Application of the ferrous oxidation–xylenol orange assay for the screening of 5-lipoxygenase inhibitors

ANALYTICAL BIOCHEMISTRY Analytical Biochemistry 351 (2006) 62–68 www.elsevier.com/locate/yabio

Application of the ferrous oxidation–xylenol orange assay for the screening of 5-lipoxygenase inhibitors Young Sik Cho, Hyo Sun Kim, Chi Hyun Kim, Hyae Gyeong Cheon ¤ Laboratory of Molecular Pharmacology and Physiology, Medicinal Science Division, Korea Research Institute of Chemical Technology, Jang-dong 100, Yuseong-gu, Taejon 305-600, Republic of Korea Received 13 October 2005 Available online 13 January 2006

Abstract 5-Lipoxygenase (5-LO) is the key enzyme involved in leukotriene synthesis and its improper regulation is implicated in several inXammatory diseases. A rapid and sensitive assay for 5-LO activity suitable for high-throughput format is not yet available. In this study, we examined whether the ferrous oxidation–xylenol orange (FOX) assay could be applicable for the high-throughput screening of 5-LO inhibitors. Using insect cell lysates overexpressing rat 5-LO, the eVects of cofactors of 5-LO such as ATP, Ca2+, and L--phosphatidylcholine (PC) on the color development of FOX reagents were investigated. ATP quenched substantially color development by hydroperoxide, an intermediate of 5-LO reaction, and an optimum concentration of ATP with little interference was determined as 20 M. Ethylenediaminetetraacetate (0.4 mM) also aVected the complex formation with FOX reagents. On the other hand, neither Ca2+ nor PC inXuenced complex formation with FOX reagents. Under optimized assay conditions, zileuton, a 5-LO-speciWc inhibitor, exhibited inhibitory potency (IC50 values of 0.1–0.2 M) similar to that determined by the conventional spectrophotometric assay. Taken together, this study shows that the FOX assay with some modiWcations can be employed for high-throughput assay format for the measurement of 5LO activity at the stage of primary screening. © 2005 Elsevier Inc. All rights reserved. Keywords: FOX assay; 5-Lipoxygenase; Zileuton; High-throughput screening

5-Lipoxygenase (5-LO)1 is the Wrst enzyme involved in the synthesis of leukotrienes from arachidonic acid [1]. The resulting leukotrienes have been implicated in the pathogenesis of several inXammatory diseases, most notably asthma, psoriasis, rheumatoid arthritis, and inXammatory bowel disease [2,3]. Due to the critical role of leukotrienes as inXammatory mediators of various diseases, modulation of leukotrienes production has been an emerging approach *

Corresponding author. Fax: +82 42 861 0770. E-mail address: [email protected] (H.G. Cheon). 1 Abbreviations used: FOX, ferrous oxidation–xylenol orange; 5-LO, 5lipoxygenase; PC, L--phosphatidylcholine; 5-HPETE, 5(S)-hydroperoxy6-trans-8,11,14-cis-eicosatetraenoic acid; HTS, high-throughput screening; NDGA, nordihydroguaiaretic acid; PCR, polymerase chain reaction; PAGE, polyacrylamide gel electrophoresis; -ME, -mercaptoethanol; DTT, dithiothreitol, FLAP, Wve-lipoxygenase activating protein; MAPKAP, mitogen-activated protein kinase-activated protein kinase; SFM, serum-free medium; PVDF, polyvinylidene diXuoride. 0003-2697/$ - see front matter © 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.ab.2005.12.025

to treat related diseases. Therefore, a rapid and sensitive assay system for 5-LO is prerequisite to discover potent and novel 5-LO inhibitors. To date, available techniques to measure 5-LO activity are spectrophotometric measurement at 234 nm and determination of oxygen consumption rate [4,5]. These methods are so laborious and time consuming that they are unsuitable for high-throughput format. Thus, there has been a need for new assay systems adaptable for high-throughput screening (HTS). A ferrous oxidation–xylenol orange (FOX) assay was previously used for platelet 12-LO assay [6], and we examined in the present study whether the FOX assay is applicable for 5-LO enzyme assays in a high-throughput format. It is based on the principle that Fe3+ oxidized from Fe2+ by lipid hydroperoxide leads to stable complex with xylenol orange that absorbs strongly in the visible region. In contrary to other lipoxygenases, 5-LO requires several cofactors such as Ca2+, L--phosphatidylcholine (PC),

