence of the resultant plasmid, pQE-8-LOX, was con¢rmed to rule out mutations during subcloning procedures. The culture of transformed E. coli was performed at 25³C, and induced by 0.1 mM isopropyl L-D-thiogalactoside. The 105 000Ug supernatant of the E. coli was subjected to ammonium sulfate fractionation and nickelNTA a¤nity chromatography. Speci¢c activities of the sonicated cells, high-speed supernatant, ammonium sulfate fraction, and puri¢ed enzyme were 0.5, 2.1, 17.3 and 192 nmol/45 min per mg protein,

Table 1 Substrate speci¢city of 8-lipoxygenase Fatty acid Arachidonic (20:4) Linoleic (18:2) K-Linolenic (18:3) Q-Linolenic (18:3) Eicosapentaenoic (20:5) Docosahexaenoic (22:6)

Crude enzyme

Puri¢ed enzyme

nmol/45 min/mg (%) nmol/45 min/mg (%) 5.7 þ 1.6 (100) 192 þ 14 (100) 1.0 þ 0.2 (17) 41 þ 4 (21) 6.1 þ 2.4 (106) 179 þ 5 (91) 2.5 þ 0.85 (43) 82 þ 7 (42) 3.4 þ 0.35 (56) 133 þ 4 (69) 5.0 þ 0.07 (86)

nmol/5 min/mg (%) Arachidonic (20:4) ND 5-HPETE ND

178 þ 3.5 (93)

nmol/5 min/mg (%) 32 þ 4.8 (100) 2.3 þ 0.5 (7)

The crude 8-lipoxygenase expressed in COS-7 cells and puri¢ed enzyme expressed in E. coli were incubated with 25 WM fatty acids at 37³C for 45 min. Reactions with 5-HPETE were carried out for 5 min (see text). Products were separated by reverse-phase HPLC, and quanti¢ed as described in Section 2. Numbers of carbon atom and double bond of fatty acid are shown in parentheses after the names. Each value represents the mean þ S.D. of four independent experiments. ND, not determined.

Fig. 2. HPLC analysis of products by the puri¢ed 8-lipoxygenase incubated with arachidonic acid or 5-HPETE. (A) Puri¢ed 8-lipoxygenase (5 Wg of protein) was incubated with 25 WM arachidonic acid at 37³C for 45 min. (B) The enzyme (40 Wg of protein) was incubated with 25 WM 5-HPETE at 37³C for 5 min. (C) Heat-inactivated enzyme was incubated with 25 WM 5-HPETE. Products were analyzed by reverse-phase HPLC with a solvent system of methanol/water/acetic acid (75:25:0.01, by volume) at a £ow rate of 1 ml/min. UV absorbance was monitored at 270 nm for 32 min and at 235 nm thereafter. Inset shows the analysis of the peak I in A by straight-phase HPLC to separate 8- and 12-HETEs. Arrows indicate retention times of authentic compounds. IS denotes 15-hydroxy-11,13-eicosadienoic acid as an internal standard.

respectively. Thus, over a 380-fold puri¢cation of 8-lipoxygenase was achieved by these steps. 3.2. Enzyme properties of 8-lipoxygenase It is well known that most mammalian lipoxygenases show a suicide inactivation during catalysis [28]. Namely, the reaction ceases within a few minutes although the su¤cient amount of the substrate

