Cholesterol and its anionic derivatives inhibit 5-lipoxygenase activation in polymorphonuclear leukocytes and MonoMac6 cells Dmitry A. Aleksandrov 1 , Anna N. Zagryagskaya 1 , Marina A. Pushkareva 1 , Markus Bachschmid 2 , Marc Peters-Golden 3 , Oliver Werz 4 , Dieter Steinhilber 4 and Galina F. Sud’ina 1 1 A.N. Belozersky Institute of Physicochemical Biology, Moscow State University, Russia 2 Faculty of Biology, University of Konstanz, Germany 3 Pulmonary and Critical Care Medicine Division, University of Michigan, Ann Arbor, USA 4 Institute of Pharmaceutical Chemistry ⁄ ZAFES, University of Frankfurt, Germany Mammalian 5-lipoxygenase (5-LO) converts arachi- donic acid (AA) to 5-hydroperoxyeicosatetraenoic acid (5-HPETE) and further to leukotriene A4 (LTA4) [1]. This substance is a key intermediate in the biosynthesis of two leukotriene families that act as potent media- tors of cell proliferation [2], apoptosis [3], tumorigen- esis and inflammatory processes such as allergy, atherosclerosis and asthma [4,5]. 5-LO is activated by intracellular Ca 2+ influx that leads to translocation and binding of 5-LO to the nuclear membrane [6–11]. It is commonly observed that the N-terminal domain of the enzyme may function like the protein kinase C C2 domain and facilitates Ca 2+ -mediated membrane binding [12–16]. There are two main physicochemical factors influen- cing translocation of 5-LO to the membrane. First, the calcium-ion binding counteracts electrostatic repulsion of anionic C2 domain of 5-LO at the surface of ani- onic membranes [15]. Secondly, it may be regulated by membrane surface charge and membrane lipids. It was shown that phosphatidyl choline (PC) head groups facilitate the 5-LO C2-like domain binding to mem- branes, and that Trp13, Trp75, Trp102 of 5-LO are involved in these protein–lipid interactions [15]. The Keywords 5-lipoxygenase; cholesterol sulfate; leukotrienes; neutrophil Correspondence G. F. Sud’ina, A.N. Belozersky Institute of Physicochemical Biology, Moscow State University, Russia Fax: + 7 095 9393181 Tel: +7 095 9393184 E-mail: sudina@genebee.msu.ru (Received 19 September 2005, revised 23 November 2005, accepted 5 December 2005) doi:10.1111/j.1742-4658.2005.05087.x 5-Lipoxygenase (5-LO) is the key enzyme in the biosynthesis of leukotrie- nes (LTs), biological mediators of host defense reactions and of inflamma- tory diseases. While the role of membrane binding in the regulation of 5-LO activity is well established, the effects of lipids on cellular activity when added to the medium has not been characterized. Here, we show such a novel function of the most abundant sulfated sterol in human blood, cho- lesterol sulfate (CS), to suppress LT production in human polymorpho- nuclear leukocytes (PMNL) and Mono Mac6 cells. We synthesized another anionic lipid, cholesterol phosphate, which demonstrated a similar capacity in suppression of LT synthesis in PMNL. Cholesteryl acetate was without effect. Cholesterol increased the effect of CS on 5-LO product synthesis. CS and cholesterol also inhibited arachidonic acid (AA) release from PMNL. Addition of exogenous AA increased the threshold concentration of CS required to inhibit LT synthesis. The effect of cholesterol and its ani- onic derivatives can arise from remodeling of the cell membrane, which interferes with 5-LO activation. The fact that cellular LT production is regulated by sulfated cholesterol highlights a possible regulatory role of sulfotransferases ⁄ sulfatases in 5-LO product synthesis. Abbreviations AA, arachidonic acid; CA, cholesterol acetate; CP, cholesterol phosphate; CS, cholesterol sulfate; HBSS, Hank’s balanced salt solution; 5-HETE, 5-hydroxyeicosatetraenoic acid; LT, leukotrienes; LTB4, leukotriene B4; iso-LTB4, 5(S),12(S,R)-dihydroxy-all-trans-eicosatetraenoic acids; 5-LO, 5-lipoxygenase; MbCD, methyl-b-cyclodextrin; PMNL, polymorphonuclear leukocyte; ROS, reactive oxygen species. 548 FEBS Journal 273 (2006) 548–557 ª 2006 The Authors Journal compilation ª 2006 FEBS specific role of tryptophan (Trp) residues in facilitating membrane binding has been documented [17]. At phy- siological temperature the PC-lipid is preferably distri- buted to perinuclear region of cells [18]. In cell-free assays with lipids in the form of large unilamellar vesicles, it was found that an increase in the molar percentage of cationic lipid 1,2-dimyristoyl- glycero-3-ethylphosphocholine increased, but that the anionic lipid 1-palmitoyl-2-oleoyl-glycero-3-phospho- glycerol decreased 5-LO activity [19]. Just recently, it was published by these authors that membrane fluidity is a key modulator of membrane binding and activity of 5-lipoxygenase [20]. Earlier, it was demonstrated by our group that cellular 5-LO activity is suppressed by anionic lipid sulfatides [21]. To further test the role of anionic lipids as 5-LO inhibitors, we investigated ani- onic derivatives of cholesterol. Cholesterol and its sul- fated product cholesterol sulfate (CS) are also known as agents that increase membrane rigidity. Being an important constituent of cell membranes, free cholesterol is converted to its ester in blood plasma. In view of the role