Ferrous oxidation–xylenol orange assay for 5-lipoxygenase / Y.S. Cho et al. / Anal. Biochem. 351 (2006) 62–68

and ATP for its optimum activity [7]. This study was aimed at the optimization of the FOX assay for HTS format of 5LO assays. To this end, insect cells overexpressing recombinant rat 5-LO were prepared as a 5-LO enzyme source. To examine the applicability of the FOX assay to HTS for 5LO inhibitors, the eVects of cofactors on color development by FOX reagents were investigated. In addition, under the optimized condition, the inhibitory potency of zileuton, a 5LO-speciWc inhibitor, was determined by the FOX assay and compared with the results of the spectrophotometric assay which measures strong absorbance at 234 nm of conjugated double bonds of arachidonic acid metabolites produced by lipoxygenase. In both assay systems, zileuton inhibited enzyme activity with IC50 values of 0.1–0.2 M. Materials and methods Materials Spodoptera frugiperda (Sf-9) cells (ATCC-CRL-1711) were obtained from American Type Culture Collection and maintained in Sf-900 II serum-free medium (SFM). Human recombinant 5-LO and polyclonal antibody against 5-LO were purchased from Cayman (Ann Arbor, MI). Zileuton was synthesized in Korea Research Institute of Chemical Technology. All the restriction enzymes were obtained from Boeringer Mannheim (Mannheim, Germany) unless otherwise stated. Nordihydroguaiaretic acid (NDGA) and ebselen were from Calbiochem (La Jolla, CA). Hinokitiol was from Wako (Osaka, Japan). Baculovirus expression systems (BAC-TO-BAC) and Sf-900 II SFM were from Invitrogen (Carlsbad, CA). All other reagents were from Sigma (St. Louis, MO). Construction of recombinant baculoviruses for rat 5-LO expression Baculoviruses expressing rat 5-LO were constructed by subcloning the cDNAs encoding full-length proteins into pFASTBAC (pFB) (Invitrogen). The cDNA for rat 5-LO was prepared by PCR ampliWcation using primers (forward primer; 5⬘-GGAATTCATGCCTTCCTACACTGTCAC C-3⬘, reverse primer; 5⬘-GCGTCGACCTACACTGTTTG GAATCTGTCTG-3⬘) under the following condition: 94 °C for 1 min, 59 °C for 1 min, and then 72 °C for 4.5 min with 30 cycles. PCR products digested with EcoRI and SalI were ligated into pFB with cutting hanger at both ends. The following procedures for the expression of recombinant 5LO in Sf-9 cells were carried out according to the protocol provided from the supplier (Invitrogen). Polyacrylamide gel electrophoresis and immunoblots Total lysates from insect cells harboring 5-LO expression vector were prepared by sonication (Sonics Vibro Cell, setting at 40 amplitude) for two pulsatile 1-s bursts in an ice/water bath. The lysates were then centrifuged at 15,000g