BBAMCB 55424 8-4-99

136

N. Qiao et al. / Biochimica et Biophysica Acta 1438 (1999) 131^139

remains. The exception is a 12-lipoxygenase present in platelets of various species, the reaction of which proceeds almost linearly over 30 min [28]. To characterize catalytic properties of mouse 8-lipoxygenase, we ¢rst examined reaction time course. Both crude enzyme expressed in COS-7 cells and puri¢ed enzyme expressed in E. coli were incubated with 25 WM [1-14 C]arachidonic acid at 37³C. As shown in Fig. 3A, the reaction of 8-lipoxygenase proceeded almost linearly for at least 40 min (open and closed circles). These time courses were similar to that of the human platelet 12-lipoxygenase (open squares), and completely di¡erent from porcine leukocyte 12-lipoxygenase which showed a typical suicide inactivation (closed squares). The crude enzyme preparation from mock-transfected COS-7 cells hardly reacted on arachidonic acid. We examined the substrate speci¢city of 8-lipoxygenase expressed in E. coli as well as COS-7 cells. As shown in Table 1, linoleic acid was a poor substrate; approximately 20% as active as arachidonic acid, con¢rming the previous data [11]. However, the recombinant 8-lipoxygenase reacted on K-linolenic and docosahexaenoic acids at a rate comparable to arachidonic acid, and eicosapentaenoic and Q-linolenic acids were oxygenated at about half reaction rates. It should be noted that almost the same substrate speci¢city was observed with the crude 8-lipoxygenase expressed in COS-7 cells and the puri¢ed enzyme having six histidine residues at the N-terminal. Reaction of the 8-lipoxygenases with various fatty acids proceeded in an almost linear fashion for more than 30 min as observed with arachidonic acid (data not shown). Thus, the mouse skin 8-lipoxygenase showed a broad substrate speci¢city in terms of carbon-chain length and double bonds. As for cofactor requirement of the 8-lipoxygenase, reaction rate of the crude enzyme expressed in COS-7 cells increased by 1.5-fold in the presence of 2 mM CaCl2 , but the e¡ect was not signi¢cant with the puri¢ed enzyme of E. coli. An apparent ATP stimulation was not observed with the crude or puri¢ed 8-lipoxygenase preparation. When the puri¢ed 8-lipoxygenase was incubated with arachidonic acid in the presence of 2 WM 13-hydroperoxy-9,11-octadecadienoic acid, the time course and extent of the reaction were essentially identical to those without the hydroperoxy compound (data not shown).

Fig. 3. Time courses of various lipoxygenases reacting with arachidonic acid (A) and 5-HPETE (B). (A) [1-14 C]Arachidonic acid was incubated at 37³C with the puri¢ed mouse 8-lipoxygenase expressed in E. coli (16 Wg of protein, closed circles) and crude mouse 8-lipoxygenase expressed in COS-7 cells (400 Wg, open circles). Reactions were also performed with human platelet 12-lipoxygenase (16 Wg, open squares) and porcine leukocyte 12-lipoxygenase (70 Wg, closed squares), which were expressed in COS-7 cells. Reaction products were separated by thin-layer chromatography, and radioactivity was determined by a BAS 1000 imaging analyzer. (B) The puri¢ed 8-lipoxygenase (60 Wg) was incubated with 25 WM 5-HPETE. Products were separated by reverse-phase HPLC and quanti¢ed using 15-hydroxy-11,13eicosadienoic acid as an internal standard. An arrow indicates the addition of the same amount of 5-HPETE. Data are representative of three separate experiments.

3.3. Reaction with 5-HPETE We examined whether 5-HPETE could be a substrate for the 8-lipoxygenase, because this compound is converted to LTA4 by the abstraction of a hydrogen atom from C-10 and the 8-lipoxygenase can withdraw that from the same position of arachidonic acid [13]. The puri¢ed 8-lipoxygenase was incubated with 25 WM 5-HPETE at 37³C for 5 min. Products were reduced and analyzed by reverse-phase HPLC monitoring at 270 nm for a conjugated triene. As shown in Fig. 2B, a doublet peak appeared with