of cholesterol lipids in atherogenesis, we sought to determine the influence of cholesterol and its derivatives on cellular 5-LO activity. We studied CS, cholesterol phosphate (CP) and choles- terol acetate (CA), the latter being chosen as a refer- ence ester compound. The 5-lipoxygenase pathway has recently been implicated as an important component in the pathogenesis of atherosclerosis [22]. It is likely that neutrophils and monocytes largely dictate the levels of LTs in the vasculature, and these mediators play an important role in leukocyte adhesion to vessel walls, ROS production and atherothrombosis. The regulation of LT synthesis by cholesterol and its derivatives in polymorphonuclear leukocytes (PMNL) has not been previously characterized. Among the two anionic cholesterol derivatives inves- tigated here, CS is a natural compound. CS is synthes- ized by the sulfotransferase SULT2B1b from free cholesterol. It is quantitatively the most important and abundant sterol sulfate in human plasma, where it is present at concentrations of about 3 lgÆmL )1 [23]. The concentration of CS may vary widely during disease states [24,25]. Infiltration of neutrophils and monocytes into gastric mucosa is a hallmark of chronic gastritis caused by Helicobacter pylori, with neutrophil stimula- tion that results in the damage of the gastric epithe- lium. Human gastric fluid and gastric epithelia contain up to 500 lg sulfatides and 700 lg CS per gram of dry weight [26]. CS exhibits a gastroprotective activity after administration of sulfolipid-containing liposomes [26]. As a component of cell membranes, CS plays a stabil- izing role [27], prevents osmotic lysis and supports cell adhesion [25]. CS is known to regulate the activity of serine proteases, especially protein kinase C isoforms [28], phosphatidylinositol-3-kinase [29], chymotrypsin [30], pancreatic elastase [31] and DNase I [32]. CS also regulates transcription of transglutaminase I [33]. Recently, it was shown that CS is a specific ligand for retinoic acid related orphan receptor alpha (RORa) [34]. In this work, we report that CS suppresses LT syn- thesis in two cell types, PMNL and MM6. Results Free cholesterol and its anionic derivatives inhibit cellular 5-LO product formation We determined the effect of free cholesterol, CS, CP and CA on LT production in whole cells. Isolated human PMNL were collected and suspended in PGC buffer with different cholesterol derivatives. This mix- ture was preincubated for 30 min at 37 °C and leuko- triene production was stimulated for 10 min with 2 lm A23187 calcium ionophore. The formation of LTB4, its trans- and epi-trans-isomers and 5-HETE was deter- mined. The data are presented in Fig. 1. The anionic derivative CS inhibited LT synthesis at concentrations as low as 10 lgÆmL )1 (Fig. 1A). Cholesterol had no significant effect at 10 lgÆmL )1 , but inhibited at 25 lgÆmL )1 (Fig. 1B). Both lipids were inactive in the presence of the cholesterol acceptor methyl-b-cyclodex- trin (MbCD) (Fig. 1C). MbCD at 1 mm had no effect, but at the optimal concentration of 2 mm, it signifi- cantly increased AA release and LT synthesis. The synthetic anionic cholesterol derivative CP was an even stronger inhibitor of LT synthesis than CS (Fig. 1B). The naturally occurring anionic lipid CS was analyzed in more detail. Combined exposure to cholesterol and CS at 25 lgÆmL )1 of lipids did not have a greater effect on 5-LO product synthesis than the exposure to 25 lgÆmL )1 of CS alone (Fig. 2). How- ever, when added with CS and AA, cholesterol signifi- cantly suppressed 5-LO product synthesis (lower panel), resulting in no difference between the sum of 5-LO products in the absence and presence of AA at 25 lgÆmL -1 CS + 25 lgÆmL )1 cholesterol. At fixed concentrations of CS, MbCD significantly increased 5-LO product synthesis in the absence of added AA (Fig. 2, upper panel), thus increasing the threshold concentration of CS to inhibit LT synthesis. Similarly, the sensitivity to CS decreased in the presence of AA (Fig. 2, lower panel). From Fig. 2 it can be seen that agents like cholesterol, which increase membrane rigid- ity, enhanced the inhibitory effects of CS on 5-LO D. A. Aleksandrov et al. Cholesterol sulfate inhibits 5-lipoxygenase FEBS Journal 273 (2006) 548–557 ª 2006 The Authors Journal compilation ª 2006 FEBS 549 product formation, and that agents that decrease mem- brane rigidity (MbCD, AA) attenuated the effect of CS on 5-LO product synthesis. In a wide concentration range, CS significantly inhibited 5-LO product synthe- sis in PMNL and MM6 cells also in the presence of AA in the medium (Fig. 3). CS down-regulates AA-release in PMNL Experiments performed with PMNL labeled with radioactive 14 C-AA revealed the mode of inhibition of LT production by CS. Stimulation of PMNL with ionophore induced the release of 14 C-activity into the supernatant and the formation of radioactively labe- led 5-LO products. After a 30-min incubation of PMNL with CS, the release of 14 C-AA and 14 C- labelled 5-LO products was suppressed in a concen- tration-dependent manner (Fig. 4A). Interestingly, the release of endogenous 14 C-AA was inhibited