63

for 30 min to remove the supernatants, with which enzyme assays were carried out. Protein samples (50 g) of either rat 5-LO or commercially available human 5-LO were subjected to 10% polyacrylamide gel electrophoresis according to the method of Laemmli [8]. For immunoblot analysis, resolved proteins on PAGE were electrophoretically transferred to polyvinylidene diXuoride (PVDF) and probed with polyclonal anti-5-lipoxygenase antiserum. Spectrophotometric assay The kinetics of 5(S)-hydroperoxy-6-trans-8,11,14-ciseicosatetraenoic acid (5-HPETE) production was measured by the increase in absorbance at 234 nm upon incubation of the lysates with arachidonic acid, as previously described for the assay of porcine 5-LO [4]. The reaction was measured at room temperature in 50 mM Tris–HCl buVer (pH 7.4) containing 0.4 mM CaCl2, 24 g/ml PC, 40 M arachidonic acid, and 0.2 mM ATP and was initiated by the addition of enzyme. Unless otherwise stated, amounts of enzyme added into the reaction buVer were 0.4 unit (U) and 30 g for human and rat 5-LO, respectively. The optimal velocity of the reaction was determined from the rate of conjugated diene formation (A234) using an  D 23,000 M¡1 cm¡1 for 5-HPETE [9], and its enzyme activity is expressed as pmol of 5-HPETE formed/min/l. FOX assay FOX assays were carried out with modiWcations of the previously reported method [6]. The FOX assay was based on the complex formation of Fe3+/xylenol orange with absorption at visible light. The absorption spectra of mixtures of FOX reagents and 5-HPETE were scanned under visible light ranging from 350 to 800 nm. The insect lysates expressing rat 5-LO or commercially available human 5LO were preincubated with lipoxygenase inhibitors in 50 l of 50 mM Tris–HCl buVer (pH 7.4) containing 0.4 mM CaCl2, 24 g/ml PC, and 20 M ATP at room temperature for 4 min. The reactions were started by the addition of 40 M arachidonic acid and kept for another 4 min. The assay was terminated by the addition of 100 l FOX reagent: sulfuric acid (25 mM), xylenol orange (100 M), iron(II) sulfate (100 M), methanol:water (9:1). Blank assay was done without substrate during incubation but substrate was added after termination. For H2O2-mediated FOX reaction, FOX reagents were added to various concentrations of H2O2 to produce color change. To examine the eVect of ATP on FOX assay measurements, increasing concentrations of ATP were added either before or after the addition of FOX reagent. Also, either reducing agents or chelating agents were introduced into both assay systems to investigate their direct eVects on nonheme iron in the active site of 5-LO or their indirect eVects on the formation of Fe3+/xylenol orange complex. To determine the IC50 of zileuton for lipoxygenase, various concentrations of zileuton were preincubated with the enzyme for

64

Ferrous oxidation–xylenol orange assay for 5-lipoxygenase / Y.S. Cho et al. / Anal. Biochem. 351 (2006) 62–68

4 min under the same condition as in the spectrophotometric assay except for using ATP at 20 M. Enzyme reaction was initiated by the addition of arachidonic acid and terminated with FOX reagent for the color development. To compare the inhibitory spectra of lipoxygenase inhibitors on human and rat 5-LOs, NDGA (a general lipoxygenase inhibitor), zileuton (a 5-LO-speciWc inhibitor), hinokitiol (a 12-LO inhibitor), or ebselen (a 15-LO inhibitor) was incubated with 5-LO for 4 min, and then enzyme activities were monitored by the FOX assay. Statistical analysis Results were expressed as means § SE. Statistical signiWcance was evaluated using Student’s t test, and p < 0.05 was considered statistically signiWcant. Results Rat 5-LO enzyme preparation Baculoviruses overexpressing rat 5-LO were constructed by subcloning the cDNAs encoding full-length proteins into pFASTBAC (Fig. 1A). A cDNA for rat 5-LO was prepared by PCR ampliWcation using speciWc primers described under Materials and methods. To conWrm the expression of recombinant rat 5-LO protein, cell lysates

from Sf9 cells infected with 5-LO-expressing baculovirus were prepared, subjected to SDS–PAGE, and transferred to PVDF membrane for immunoblotting with antibody against 5-LO. A single band of recombinant rat 5-LO corresponding to MW of about 78 kDa was detected by Western blot, and its size was a little smaller than that of human recombinant 5-LO available commercially (Fig. 1B). Small aliquots (30 g) were tested for the conversion of substrate arachidonic acid to 5-HPETE by the spectrophotometric assay. Functional activity of recombinant rat 5-LO was comparable to that of human 5-LO (Fig. 1C). The enzyme activity of recombinant 5-LO was determined to be 165 pmol 5-HPETE/min/l. EVects of 5-HPETE on the FOX assay 5-LO-mediated intermediate 5-HPETE was incubated in the absence or presence of FOX reagent for 4 min, and then the resulting products were scanned at wavelengths ranging from 300 to 800 nm. A red shift in the absorption spectra of Fe3+/xylenol orange complex was made in the presence of 5-HPETE, indicating that a maximum peak was shifted from 475 to 575 nm (Fig. 2A). Intrinsic absorbance of 5-HPETE at 234 nm was correlated with increased concentrations of 5-HPETE, and complex formation with xylenol orange was also readily detected at 575 nm (Fig. 2B).