BBAMCB 55424 8-4-99

N. Qiao et al. / Biochimica et Biophysica Acta 1438 (1999) 131^139

retention times of 9.7 min and 10.4 min which cochromatographed with authentic 6-trans-LTB4 and 12-epi-6-trans-LTB4 . The result indicated that the puri¢ed 8-lipoxygenase transformed 5-HPETE to unstable LTA4 which nonenzymatically degraded to the 6-trans-LTB4 epimers [29]. The reaction rate of 8-lipoxygenase with 5-HPETE was about 7% as compared with arachidonic acid (Table 1). The peak eluted at 42.8 min was collected, and the UV spectrum was examined. It showed an absorption peak at 234 nm, suggesting that it retained the diene structure of 5-HPETE. Further characterization was not performed due to a limited amount of this compound. It should be mentioned that production of this compound leveled o¡ within 10 min (data not shown) as was the case in LTA4 production from 5HPETE (see below for Fig. 3B). The heat-denatured enzyme did not convert 5-HPETE to LTs (Fig. 2C). Since conversion of 5-HPETE to LTA4 was catalyzed by 5-lipoxygenase [5,6] as well as our 8-lipoxygenase, e¡ects of a speci¢c inhibitor of 5-lipoxygenase, AA861 [30], was examined. AA861 suppressed the reaction rate of the puri¢ed 8-lipoxygenase with 5-HPETE to 40% at 20 WM and to 5% at 50 WM. Furthermore, the 8-lipoxygenase reaction of arachidonic acid was almost completely inhibited by 20 WM of AA861 (data not shown). The crude 8-lipoxygenase expressed in COS-7 cells was also incubated with 5-HPETE, but signi¢cant peaks of 6-trans-LTB4 epimers were not observed, probably due to low enzymatic activity. Fig. 3B shows the reaction time course of the puri¢ed 8-lipoxygenase with 5-HPETE. The reaction rate decreased during catalysis, and ceased within 10 min, suggesting suicide inactivation of the enzyme. To con¢rm this possibility, 5-HPETE was added to the reaction mixture after 15 min, but it did not increase the production of 6-trans-LTB4 epimers (an arrow in Fig. 3B). 4. Discussion In this study we obtained cDNA for mouse skin 8lipoxygenase by PCR, and expressed it in COS-7 cells. The speci¢c activity of the crude 8-lipoxygenase was only 3% of that of human platelet 12-lipoxygenase expressed in COS-7 cells at the same time (Table 1 and Fig. 3). In accordance with this observation,

137

Jisaka et al. reported di¤culties in transient 8-lipoxygenase expression in HEK293 cells, whereas human 15-lipoxygenase showed easily detectable activity in the same expression system [11]. The 8-lipoxygenase was also expressed in E. coli employing a pQE-32 expression vector containing six histidines at its Nterminal. The enzyme could be easily puri¢ed by nickel-NTA a¤nity chromatography to a speci¢c activity of approximately 0.2 Wmol/45 min per mg protein. It was reported that the N-terminal addition of histidine residues to porcine leukocyte 12-lipoxygenase signi¢cantly altered enzyme properties such as sensitivity to inhibitors and reactivity with linoleic acid [31,32]. Therefore, we carefully compared the properties of the puri¢ed histidine-tagged 8-lipoxygenase expressed in E. coli and the crude enzyme expressed in COS-7 cells. The reaction time courses of 8-lipoxygenase with arachidonic acid were almost linear for 40 min in the both preparations (Fig. 3A). Furthermore, these two enzymes showed the same order of relative activity toward various unsaturated fatty acids (Table 1). These observations indicated that the addition of six histidines to N-terminal of 8-lipoxygenase did not signi¢cantly change the catalytic properties. A reason for the di¡erences in e¡ect of the N-terminal histidine addition between porcine leukocyte 12-lipoxygenase and mouse skin 8-lipoxygenase is unknown, but will be an interesting issue in protein chemistry of lipoxygenases. The reactions with arachidonic acid of mouse skin 8-lipoxygenase and human platelet 12-lipoxygenase proceeded almost linearly over 40 min (Fig. 3A). This time course is of interest in a sense that many other mammalian lipoxygenases show catalytic autoinactivation [28] as exempli¢ed by porcine leukocyte 12-lipoxygenase shown in Fig. 3A. In contrast, time course of the 8-lipoxygenase with 5-HPETE showed a typical suicide inactivation (Fig. 3B). As for mechanisms for turnover-dependent inactivation, human leukocyte 5-lipoxygenase was inactivated by covalent binding to LTA4 formed during catalysis [33]. It was also suggested that 14,15-LTA4 inactivated porcine 12-lipoxygenase by covalent modi¢cation by this compound [34]. By analogy to the proposed mechanism for catalytic autoinactivation, the inactivation of 8-lipoxygenase during the reaction with 5-HPETE might be mediated by LTA4 or its derivatives. In fact, the puri¢ed 8-lipoxygenase was preincubated