in a similar way in the presence or absence of 20 lm AA in the medium. Figure 4B presents the total product synthesis under the same assay conditions using labe- led and exogenously added nonlabeled AA, and demonstrates the crucial alteration of the dose– response curve in the presence of exogenously added AA. This fact demonstrates that CS is likely to inhi- bit LT production at the level of substrate availabil- ity. The data demonstrate that AA added together with CS masks the effect of CS up to 50 lgÆmL )1 CS. When no AA is added in preincubations, CS effectively suppressed (50% inhibition) AA release and 5-LO product formation at concentrations of 50 lgÆmL )1 CS. Experiments performed in the pres- ence of exogenous AA showed that 5-LO was inhib- ited by 50% at about 100 lgÆmL )1 CS in Mono Mac6 cells (Fig. 3) and in PMNL (Fig. 4B). A B C Fig. 1. Comparative effects of cholesterol (chol) and its esters on 5-LO product synthe- sis in PMNL. PMNLs were preincubated for 30 min at 37 °C with indicated lipids, and then stimulated for 10 min with 2 l M A23187 calcium ionophore. (A) and (B) Lipid concentration is 10 lgÆmL )1 (A) and 25 lgÆmL )1 (B). (C) MbCD was added together with indicated lipids taken at 25 lgÆmL )1 . The LT synthesis and the total 5-LO product release (as measured by radio- activity) are presented as percentage of the control. *P<0.05; **P<0.01, when corres- ponding data compared to control. Cholesterol sulfate inhibits 5-lipoxygenase D. A. Aleksandrov et al. 550 FEBS Journal 273 (2006) 548–557 ª 2006 The Authors Journal compilation ª 2006 FEBS Cytochalasin D and chlorpromazine inhibit the effect of CS on LT synthesis in PMNL We hypothesized that the effect of CS depended on its internalization and trafficking to ER. This vesicular transport is sensitive to a class of amphipathic amines such as U18666A [35] and chlorpromazine [36,37]. Chlorpromazine (CPZ), as well as an inhibitor of endocytosis, cytochalasin D (Cyto D), completely pre- vented CS-induced inhibition of LT synthesis in PMNL (Fig. 5). Inhibitory effect of CS in a 5-LO cell-free assay We investigated the influence of CS on 5-LO activity under cell-free conditions. 5-LO isolated from PMNL was dissolved in PGC buffer containing AA and differ- ent concentrations of CS. CS in a concentration range of 10–100 lgÆmL )1 inhibited the production of LTs and 5-HETE (Fig. 6). Fig. 2. Cumulative effect of CS and choles- terol (chol) in joined incubations, without (above) or with 20 l M AA (below). PMNLs were preincubated for 30 min at 37 °C with the indicated lipids, with or without AA, and then stimulated for 10 min with 2 l M cal- cium ionophore A23187. MbCD was added at 0.5 m M.*P<0.05; **P<0.01, when data are compared with the control when no lipid added. (*)P < 0.05, when data are compared with corresponding control at the fixed concentration of CS. Fig. 3. CS inhibits 5-LO product synthesis in MM6 and PMNL in a concentration-dependent manner. MM6 and PMNL were preincu- bated for 30 min at 37 °C with CS at the indicated concentrations with 20 l M AA, and then stimulated for 10 min with 1 lM A23187 calcium ionophore. The data presented as the sum of LTs and 5-HETE. D. A. Aleksandrov et al. Cholesterol sulfate inhibits 5-lipoxygenase FEBS Journal 273 (2006) 548–557 ª 2006 The Authors Journal compilation ª 2006 FEBS 551 CS suppresses 5-LO binding to the nuclear membrane in PMNL The influence of CS on intracellular 5-LO localization in human PMNL was investigated after cell stimula- tion with calcium ionophore A23187. Cells were preincubated in PGC buffer with 100 lgÆmL )1 of CS for 15 min at 37 °C. Ionophore was then added and the cells were incubated for an additional 10 min. The cells were then centrifuged and lyzed. Membrane and cytosolic fractions were separated and analyzed by SDS ⁄ PAGE and western blotting technique. The results are presented in Fig. 7. In intact cells, 5-LO was mostly located in the cytosolic fraction. Calcium ionophore induced membrane association of 5-LO. CS prevented the translocation and binding of 5-LO to the membranes to a significant extent (Fig. 7). Discussion We reveal a novel role of CS as a regulatory molecule in the production of LTs. We propose that CS affects LT production by several mechanisms. The liberation of free arachidonic acid is often the initial, rate-limit- ing step in the biosynthesis of eicosanoids. Cytosolic PLA2 (cPLA2), like 5-LO, is a C2 domain-containing A B Fig. 4. CS suppresses AA release (A) and the 5-LO product forma- tion (B) in PMNL in a concentration dependent manner. (A) Thirty- minute incubation of PMNL with CS decreases A23187-induced release of 14 C-labelled 5-LO products and free AA. This effect occurs without (empty columns) and with exogenously added 20 l M AA (hatched columns). (B) The dependence of 5-LO product synthesis on the concentration of CS. Data are given in the pres- ence (hatched columns) and in the absence (empty columns) of 20 l M AA in the medium. Fig. 5. The CS effect on LT synthesis is inhibited by cytochalasin D and chlorpromazine. The inhibitory influence of 25 lgÆmL )1 CS on LT formation in PMNL was abolished in the presence of 10 l M cytochalasin D (Cyto D) and 30 lM chlorpromazine (CPZ). Fig. 6. CS suppresses 5-LO product synthesis in a cell-free assay. 