Fig. 1. Construction scheme of baculoviral vector for rat 5-LO production (A), expression of rat 5-LO in insect cell Sf-9 (B), and veriWcation of enzyme activity from insect cell lysates (C). A cDNA for rat 5-LO was subcloned into expression vector pFastBac1 to overexpress functionally active proteins in insect cells. Fifty micrograms of cell lysates were resolved on 10% SDS–PAGE, and the expression of 5-LO was examined by staining and immunoblotting. Enzyme activity was tested with 30 g of lysates prepared from 5-LO-overexpressing insect cells in a standard spectrophotometric assay. The values shown in (C) are the means § SE from three experiments done in duplicate.

Ferrous oxidation–xylenol orange assay for 5-lipoxygenase / Y.S. Cho et al. / Anal. Biochem. 351 (2006) 62–68

65

Fig. 2. Absorption spectra of complex formation of xylenol orange with Fe3+ oxidized by 5(S)-hydroperoxy-6-trans-8,11,14-cis-eicosatetraenoic acid (5HPETE) (A) and comparative absorbances of intrinsic 5-HPETE and 5-HPETE-mediated Fe3+/xylenol orange complex products determined at 234 and 575 nm, respectively (B). For absorption spectra of Fe3+/xylenol orange complex, 5-HPETE (5 M) was mixed with FOX reagent for 4 min and resulting products were scanned from 350 to 800 nm. Also, diVerent concentrations of 5-HPETE were incubated with FOX reagents for 4 min, and then complex formation between 5-HPETE-oxidized Fe3+ and xylenol orange was determined at 575 nm. For comparison, intrinsic absorbance of 5-HPETE was measured in a reaction mixture at 234 nm. Each plot represents mean § SE of three runs.

Comparison of human and rat 5-LOs in the spectrophotometric and FOX assays Both rat and human 5-LOs had similar kinetics and titration proWles in the spectrophotometric assay. Enzymes converted rapidly substrate arachidonic acid to lipid peroxide within 2 min, and enzyme reaction reached a plateau from 2 min through 4 min (Figs. 3A and B). However, the Fe3+/xylenol orange complex formation by hydroperoxides was little detected in the FOX assay under the same buVer condition as described for the spectrophotometric assay (data not shown). This result indicates that some components present in the reaction buVer would interfere with either oxidation of ferrous iron by 5LO-generated lipid hydroperoxide or complex formation of xylenol orange with Fe3+. EVects of cofactors on the FOX assay To determine which cofactors in the assay buVer might contribute to abolish 5-LO-mediated FOX reaction, each was added to enzyme reaction before the addition of FOX reagent. In a spectrophotometric assay, ATP among three cofactors was critical for stimulation of rat 5-LO with its maximum eVects observed at 0.2 mM (Fig. 4C) although other cofactors also had stimulating eVects on enzyme activity (Fig. 4A). On the other hand, ATP acted as a strong quencher of color change in the FOX assay (Fig. 4B). ATP strongly inhibited complex formation in FOX reaction in a concentration-dependent manner over the ranges 25– 200 M (Fig. 4D). To elucidate the quenching eVect of ATP on the FOX reaction mediated by lipid peroxide, we directly employed hydroperoxide to induce FOX reaction. Hydroperoxide

Fig. 3. Comparison of 5-LO enzyme activity of commercially available human 5-LO (A) and rat 5-LO (B) from insect lysates in the spectrophotometric assay. Concentration- and time-dependent enzyme activities of human and rat 5-LOs were examined in the reaction buVer containing 50 mM Tris–HCl buVer (pH 7.4), 0.4 mM CaCl2, 24 g/ml PC, 40 M arachidonic acid, and 0.2 mM ATP. Each plot represents mean § SE of three experiments.

itself rapidly catalyzed the oxidation of ferrous to ferric ion form, which in turn produced strong complex with xylenol orange (Fig. 5A). However, the addition of ATP abolished peroxide-mediated FOX reaction in a concentration-dependent manner (Fig. 5B), and it quenched color development to nearly 65% of control group even if it was added after the termination of enzyme reaction

66

Ferrous oxidation–xylenol orange assay for 5-lipoxygenase / Y.S. Cho et al. / Anal. Biochem. 351 (2006) 62–68