BBAMCB 55424 8-4-99

138

N. Qiao et al. / Biochimica et Biophysica Acta 1438 (1999) 131^139

at 37³C for 5 min with 5 or 10 WM LTA4 , the enzyme activity decreased to 68 þ 3% or 43 þ 4%, respectively (n = 3, data not shown). The possibility that LTA4 covalently binds to the 8-lipoxygenase is currently under investigation. Interestingly, the broad substrate speci¢city of the mouse skin 8-lipoxygenase expressed in COS-7 cells and E. coli was elucidated in our study (Table 1). However, linoleic acid was a poor substrate for the enzyme, being consistent with the previous ¢ndings [11]. In contrast, linoleic acid was e¤ciently oxygenated by 12- and 15-lipoxygenases from various species [28]. Such a broad substrate speci¢city of the 8lipoxygenase suggests that arachidonic acid would not be the only natural substrate for the enzyme. From this point of view, we examined 5-HPETE as a possible substrate for the 8-lipoxygenase, and found that the puri¢ed enzyme catalyzed transformation of 5-HPETE to LTA4 . It was previously demonstrated that human and mouse 5-lipoxygenases exhibited 8S-lipoxygenase activity when 8,11,14eicosatrienoic acid was used as a substrate [35,36]. Therefore, it is conceivable that reaction mechanism on LTA synthase catalyzed by both 8-lipoxygenase and 5-lipoxygenase are the same in terms of hydrogen abstraction. In accordance to this notion, AA861 at a 20 WM concentration, a speci¢c inhibitor of 5lipoxygenase, suppressed almost completely our 8lipoxygenases expressed in COS-7 cells and E. coli (data not shown). These ¢ndings, taken together, suggest that the active centers of 8-lipoxygenase and 5-lipoxygenase are of similar secondary structure. The relative reaction rate of LTA synthase activity of the puri¢ed 8-lipoxygenase (7% of arachidonic acid oxygenation) was slightly greater than that previously reported for porcine leukocyte 5-lipoxygenase (2^3%) [37]. Considering LT biosynthesis in epidermis, it is of interest that 5-lipoxygenase was expressed in mouse epidermis [38]. Since 8-lipoxygenase was an inducible enzyme in epidermis, it may regulate the synthesis of LTA4 from 5-HPETE produced by the 5-lipoxygenase in mouse skin under certain pathophysiological circumstances. Acknowledgements We are indebted to Dr. Y. Ashida of Takeda

Chemical Company for the generous gift of AA861. This work was supported by grants-in-aid from the Ministry of Education, Science and Culture of Japan, and grants from Ono Pharmaceutical Company, the Mochida Memorial Foundation for Medical and Pharmaceutical Research, the Ichiro Kanehara Foundation, and the Kisshokai Foundation.

References [1] S. Yamamoto, Mammalian lipoxygenases: molecular structures and functions, Biochim. Biophys. Acta 1128 (1992) 117^131. [2] T. Yoshimoto, S. Yamamoto, Arachidonate 12-lipoxygenase, J. Lipid Mediat. Cell Signal. 12 (1995) 195^212. [3] C.D. Funk, The molecular biology of mammalian lipoxygenases and the quest for eicosanoid functions using lipoxygenase-de¢cient mice, Biochim. Biophys. Acta 1304 (1996) 65^84. [4] E. Sigal, The molecular biology of mammalian arachidonic acid metabolism, Am. J. Physiol. 260 (1991) L13^28. [5] C.A. Rouzer, E. Rands, S. Kargman, R.E. Jones, R.B. Register, R.A. Dixon, Characterization of cloned human leukocyte 5-lipoxygenase expressed in mammalian cells, J. Biol. Chem. 263 (1988) 10135^10141. [6] C.D. Funk, H. Gunne, H. Steiner, T. Izumi, B. Samuelsson, Native and mutant 5-lipoxygenase expression in a baculovirus/insect cell system, Proc. Natl. Acad. Sci. USA 86 (1989) 2592^2598. [7] B. Samuelsson, S.E. Dahlen, J.A. Lindgren, C.A. Rouzer, C.N. Serhan, Leukotrienes and lipoxins: structures, biosynthesis, and biological e¡ects, Science 237 (1987) 1171^1176. [8] A.W. Ford Hutchinson, Regulation of leukotriene biosynthesis, Cancer Metastas. Rev. 13 (1994) 257^267. [9] M. Gschwendt, G. Furstenberger, W. Kittstein, E. Besemfelder, W.E. Hull, H. Hagedorn, H.J. Opferkuch, F. Marks, Generation of the arachidonic acid metabolite 8-HETE by extracts of mouse skin treated with phorbol ester in vivo; identi¢cation by 1 H-n.m.r. and GC-MS spectroscopy, Carcinogenesis 7 (1986) 449^455. [10] G. Furstenberger, H. Hagedorn, T. Jacobi, E. Besemfelder, M. Stephan, W.D. Lehmann, F. Marks, Characterization of an 8-lipoxygenase activity induced by the phorbol ester tumor promoter 12-O-tetradecanoylphorbol-13-acetate in mouse skin in vivo, J. Biol. Chem. 266 (1991) 15738^15745. [11] M. Jisaka, R.B. Kim, W.E. Boeglin, L.B. Nanney, A.R. Brash, Molecular cloning and functional expression of a phorbol ester-inducible 8S-lipoxygenase from mouse skin, J. Biol. Chem. 272 (1997) 24410^24416. [12] P. Krieg, A. Kinzig, M. Heidt, F. Marks, G. Furstenberger, cDNA cloning of a 8-lipoxygenase and a novel epidermistype lipoxygenase from phorbol ester-treated mouse skin, Biochim. Biophys. Acta 1391 (1998) 7^12.