5-LO was isolated as described in the Experimental procedures and diluted into NaCl ⁄ P i ⁄ EDTA buffer. CS was added for 15 min, and then LT synthesis was induced by addition of Ca 2+ at a final con- centration 2 m M. Fig. 7. CS inhibits 5-LO translocation to membranes in human PMNL. A, Ionophore A23187; CS, cholesterol sulfate, 100 lgÆmL )1 ; M, membrane (nuclear) fraction; C, cytosolic fraction. PMN were incubated and stimulated as described in Experimental procedures. The result is representative for three experiments. Cholesterol sulfate inhibits 5-lipoxygenase D. A. Aleksandrov et al. 552 FEBS Journal 273 (2006) 548–557 ª 2006 The Authors Journal compilation ª 2006 FEBS enzyme that preferentially binds to the nuclear mem- brane [38,39]. The C2 domain of cPLA2 [40] and 5-LO [15] strongly favor the zwitterionic phosphatidylcholine and these protein–lipid interactions may thus be inhib- ited by cholesterol and its anionic derivatives, decreas- ing the availability of substrate for 5-LO product formation and thus reducing cellular 5-LO activity. The results obtained in the present work indicate that CS affects LT production at the level of substrate availability (Fig. 4). We observed inhibition of the A23187-induced 5-LO binding to the membrane frac- tion in CS-treated cells (Fig. 7), which can reduce 5-LO function due to lack of colocalization with the substrate AA that is released from the membrane. These two effects of CS, on AA release and 5-LO binding to the membrane, may be linked in light of new published data [41], suggesting that AA can pro- mote membrane association. CS also inhibited LT syn- thesis in a cell-free assay (Fig. 6). This can arise from a direct effect of CS on 5-LO. The CS structure is sim- ilar to that of tirucallic acid, for which direct binding to the 5-LO protein was shown [42]. Cholesterol is especially abundant in the plasma mem- brane of mammalian cells. The importance of maintain- ing the cholesterol–phospholipid (C ⁄ PL) ratio is illustrated by studies which show a marked alteration in the activity of transmembrane proteins and cell function following alterations in the C ⁄ PL ratio [43]. The effects of cholesterol on membrane and cell function are pre- sumably related to its ability to modulate the biophysi- cal properties of membranes. Insertion of cholesterol derivatives would therefore strikingly influence the membrane properties and exert complex effects on several parameters that influence LT biosynthesis: AA release, 5-LO translocation and 5-LO enzyme activity. Recent publications investigated the regulation of 5-LO activity by lipids and it was found that structur- ally diverse lipids showed a similar behavior: both the acidic lipid cardiolipin and the zwitterionic 1-palmi- toyl-2-oleoyl-sn-glycero-3-phosphocholine (POPC) faci- litated 5-LO activation in contrast to PIP and ceramide [19]. We tested the effect of modification of the polar hydroxy group of cholesterol on cellular 5-LO activity. Cholesterol acetate was the most favora- ble substitute. Increasing or decreasing free cholesterol was found to affect 5-LO activity, with clear inhibition at high cholesterol concentrations. The anionic deriva- tives CS and CP inhibited LT synthesis. ER membranes have a low C ⁄ PL ratio in compar- ison with plasma membranes and are consequently the most ‘fluid’ membranes in the cell. Cholesterol addi- tion is known to decrease membrane fluidity. Inclusion of cholesterol reduces the permeability of membranes, increases the molecular order and produces a conden- sing effect on membranes [44]. CS possesses an anionic sulfate moiety at the 3b-position. The electric charge at the membrane surface influences the orientation of phosphatidylcholine polar head groups in membranes: when the charge at the membrane surface is negative, the positively charged end of the choline group moves towards the membrane interior because of electrostatic attraction [45]. Incorporation of CS stabilized mem- brane vesicles and decreased the elasticity of the lipid bilayer [46]. Our results on the inhibition of cellular 5-LO by cholesterol and its derivatives fit well with the recently published concept that membrane fluidity is a key modulator of the activity of 5-lipoxygenase [20]. Chlorpromazine known to decrease ER cholesterol content abolished the effect of CS in our assay. Cholesterol sulfate has been known to be a normal constituent of blood platelets and to modulate platelet function [47]. The sulfotransferase SULT2B1b is selec- tively expressed in human platelets and 566 pmol of CS can be found per 10 9 untreated platelets [48]. Addi- tion of the sulfate donor 3¢-phosphoadenosine-5-phos- phosulfate (PAPS) induced a 300-fold increase in platelet CS content. HDL and specifically apoA-I sta- bilized and maintained the level of platelet SULT2B1b mRNA [48]. We speculate that platelets might present their CS to PMNL, and inhibition of LT synthesis may be a consequence of such a lipid exchange. Physiological concentrations of platelets inhibited LT biosynthesis