Fig. 4. EVects of Ca2+, PC, and ATP on rat 5-LO activity in the spectrophotometric and FOX assays. Enzyme assays were started by the addition of either CaCl2, PC, or ATP for 4 min, and lipid hydroperoxides produced by 5-LO were directly measured at 234 nm (A) or at 575 nm (B) after the generation of Fe3+/xylenol orange complex. To determine the eVect of ATP on 5-LO activity in both assays, various concentrations of ATP were added to the reaction mixture. Kinetics of 5-LO activity was obtained from the increase of absorbance at 234 nm in the presence of diVerent concentrations of ATP (C). Also, titration curves of ATP for 5-LO were plotted from the data compiled in the FOX assay. Remaining activity was expressed as the percentage of full activity shown in the presence of two cofactors such as CaCl2 and PC (D). The results are expressed on the semilog scale in x axis and represent the means § SE of three independent experiments. ¤¤p < 0.01 relative to the control activity (no ATP).

with FOX reagent. The optimized concentration of ATP in the FOX assay was determined to be 20 M, at which loss of enzyme activity was minimized in a spectrophotometric assay. Since 5-LO enzyme has a coordinated iron within the catalytic pocket, the eVects of reducing agents and chelating agents were examined for 5-LO activity. The reducing agents such as -ME and DTT did not aVect diene production from arachidonic acid by enzyme, but EDTA exhibited a little inhibition on intrinsic 5-LO enzyme activity (Fig. 6A). As for FOX assay, no inhibition was made with reducing agents while EDTA completely suppressed color development (Fig. 6B). This result implies that compounds with chelating properties such as EDTA may interfere with the FOX assay, producing false-positive results in high-throughput screening using chemical library. Kinetic proWles of 5-LO activity were measured with increasing concentrations of a speciWc 5-LO inhibitor zileuton to determine its IC50 values. Zileuton reduced the initial velocities of enzyme activities in a concentration-dependent manner with approximate IC50 values of 0.1–0.2 M for rat and human 5-LOs (Fig. 7A). Comparable inhibition of 5LO activity by zileuton was observed using the FOX assay (Fig. 7B). In addition, both rat and human 5-LO enzymes were inhibited in a similar extent by other lipoxygenase inhibitors such as NDGA, hinokitiol, and ebselen in the FOX assay system (Fig. 7C).

Discussion The present study was carried out to evaluate whether the FOX assay is sensitive, rapid, and reproducible for the measurement of 5-LO activity. Based on the present results, the FOX assay with some modiWcations can be employed as a primary assay for detecting 5-LO activity, and this Wnding would provide important means for the identiWcation of novel 5-LO inhibitors. To express functional rat 5-LO, baculoviral vector was constructed and transfected into insect cells. The expression of 5-LO in insect cells appears to have advantages of high product yield and posttranslation modiWcations over bacteria or yeast expression systems, which have been successfully used to obtain functionally active proteins [10,11]. Using baculovirus expression system in the present study, rat 5-LO with molecular weight around 78 kDa was highly expressed and catalyzed the conversion of arachidonic acid to 5-HPETE with linear increase of absorbance at 234 nm untill 4 min. Based on the principle that hydroperoxide-mediated ferric iron (Fe3+) can produce complex formation with xylenol orange, it would be theoretically possible that the FOX assay once designed for HTS of 12-LO inhibitors could be applicable to assay for 5-LO. In fact, 5-HPETE itself, an intermediate produced by 5-LO, had a peak absorption at 475 nm, and, once 5-HPETE was mixed with FOX reagents, peak absorption was red-shifted to 575 nm, implying that 5-

Ferrous oxidation–xylenol orange assay for 5-lipoxygenase / Y.S. Cho et al. / Anal. Biochem. 351 (2006) 62–68

67

Fig. 6. EVects of reducing agents and EDTA on the spectrophotometric and FOX assays. After rat 5-LO was preincubated with reducing agents ME or DTT or a chelating agent EDTA, 40 M arachidonic acid was added to trigger the enzyme reaction and allowed to incubate for 4 min. Enzyme activity was measured with times at 234 nm (A) or at 575 nm (B) after termination of the reaction with the addition of FOX reagent. Values represent the mean § SE of three replicates. Fig. 5. Titration curve of FOX reaction mediated by H2O2 (A) and quenching eVect of ATP on H2O2-mediated Fe3+/xylenol orange complex formation (B). The increasing concentrations of H2O2 were added to the reaction buVer and subsequently peroxide-mediated color development was started by the addition of FOX reagent. Various concentrations of ATP were included in H2O2-driven Fe3+/xylenol orange complex reaction and then changes in the absorbance of Fe3+/xylenol orange complex at 575 nm were monitored on a microplate reader (Synergy HT). The results are expressed on the semilog scale in x axis and represent the means § SE (n D 3). $$p < 0.005 versus control (no H2O2), ¤p < 0.05; ¤¤p < 0.005 versus control (no ATP).