BBAMCB 55424 8-4-99

N. Qiao et al. / Biochimica et Biophysica Acta 1438 (1999) 131^139 [13] M.A. Hughes, A.R. Brash, Investigation of the mechanism of biosynthesis of 8-hydroxyeicosatetraenoic acid in mouse skin, Biochim. Biophys. Acta 1081 (1991) 347^354. [14] R.L. Maas, C.D. Ingram, D.F. Taber, J.A. Oates, A.R. Brash, Stereospeci¢c removal of the DR hydrogen atom at the 10-carbon of arachidonic acid in the biosynthesis of leukotriene A4 by human leukocytes, J. Biol. Chem. 257 (1982) 13515^13519. [15] T. Uetsuki, A. Naito, S. Nagata, Y. Kaziro, Isolation and characterization of the human chromosomal gene for polypeptide chain elongation factor-1 alpha, J. Biol. Chem. 264 (1989) 5791^5798. [16] S. Mizushima, S. Nagata, pEF-BOS, a powerful mammalian expression vector, Nucleic Acids Res. 18 (1990) 5322. [17] M.A. Lopata, D.W. Cleveland, W.B. Sollner, High level transient expression of a chloramphenicol acetyl transferase gene by DEAE-dextran mediated DNA transfection coupled with a dimethyl sulfoxide or glycerol shock treatment, Nucleic Acids Res. 12 (1984) 5707^5717. [18] C.D. Funk, L. Furci, G.A. FitzGerald, Molecular cloning, primary structure, and expression of the human platelet/ erythroleukemia cell 12-lipoxygenase, Proc. Natl. Acad. Sci. USA 87 (1990) 5638^5642. [19] T. Izumi, S. Hoshiko, O. Radmark, B. Samuelsson, Cloning of the cDNA for human 12-lipoxygenase, Proc. Natl. Acad. Sci. USA 87 (1990) 7477^7481. [20] T. Yoshimoto, Y. Yamamoto, T. Arakawa, H. Suzuki, S. Yamamoto, C. Yokoyama, T. Tanabe, H. Toh, Molecular cloning and expression of human arachidonate 12-lipoxygenase, Biochem. Biophys. Res. Commun. 172 (1990) 1230^ 1235. [21] T. Yoshimoto, H. Suzuki, S. Yamamoto, T. Takai, C. Yokoyama, T. Tanabe, Cloning and sequence analysis of the cDNA for arachidonate 12-lipoxygenase of porcine leukocytes, Proc. Natl. Acad. Sci. USA 87 (1990) 2142^2146. [22] O.H. Lowry, N.J. Rosenberg, A.L. Farr, A.L. Randall, Protein measurement with the folin phenol reagent, J. Biol. Chem. 193 (1951) 265^275. [23] T. Yoshimoto, Y. Miyamoto, K. Ochi, S. Yamamoto, Arachidonate 12-lipoxygenase of porcine leukocyte with activity for 5-hydroxyeicosatetraenoic acid, Biochim. Biophys. Acta 713 (1982) 638^646. [24] J.M. Boeynaems, J.A. Oates, W.C. Hubbard, Preparation and characterization of hydroperoxy-eicosatetraenoic acids (HPETEs), Prostaglandins 19 (1980) 87^97. [25] M. Furukawa, T. Yoshimoto, K. Ochi, S. Yamamoto, Studies on arachidonate 5-lipoxygenase of rat basophilic leukemia cells, Biochim. Biophys. Acta 795 (1984) 458^465. [26] M. Claeys, G.A. Kivits, E. Christ Hazelhof, D.H. Nugteren, Metabolic pro¢le of linoleic acid in porcine leukocytes through the lipoxygenase pathway, Biochim. Biophys. Acta 837 (1985) 35^51.