in human neutrophils in an adhesion- dependent manner [49,50]. After degranulation by thrombin under nonaggregating conditions, platelets lost their inhibitory activity [50]. CS might be a medi- ator of PMNL–platelet interaction and platelet-medi- ated inhibition of leukotriene synthesis in PMNL. It is a published report that a cytoprotective effect of CS exists in gastric ulcer models [26]. Considering the role of PMNL in gastric inflammation and gastric epithelial damage, one might propose that inhibition of LT synthesis by CS contributes to the suppressive effect of CS in gastric inflammation. Regarding the role of CS in inflammation and athero- genesis, other activities of CS should also be considered. Key events in atherogenesis are increased ROS genera- tion and lipid peroxidation. Thus, the role of CS in ROS production should be elucidated. Altogether, our findings show that CS is a potent inhibitor of LT production and that different mecha- nisms are involved in the inhibitory process. These observations provide us with new insights and approa- ches to modify inflammatory processes by means of cholesterol amount and its conversion to anionic deriv- atives. D. A. Aleksandrov et al. Cholesterol sulfate inhibits 5-lipoxygenase FEBS Journal 273 (2006) 548–557 ª 2006 The Authors Journal compilation ª 2006 FEBS 553 Experimental procedures Materials RPMI-1640 medium was from Gibco (Grand Island, NY, USA), and fetal bovine serum (FBS) was obtained from Boehringer Mannheim (Mannheim, Germany). CP was kindly provided by E. Volkov (Moscow State University, Belozersky Institute of Physicochemical Biology). Insulin was a gift from Aventis (Frankfurt, Germany). Human transforming growth factor-b1 (TGF-b1) was purified from outdated platelets as described [51]. Calcitriol was kindly provided by H. Wiesinger (Schering AG, Berlin, Germany). 5-lipoxygenase polyclonal antiserum was kindly provided by O. Radmark (Karolinska Institute, Stockholm, Sweden). CS, CA, calcium ionophore A23187, sucrose, and AA were from Sigma Chemical Co. (Deisenhofen, Germany). High- pressure liquid chromatography (HPLC) solvents were from Merck (Darmstadt, Germany). Cells MM6 cells were maintained in RPMI 1640 medium with glutamine supplemented with 10% fetal calf serum, 100 lgÆmL )1 streptomycin, 100 UÆmL )1 penicillin, 1 mm sodium pyruvate, nonessential amino acids, 1 mm oxalace- tic acid, and 10 lgÆmL )1 bovine insulin. All cultures were seeded at a density of 2 · 10 5 cellsÆmL )1 . MM6 cells were treated with 2 ngÆmL )1 TGF-b1 and 50 nm vitamin D3 (VD3) for 4 days. Cells were harvested by centrifugation (200 g for 10 min at room temperature) and washed once in NaCl ⁄ P i , pH 7.4. Human PMNL were isolated from freshly drawn citrate- anticoagulated donor blood. Leukocyte-rich plasma was prepared by Dextran sedimentation. Granulocytes were obtained from leukocyte-rich plasma by centrifugation on Ficoll–Paque and hypotonic lysis of erythrocytes. PMNL were resuspended (5 · 10 6 cellsÆmL )1 ; purity > 96–97%) in phosphate-buffered saline (NaCl ⁄ P i ) plus 1 mgÆmL )1 glu- cose (PG buffer) or alternatively, in PG and 1 mm CaCl 2 (PGC buffer) as indicated. For incubations, cells were finally resuspended in PGC buffer. The viability of cells was tested by trypan blue exclusion, and more than 95% of cells were viable up to a concentra- tion of 100 lg ÆmL )1 CS and 40 lgÆmL )1 CP; more than 90% were viable at a concentration of 200 lgÆmL )1 CS. 5-LO isolation and cell-free assay PMNL (20–80 · 10 6 ) were suspended in 1 mL of buffer 1 [50 mm KH 2 PO 4 , 0.1 m NaCl, 2 mm EDTA, 1 mm dithio- threitol, 0.5 mm phenylmethylsulfonyl fluoride, and 60 lgÆmL )1 soybean trypsin inhibitor (STI), pH 7.1] and sonicated on ice for three times 10 s. Cell sonicates were centrifuged at 10 000 g for 10 min at 4 °C to remove nuclei and unbroken cells. The supernatant was removed and cen- trifuged again at 100 000 g for 60 min at 4 °C. The pellet was rinsed with 1 mL of buffer 2 (20 mm KH 2 PO 4 ,2mm EDTA, 1 mm dithiothreitol, pH 7.1) to remove residual cytosol. The washed pellet (membrane fraction) was then resuspended in 1 mL of buffer 1 by sonication on ice twice for 10 s. containing 0.2 m Tris ⁄ HCl (pH 7.5), 1.6 mm EDTA, 1.8 mm ATP, aliquot of cytosol and respective con- centrations of CS. These mixtures were preincubated at 37 °C for 15 min, and the reaction was initiated by the addition of CaCl 2 (3 mm in the media) and AA (20 lm in the media). After 15 min, the reaction was stopped by addi- tion of 1 mL methanol. 