LO activity could be measured indirectly by color change from yellow to blue. Absorbance of Fe3+/xylenol orange complex mediated by 5-HPETE was higher than that of intrinsic 5-HPETE over all the ranges used. Also, the color developed was stable for at least 1 h, consistent with the previous report using 12-HPETE [6]. The catalytic activity of rat 5-LO expressed in insect cells was comparable to that of human 5-LO in the spectrophotometric assay. On the contrary, under the same condition as in the spectrophotometric assay, production rate of 5HPETE catalyzed by 5-LO was not correlated with the complex formation of xylenol orange/Fe3+. While Ca2+, ATP, and PC were required for the maximum activity of 5LO in the spectrophotometric assay [7], ATP had suppressive eVects on the FOX assay, and the quenching eVect of ATP on FOX reaction seems to be irrespective of lipid hydroperoxide or H2O2. It was previously proposed that Fe3+ could bind to the N1 and N7 sites of ATP base, depending on the pH [12]. Therefore, ATP might selectively bind to Fe3+ oxidized by enzyme or chemical, thereby preventing complex formation between xylenol orange and Fe3+.

The active site of 5-LO contains nonheme iron so that it is liable to oxidoreduction or chelation by chemicals. In the present study, both -ME and DTT did not aVect enzyme activity in both assays, while EDTA (0.4 mM) strongly suppressed FOX reaction. EDTA is thought to function as a direct chelator for Fe2+ or as a remover of Fe3+ present in FOX reagents, implying that compounds with chelating properties such as EDTA may produce false-positive results in high-throughput screening. In addition to factors examined in the present study, the eVects of other factors such as FLAP [13], posttranslational modiWcation by p38 kinasedependent MAPKAP kinases [14], and translocation to the nuclear envelope [15] should be further investigated. Zileuton, a representative 5-LO inhibitor, inhibited enzyme activity in a similar extent in both assays with IC50 values of approximately 0.1–0.2 M. The inhibitory spectra of other lipoxygenase inhibitors including 12-LO and 15LO were similar irrespective of rat or human 5-LO enzymes using the FOX assays. From the deduced primary structure, rat 5-LO encodes a 673-amino-acid protein with a calculated molecular weight of 77,839 and shares around 93% amino acid homology with human 5-LO [16]. Therefore, highly expressed rat 5-LO could replace human enzyme for the primary screening of inhibitors. Although FOX assay is based on the spectra shift rather than the appearance of color against colorless blank, the color change occurred very rapidly within 4 min and remained stable at room temperature after termination. The color development was linear up to 10 M 5-HPETE used in this work, and Fe3+/ xylenol orange complex driven by 5-HPETE displayed higher absorbance than did intrinsic 5-HPETE.

68

Ferrous oxidation–xylenol orange assay for 5-lipoxygenase / Y.S. Cho et al. / Anal. Biochem. 351 (2006) 62–68

Fig. 7. Determination of IC50 values of zileuton for human and rat 5-LOs by the spectrophotometric assay (A) and the FOX assay (B) and the eVects of 5LO-, 12-LO-, and 15-LO-speciWc inhibitors against human and rat 5-LOs in the FOX assay (C). To calculate IC50 of zileuton, titration curves of zileuton for 5-LO inhibition were plotted from the spectrophotometric and FOX assays, respectively. To examine the eVects of lipoxygenase isozyme inhibitors on human and rat 5-LOs, each enzyme was preincubated with 5-LO-speciWc inhibitor zileuton (Zil), general LO inhibitor nordihydroguaiaretic acid (NDGA), 12-LO inhibitor hinokitiol (Hino), and 15-LO inhibitor ebselen at 1 M prior to the addition of substrate. The results are the means § SE of three experiments.