139

[27] T. Yoshimoto, R.J. Soberman, R.A. Lewis, K.F. Austen, Isolation and characterization of leukotriene C4 synthetase of rat basophilic leukemia cells, Proc. Natl. Acad. Sci. USA 82 (1985) 8399^8403. [28] S. Yamamoto, H. Suzuki, N. Ueda, Arachidonate 12-lipoxygenases, Prog. Lipid Res. 36 (1997) 23^41. [29] P. Borgeat, B. Samuelsson, Arachidonic acid metabolism in polymorphonuclear leukocytes: unstable intermediate in formation of dihydroxy acids, Proc. Natl. Acad. Sci. USA 76 (1979) 3213^3217. [30] T. Yoshimoto, C. Yokoyama, K. Ochi, S. Yamamoto, Y. Maki, Y. Ashida, S. Terao, M. Shiraishi, 2,3,5-Trimethyl-6(12-hydroxy-5,10-dodecadiynyl)-1,4-benzoquinone (AA861), a selective inhibitor of the 5-lipoxygenase reaction and the biosynthesis of slow-reacting substance of anaphylaxis, Biochim. Biophys. Acta 713 (1982) 470^473. [31] R.G. Reddy, T. Yoshimoto, S. Yamamoto, C.D. Funk, L.J. Marnett, Expression of porcine leukocyte 12-lipoxygenase in a baculovirus/insect cell system and its characterization, Arch. Biochem. Biophys. 312 (1994) 219^226. [32] R.G. Reddy, T. Yoshimoto, S. Yamamoto, L.J. Marnett, Expression, puri¢cation, and characterization of porcine leukocyte 12-lipoxygenase produced in the methylotrophic yeast, Pichia pastoris, Biochem. Biophys. Res. Commun. 205 (1994) 381^388. [33] R.A. Lepley, F.A. Fitzpatrick, Irreversible inactivation of 5lipoxygenase by leukotriene A4 . Characterization of product inactivation with puri¢ed enzyme and intact leukocytes, J. Biol. Chem. 269 (1994) 2627^2631. [34] K. Kishimoto, M. Nakamura, H. Suzuki, T. Yoshimoto, S. Yamamoto, T. Takao, Y. Shimonishi, T. Tanabe, Suicide inactivation of porcine leukocyte 12-lipoxygenase associated with its incorporation of 15-hydroperoxy-5,8,11,13-eicosatetraenoic acid derivative, Biochim. Biophys. Acta 1300 (1996) 56^62. [35] T. Shimizu, T. Izumi, Y. Seyama, K. Tadokoro, O. Radmark, B. Samuelsson, Characterization of leukotriene A4 synthase from murine mast cells: evidence for its identity to arachidonate 5-lipoxygenase, Proc. Natl. Acad. Sci. USA 83 (1986) 4175^4179. [36] P. Borgeat, M. Hamberg, B. Samuelsson, Transformation of arachidonic acid and homo-Q-linolenic acid by rabbit polymorphonuclear leukocytes, J. Biol. Chem. 251 (1976) 7816^ 7820. [37] N. Ueda, S. Kaneko, T. Yoshimoto, S. Yamamoto, Puri¢cation of arachidonate 5-lipoxygenase from porcine leukocytes and its reactivity with hydroperoxyeicosatetraenoic acids, J. Biol. Chem. 261 (1986) 7982^7988. [38] C.D. Funk, D.S. Keeney, E.H. Oliw, W.E. Boeglin, A.R. Brash, Functional expression and cellular localization of a mouse epidermal lipoxygenase, J. Biol. Chem. 271 (1996) 23338^23344.

BBAMCB 55424 8-4-99