30 lL1m HCl and 200 ng of pros- taglandin B 1 (PGB 1 ) were added. This mixture was prepared and analyzed by HPLC as further described. Determination of 5-LO product formation in cells PMNL (10 · 10 6 ) and MM6 cells (3 · 10 6 ) were finally resuspended in 1 mL (MM6) or 6 mL (PMNL) of PGC buffer and preincubated at 37 °C with indicated additives as described. The lipids were added as ethanol solutions (cholesterol, CA and CP) or ethanol ⁄ dimethyl sulfoxide (1 : 3) for CS. After 30 min at 37 °C the reaction was star- ted by the addition of 2 lm ionophore A23187. After 10 min, the reaction was stopped with an equal volume of methanol with 200 ng PGB 2 as internal standard for PMNL or 200 ng PGB 1 as internal standard for MM6. To water ⁄ methanol extracts were added 30 l L1m HCl and 500 lL NaCl ⁄ P i per 1 mL incubation. After centrifugation (10 min, 800 g), the samples were applied to C-18 solid- phase extraction columns, which were conditioned first with methanol, then with water. The columns were washed with water, 25% methanol; 5-LO metabolites were extracted with methanol and were analyzed by HPLC as described [52] using UV light detection at 235 and 280 nm. The respective molar absorptions were used for calculation. Cysteinyl LTs (LTC4, D4, and E4) were not determined. AA-release assay Labeled PMNL were prepared by incubating cells for 1.5 h at 1 · 10 7 ⁄ mL in 45 mL PGC buffer with 5.0 lCi of 14 C- AA. At the end of the incubation, 5 mL of 1% bovine serum albumin in Dulbecco’s NaCl ⁄ P i was added for next 0.5 h, then the cells were centrifuged and resuspended at 10 7 ⁄ mL in Hank’s balanced salt solution (HBSS) ⁄ 10 mm Hepes. Medium with or without cholesterol sulfate was equilibrated at 37 °C, the cells were added at 2 · 10 6 ⁄ mL and incubated for 30 min at 37 °C. Then cells were stimula- ted with 2 lm A23187 for 10 min. The incubations were stopped by addition of an equal volume of methanol at )20 °C, with prostaglandin B2 as an internal standard. The denatured cell suspensions were centrifuged and the water ⁄ methanol extracts were removed in the supernatants. Cholesterol sulfate inhibits 5-lipoxygenase D. A. Aleksandrov et al. 554 FEBS Journal 273 (2006) 548–557 ª 2006 The Authors Journal compilation ª 2006 FEBS The water ⁄ methanol extracts were purified by solid-phase extraction using C18 Sep–Pak cartridges (500 mg). The sample was extracted with 1.5 mL methanol, evaporated, reconstituted in 300 lL methanol ⁄ water (2 : 1) and ana- lyzed by HPLC with radiochemical detection. The radio- activity in peaks of LTs, 5-HETE and AA is measured as total 14 C-AA release. Subcellular fractionation by detergent lysis Isolated human PMNL (3 · 10 7 ) were resuspended in 1 mL PGC buffer. The cells were preincubated in the presence or absence of additives at 37 ° C before the addi- tion of ionophore A23187 at the indicated concentrations. After another 5-min incubation period at 37 °C, the samples were dulled on ice for 3–5 min and centrifuged (200 g, 5 min, 4 °C). Pellets were then suspended in 500 lL, ice-cold NP-40 lysis buffer (10 mm Tris ⁄ HCl, pH 7.4; 10 mm NaCl; 3 mm MgCl 2 ;1mm EDTA; 0.1% NP-40; 1 mm phenylmethylsulfonyl fluoride; 60 lgÆmL )1 soybean trypsin inhibitor (STI), and 10 lgÆmL )1 leupep- tin), vortexed (3 · 6 s), kept on ice for 10 min, and centrifuged (800 g, 10 min, 4 °C). Resultant supernatants (non-nuclear fractions) were transferred to a new tube, and the pellets (nuclear fractions) were resuspended in 500 lL ice-cold NP-40 lysis buffer. Nuclei were disrupted by sonication (3 · 6 s). Aliquots of nuclear and non- nuclear fractions were immediately mixed with the load- ing buffer, heated for 6 min at 95 °C, and analyzed for 5-LO protein by SDS ⁄ PAGE and western blotting. Immunoblot analysis of subcellular fractions Twenty-five microliters of each fraction were mixed with 4 lL glycerol ⁄ 0.1% bromophenol blue (1 : 1, v ⁄ v) and ana- lyzed by SDS ⁄ PAGE on 10% gel. After electroblotting, nitrocellulose membranes were blocked with 5% nonfat dry milk in 50 mm Tris-HCI, pH 7.4 and 100 mm NaCl (TBS) for 1 h at room temperature. Membranes were washed and incubated with anti-5-LO antiserum overnight at 4 °C. Then, membranes were washed with TBS and incubated with a 1 : 1000 dilution of alkaline phosphatase-conjugated anti-rabbit IgG for 2 h at room temperature. After washing with TBS, and TBS with 0.1% NP40, 5-LO protein was visualized with the alkaline phosphatase substrates nitro- blue tetrazolium and 5-bromo-4-chloro-3-indolylphosphate in detection buffer (100 mm Tris-HCI, pH 9.5, 100 mm NaCl, 5 mm MgCl 2 ). Statistics Results are given as mean ± sd from at least three inde- pendent experiments. Statistical evaluation of the data was performed by one-way anova followed by Dunnett’s multiple comparisons test. P-values of < 0.05 (*) or < 0.01 (**) were considered significant. Acknowledgements This work was supported by the NATO grant LST CLG 980442 and the grant 04-04-48495 from the Rus- sian Foundation of Basic Researches. The part of this study was supported by the grant from FEBS Fellow- ship Program given to D. Aleksandrov for his work in the Institute of Pharmaceutical Chemistry, Frankfurt- am-Main, Germany. References 1 Samuelsson B (1983) Leukotrienes: mediators of imme- diate hypersensitivity reactions and inflammation. Science 220, 568–575. 2 Walker JL, Loscalzo J & Zhang YY (2002) 5-Lipoxy- genase and human pulmonary artery endothelial cell proliferation. Am J Physiol Heart Circ Physiol 282, H585–H93. 3 Ghosh J & Myers CE (2002) Molecular mechanisms of prostate cancer cell death triggered by inhibition of ara- chidonate 5-lipoxygenase: involvement of Fas death receptor-mediated signals. Adv Exp Med Biol 507, 415– 420. 4 Peters-Golden M & Brock TG (2003) 5-Lipoxygenase and FLAP. Prostaglandins Leukotrienes Essent Fatty Acids 69, 99–109. 