In conclusion, FOX assay oVers rapid and sensitive method for measuring 5-LO enzyme activity. The newly applied FOX assay with some modiWcations will be helpful for high-throughput screening using chemical library to discover novel scaVolds acting as 5-LO inhibitors. Acknowledgments This study was supported by grants from the 21C frontier program “The center for biological modulators” funded by Ministry of Science and Technology in Korea. References [1] Y.-Y. Zhang, J.L. Walker, A. Huang, J.F. Keaney Jr., C.B. Clish, C.N. Serhan, J. Loscalzo, Expression of 5-lipoxygenase in pulmonary artery endothelial cells, Biochem. J. 361 (2002) 267–276. [2] S.D. Brain, T.J. Williams, Leukotrienes and inXammation, Pharmacol. Ther. 46 (1990) 57–66. [3] R. Natarajan, J.L. Nadler, Lipid inXammatory mediators in diabetic vascular disease, Arteriosclerosis Thrombosis Vascular Biol. 24 (2004) 1542–1548. [4] W.P. Beierschmitt, J.D. McNeish, R.J. GriYths, A. Nagahisa, M. Nakane, D.E. Amacher, Induction of hepatic microsomal drugmetabolizing enzymes by inhibitors of 5-lipoxygenase (5-LO): studies in rats and 5-LO knockout mice, Toxicol. Sci. 63 (2001) 15–21. [5] J. Breton, P. Keller, M. Chabot-Fletcher, L. Hillegass, W. DeWolf Jr., D. Griswold, Use of a continuous assay of oxygen consumption to evaluate the pharmacology of 5-lipoxygenase inhibitors, Prostaglandins Leukotrienes Essent. Fatty Acids 49 (1993) 929–937.

[6] N.B Waslidge, D.J. Hayes, A colorimetric method for the determination of lipoxygenase activity suitable for use in a high throughput assay format, Anal. Biochem. 231 (1995) 354–358. [7] A.W. Ford-Hutchinson, M. Gresser, R.N. Young, 5-Lipoxygenase, Annu. Rev. Biochem. 63 (1994) 383–417. [8] U.K. Laemmli, Cleavage of structural proteins during the assembly of the head of bacteriophage T4, Nature 227 (1970) 680–685. [9] M.J. Gibian, P. Vanderberg, Product yield in oxygenation of linoleate by soybean lipoxygenase: the value of the molar extinction coeYcient in the spectrophotometric assay, Anal. Biochem. 163 (1987) 343–349. [10] X. Chen, P. Reddanna, R. Reddy, R. Kidd, G. Hildenbrandt, C. Rdddy, Expression, puriWcation, and characterization of a recombinant 5-lipoxygenase from potato tuber, Biochem. Biophys. Res. Commun. 243 (1998) 438–443. [11] M. Nakamura, T. Matsmoto, M. Noguchi, I. Yamashita, M. Noma, Expression of a cDNA encoding human 5-lipoxygenase under control of the STA1 promoter in Saccharomyces cerevisiae, Gene 89 (1990) 231–237. [12] F. Du, X.A. Ma, D.F. Li, Z.R. Liao, Binding site of Fe3+ at purine of ATP as studied by NMR, J. Inorg. Biochem. 83 (2001) 101–105. [13] R.A.F. Dixon, R.E. Diehl, E. Opas, E. Rands, P.J. Vickers, J.F. Evans, J.W. Gillard, D.K. Miller, Requirement of a 5-lipoxygenase-activating protein for leukotriene synthesis, Nature 343 (1990) 282–284. [14] O. Werz, J. Klemm, B. Samuelsson, O. Radmark, 5-Lipoxygenase is phosphorylated by p38 kinase-dependent MAPKAP kinases, Proc. Natl. Acad. Sci. USA 97 (2000) 5261–5266. [15] M. Peters-Golden, T.G. Brock, Intracellular compartmentalization of leukotriene biosynthesis, Am. J. Respir. Crit. Care Med. 161 (2000) S36–S40. [16] J.M. Balcarek, T.W. Theisen, M.N. Cook, A. Varrichio, S.M. Hwang, M.W. Strohsaker, Isolation and characterization of a cDNA clone encoding rat 5-lipoxygenase, J. Biol. Chem. 263 (1988) 13937–13941.