5 Radmark O (2003) 5-Lipoxygenase-derived leukotrienes: mediators also of atherosclerotic inflammation. Arterios- cler Thromb Vasc Biol 23, 1140–1142. 6 Rouzer CA & Samuelsson B (1987) Reversible, calcium- dependent membrane association of human leukocyte 5-lipoxygenase. Proc Natl Acad Sci USA 84 , 7393–7397. 7 Wong A, Hwang SM, Cook MN, Hogaboom GK & Crooke ST (1988) Interactions of 5-lipoxygenase with membranes: studies on the association of soluble enzyme with membranes and alterations in enzyme activity. Biochemistry 27, 6763–6769. 8 Rouzer CA & Kargman S (1988) Translocation of 5-lipoxygenase to the membrane in human leukocytes challenged with ionophore A23187. J Biol Chem 263, 10980–10988. 9 Noguchi M, Miyano M, Matsumoto T & Noma M (1994) Human 5-lipoxygenase associates with phosphati- dylcholine liposomes and modulates LTA4 synthetase activity. Biochim Biophys Acta 1215, 300–306. 10 Peters-Golden M & McNish RW (1993) Redistribution of 5-lipoxygenase and cytosolic phospholipase A2 to the nuclear fraction upon macrophage activation. Biochem Biophys Res Commun 196, 147–153. D. A. Aleksandrov et al. Cholesterol sulfate inhibits 5-lipoxygenase FEBS Journal 273 (2006) 548–557 ª 2006 The Authors Journal compilation ª 2006 FEBS 555 11 Brock TG, McNish RW & Peters-Golden M (1995) Translocation and leukotriene synthetic capacity of nuclear 5-lipoxygenase in rat basophilic leukemia cells and alveolar macrophages. J Biol Chem 270, 21652– 21658. 12 Hammarberg T, Provost P, Persson B & Radmark O (2000) The N-terminal domain of 5-lipoxygenase binds calcium and mediates calcium stimulation of enzyme activity. J Biol Chem 275, 38787–38793. 13 May C, Hohne M, Gnau P, Schwennesen K & Kindl H (2000) The N-terminal beta-barrel structure of lipid body lipoxygenase mediates its binding to liposomes and lipid bodies. Eur J Biochem 267, 1100–1109. 14 Chen XS & Funk CD (2001) The N-terminal ‘beta- barrel’ domain of 5-lipoxygenase is essential for nuclear membrane translocation. J Biol Chem 276, 811–818. 15 Kulkarni S, Das S, Funk CD, Murray D & Cho W (2002) Molecular basis of the specific subcellular locali- zation of the C2-like domain of 5-lipoxygenase. J Biol Chem 277, 13167–13174. 16 Burkert E, Arnold C, Hammarberg T, Radmark O, Steinhilber D & Werz O (2003) The C2-like beta-barrel domain mediates the Ca 2+ -dependent resistance of 5-lipoxygenase activity against inhibition by glutathione peroxidase-1. J Biol Chem 278, 42846–42853. 17 Yau WM, Wimley WC, Gawrisch K & White SH (1998) The preference of tryptophan for membrane interfaces. Biochemistry 37, 14713–14718. 18 Sleight RG & Pagano RE (1984) Transport of a fluorescent phosphatidylcholine analog from the plasma membrane to the Golgi apparatus. J Cell Biol 99, 742–751. 19 Pande AH, Moe D, Nemec KN, Qin S, Tan S & Tatulian SA (2004) Modulation of human 5-lipoxygenase activity by membrane lipids. Biochemistry 43, 14653–14666. 20 Pande AH, Qin S & Tatulian SA (2005) Membrane fluidity is a key modulator of membrane binding, inser- tion, and activity of 5-lipoxygenase. Biophys J 88, 4084– 4094. 21 Sud’ina GF, Brock TG, Pushkareva MA, Galkina SI, Turutin DV, Peters-Golden M & Ullrich V (2001) Sul- phatides trigger polymorphonuclear granulocyte spread- ing on collagen-coated surfaces and inhibit subsequent activation of 5-lipoxygenase. Biochem J 359, 621–629. 22 Mehrabian M & Allayee H (2003) 5-Lipoxygenase and atherosclerosis. Curr Opin Lipidol 14, 447–457. 23 Drayer NM & Lieberman S (1965) Isolation of choles- terol sulfate from human blood and gallstones. Biochem Biophys Res Commun 18, 126–130. 24 Ponec M & Williams ML (1986) Cholesterol sulfate uptake and outflux in cultured human keratinocytes. Arch Dermatol Res 279, 32–36. 25 Strott CA & Higashi Y (2003) Cholesterol sulfate in human physiology: what’s it all about? J Lipid Res 44, 1268–1278. 26 Iwamori M, Suzuki H, Ito N, Iwamori Y & Hanaoka K (2005) Lipid compositions of human gastric fluid and epithelium: the role of sulfated lipids in gastric cytopro- tection. J Clin Gastroenterol 39, 129–133. 27 Bleau G, Bodley FH, Longpre J, Chapdelaine A & Roberts KD (1974) Cholesterol sulfate. I. Occurrence and possible biological function as an amphipathic lipid in the membrane of the human erythrocyte. Biochim Biophys Acta 352, 1–9. 28 Denning MF, Kazanietz MG, Blumberg PM & Yuspa SH (1995) Cholesterol sulfate activates multiple protein kinase C isoenzymes and induces granular cell differen- tiation in cultured murine keratinocytes. Cell Growth Differ 6, 1619–1626. 29 Woscholski R, Kodaki T, Palmer RH, Waterfield MD & Parker PJ (1995) Modulation of the substrate specifi- city of the mammalian phosphatidylinositol 3-kinase by cholesterol sulfate and sulfatide. Biochemistry 34, 11489–11493. 30 Sato J, Denda M, Nakanishi J, Nomura J & Koyama J (1998) Cholesterol sulfate inhibits proteases that are involved in desquamation of stratum corneum. J Invest Dermatol 111, 189–193. 31 Ito N, Iwamori Y, Hanaoka K & Iwamori M (1998) Inhibition of pancreatic elastase by sulfated lipids in the intestinal mucosa. J Biochem (Tokyo) 123, 107–114. 32 Iwamori M, Suzuki H, Kimura T & Iwamori Y (2000) Shedding of sulfated lipids into gastric fluid and inhibi- tion of pancreatic DNase I by cholesterol sulfate in con- cert with bile acids. Biochim Biophys Acta 1487, 268– 274. 33 Kawabe S, Ikuta T, Ohba M, Chida K, Ueda E, Yamanishi K & Kuroki T (1998) Cholesterol sulfate activates transcription of transglutaminase 1 gene in normal human keratinocytes. J Invest Dermatol 111, 1098–1102. 34 Kallen J, Schlaeppi JM, Bitsch F, Delhon I & Fournier B (2004) Crystal structure of the human RORalpha ligand binding domain in complex with cholesterol sul- fate at 2.2. J Biol Chem 279, 14033–14038. 35 Sidera C, Parsons R & Austen B (2005) The regulation of beta-secretase by cholesterol and statins in Alzhei- mer’s disease. J Neurol Sci 229–230, 269–273. 36 Sofer A & Futerman AH (1995) Cationic amphiphilic drugs inhibit the internalization of cholera toxin to the Golgi apparatus and the subsequent elevation of cyclic AMP. J Biol Chem 270, 12117–12122. 37 Wang LH, Rothberg KG & Anderson RG (1993) Mis- assembly of clathrin lattices on endosomes reveals a reg- ulatory switch for coated pit formation. J Cell Biol 123, 1107–1117. 38 Schievella AR, Regier MK, Smith WL & Lin LL (1995) Calcium-mediated translocation of cytosolic phospholi- pase A2 to the nuclear envelope and endoplasmic reticu- lum. J Biol Chem 270, 30749–30754. Cholesterol sulfate inhibits 5-lipoxygenase D. A. Aleksandrov et al. 556 FEBS Journal 273 (2006) 548–557 ª 2006 The Authors Journal compilation ª 2006 FEBS 39 Glover S, de Carvalho MS, Bayburt T, Jonas M, Chi E, Leslie CC & Gelb MH (1995) Translocation of the 85-kDa phospholipase A2 from cytosol to the nuclear envelope in rat basophilic leukemia cells stimulated with calcium ionophore or IgE ⁄ antigen. J Biol Chem 270, 15359–15367. 40 Nalefski EA, McDonagh T, Somers W, Seehra J, Falke JJ & Clark JD (1998) Independent folding and ligand specificity of the C2 calcium-dependent lipid binding domain of cytosolic phospholipase A2. J Biol Chem 273, 1365–1372. 41 Flamand N, Lefebvre J, Surette ME, Picard S & Bor- geat P (2005) Arachidonic acid regulates the transloca- tion of 5-lipoxygenase to the nuclear membranes in human neutrophils. J Biol Chem November 7 [Epub ahead of print]. 42 Boden SE, Schweizer S, Bertsche T, Dufer M, Drews G & Safayhi H (2001) Stimulation of leukotriene synthesis in intact polymorphonuclear cells by the 5-lipoxygenase inhibitor 3-oxo-tirucallic acid. Mol Pharmacol 60, 267– 273. 43 Tulenko TN, Constantinescu D, Kikuchi T, Cox RH & Santamore WP (1989) Mutual interaction of vasocon- striction and endothelial damage in stenotic arteries. Am J Physiol 256, H881–H889. 44 McMullen TP, Wong BC, Tham EL, Lewis RN & McElhaney RN (1996) Differential scanning calori- metric study of the interaction of cholesterol with the major lipids of the Acholeplasma laidlawii B membrane. Biochemistry 35, 16789–16798. 45 Scherer PG & Seelig J (1989) Electric charge effects on phospholipid headgroups: phosphatidylcholine in mix- tures with cationic and anionic amphiphiles. Biochemis- try 28, 7720–7728. 46 van den Bergh BA, Wertz PW, Junginger HE & Bouw- stra JA (2001) Elasticity of vesicles assessed by electron spin resonance, electron microscopy and extrusion mea- surements. Int J Pharm 217, 13–24. 47 Blache D, Becchi M & Davignon J (1995) Occurrence and biological effects of cholesteryl sulfate on blood platelets. Biochim Biophys Acta 1259, 291–296. 48 Yanai H, Javitt NB, Higashi Y, Fuda H & Strott CA (2004) Expression of cholesterol sulfotransferase (SULT2B1b) in human platelets. Circulation 109, 92–96. 49 Chabannes B, Poubelle PE, Moliere P, De Medicis R, Lussier A & Lagarde M (2003) Platelets abrogate leuko- triene B(4) generation by human blood neutrophils stimulated with monosodium urate monohydrate or f-Met-Leu-Phe in vitro. Laboratory Invest 83, 491–499. 50 Chabannes B, Moliere P, Merhi-Soussi F, Poubelle PE & Lagarde M (2003) Platelets may inhibit leucotriene biosynthesis by human neutrophils at the integrin level. Br J Haematol 21, 341–348. 51 Werz O, Brungs M & Steinhilber D (1996) Purification of transforming growth factor beta 1 from human plate- lets. Pharmazie 51, 893–896. 52 Werz O & Steinhilber D (1996) Selenium-dependent per- oxidases suppress 5-lipoxygenase activity in B-lympho- cytes and immature myeloid cells: the presence of peroxidase-insensitive 5-lipoxygenase activity in differen- tiated myeloid cells. Eur J Biochem 242, 90–97. D. A. Aleksandrov et al. Cholesterol sulfate inhibits 5-lipoxygenase FEBS Journal 273 (2006) 548–557 ª 2006 The Authors Journal compilation ª 2006 FEBS 557 . Cholesterol and its anionic derivatives inhibit 5-lipoxygenase activation in polymorphonuclear leukocytes and MonoMac6 cells Dmitry A. Aleksandrov 1 , Anna N. Zagryagskaya 1 , Marina A membranes, free cholesterol is converted to its ester in blood plasma. In view of the role of cholesterol lipids in atherogenesis, we sought to determine the in uence of cholesterol and its derivatives. inhibitors, we investigated ani- onic derivatives of cholesterol. Cholesterol and its sul- fated product cholesterol sulfate (CS) are also known as agents that increase membrane rigidity. Being