Pharmacogenetics of the 5-lipoxygenase pathway in asthma

Pharmacogenetics of the 5-lipoxygenase pathway in asthma

E. SILVERMAN, K-H. IN, C. YANDAVA and J. M. DRAZEN

Pulmonary and Critical Care Division, Department of Medicine, Brigham and Women's Hospital and Harvard MedicalSchool, Boston, Massachusetts, USA

Summary

It is now well appreciated that asthma is a chronic inflammatory disease of the airways; among theinflammatory cells that have been implicated in the asthmatic lesion are eosinophils and mast cells.Although these cells have the capacity to produce a number of distinct chemical mediators, thecysteinyl leukotrienes have recently been identified as important mediators of the asthmaticresponse. The leukotrienes are derived from arachidonic acid released from membranephospholipids by the action of phospholipases. The archidonic acid so released in the presence ofthe 5-lipoxygenase (5-LO) activating protein becomes a substrate for the enzyme 5-LO. This enzymecatalyses the stereo-specific addition of molecular oxygen to arachidonic acid to form the productknown as leukotriene A4. Leukotriene A4 subsequently becomes a substrate for one of two enzymes,leukotriene A4 epoxide hydrolase or LTC4 synthase. The former catalyses the formation of LTB4

while the later catalyses the formation of the cysteinyl leukotrienes. Thus the enzyme 5-LO iscritically posed to serve as a regulator of leukotriene synthesis. 5-LO action is known to be regulatedat a number of levels; the mechanisms include regulation of action of the mature protein andregulation of 5-LO gene transcription and translation; there is good reason to believe that all formsof 5-LO regulation are highly interdependent. In this regard we describe the presence and functionalconsequences of a series of naturally occuring mutations in 5-LO core promoter. These mutationsmodify gene transcription in vitro, and may have functional consequences in vivo.

Introduction

The leukotrienes (LO) are a family of polyunsaturatedlipoxygenated eicosatetraenoic acids that are derived fromarachidonic acid and exhibit a wide range of pharmaco-logical and physiological actions [1]. In biological systems,their actions are limited by their relative rates of synthesisand degradation [2]. Of the three enzymes exclusivelyinvolved in the formation of the leukotrienes ± namely,5-lipoxygenase (5-LO), LTC4 synthase, and LTA4 epoxidehydrolase (Fig. 1) ± 5-LO is the enzyme required for theproduction of both the cysteinyl leukotrienes (LTC4,LTD4, and LTE4) and LTB4. Over the past decade, it hasbeen shown that pharmacological inhibition of the actionof 5-LO or antagonism of the action of the cysteinylleukotrienes at their receptor is associated with anamelioration of disease processes thought to derive fromthe excessive action of the leukotrienes (see below). Thus

there is reason to believe that an understanding of thefactors regulating 5-LO enzyme activity will provideinsight into the pathological processes that arise fromleukotriene excess.

Arachidonic acid, the substrate for 5-LO, becomesavailable in the intracellular microenvironment as aproduct of reactions catalysed by one of the variousforms of phospholipase A2 (see review by Dennis et al.[3] and paper by Peters-Golden and McNish [4]). Thebiosynthesis of LTB4 or the cysteinyl leukotrienes thenproceeds as a result of the sequential catalytic action of5-LO on arachidonic acid. The first step results in theformation of 5-HPETE while the second step yieldsLTA4 (5,6 oxido 7,9 trans 11,14 cis eicosatetraenoicacid). This two-step lipoxygenation process involves theformation of a complex containing a nonheme ironbound to Ile673 of 5-LO [5]. The enzymatic product,LTA4, is an unstable intermediate that rapidly degradesunless it serves as a substrate for one of two enzymes;namely, LTC4 synthase or LTA4 epoxide hydrolase. Ineosinophils and mast cells, LTA4 serves as a substratefor LTC4 synthase, which adducts glutathione to the C-6position of LTA4 to form LTC4 (5(S)-hydroxy-6(R)-

Clinical and Experimental Allergy, 1998, Volume 28, Supplement 5, pages 164±170

AhedBhedChedDhedRefmarkerFigmarkerTablemarkerRef endRef start

# 1998 Blackwell Science Ltd 164

Correspondence: Dr J. M. Drazen, Respiratory Disease Division,

Brigham and Women's Hospital, 75 Francis Street, Boston, MA

02115, USA.

Paper 015 Disc

glutathionyl-7,9-trans-11,14-cis-eicosatetraenoic acid).LTC4 and its del-glutamic acid and del-glycine con-geners, known as LTD4 and LTE4, respectively,constitute the cysteinyl leukotrienes. Alternatively, inneutrophils, LTA4 serves as a substrate for LTA4

epoxide hydrolase, which adducts molecular oxygenand oxygen from water to form the potent chemo-attractant LTB4 (5(S), 12(R)-dihydroxy-6,14-cis-8,10-trans-eicosatetraenoic acid).

5-LO (E.C. 1.13.11.34) is a calcium-, ATP-, nonhemeiron-requiring enzyme expressed primarily in cells ofmyeloid origin. The 5-LO gene is 4 85 kb in size andcontains 14 exons; the longest of these is 613 bp [6]. TheG+C-rich 5' flanking sequence has promoter activityand is notable for the absence of TATA or CCAATsequences. Several positive and negative regulatoryregions have been identified including a region & 150bp upstream from the translation start site whichcontains multiple transcription factor binding motifsand is essential for trans-activation. More detailedinformation about the function of this region of thegene is provided below. Under basal conditions, 5-LO isfound in soluble form in the nucleus and cytosol;however, when the host cell is activated, the enzymetranslocates to the perinuclear membrane, where itbecomes catalytically active [4].

Although a variety of biological actions have beenascribed to LTB4, it is predominantly a chemotacticmoiety for eosinophils and neutrophils [7]. In contrast,the cysteinyl leukotrienes were identified as a result ofthe search for the chemical structure of the materialsthat constituted the smooth muscle contractile activityknown as slow-reacting substance of anaphylaxis (SRS-A) [1]. Since their discovery in 1979 and 1980, thecysteinyl leukotrienes have been demonstrated to beamong the most potent bronchoconstrictor substancesever identified [8]. In addition, a myriad other, mostlyproinflammatory biological activities have been attrib-uted to the cysteinyl leukotrienes, from mitogenic effects

to the capacity to modify vascular integrity. Among thehuman disease conditions thought to result from excessavailability of the cysteinyl leukotrienes is asthma [9].

Prior to the structural identification of the leuko-trienes, it had been established that SRS-A was a potentcontractile mediator for isolated airway smooth muscle[10]. Shortly after the elucidation of the chemicalstructure of SRS-A as the cysteinyl leukotrienes it wasshown that LTC4 and LTD4 were potent contractileagonists on isolated human airway smooth muscle andin intact humans [11,12]. The cysteinyl leukotrienes were3000±10 000 times more potent as bronchoconstrictorsubstances than histamine or methacholine; theseobservations heightened interest in the role of thecysteinyl leukotrienes as mediators of asthma. However,the most convincing data for the role of the leukotrienesin asthma are based on the observation that inhibitionof the synthesis or action of the leukotrienes isassociated with an improvement in asthma control[13±16]. Of particular relevance to pharmacogenetics isthe observation from these studies that the 30±90%inhibition of 5-LO activity achieved by zileuton isassociated with clinically significant improvements inasthma outcome. This finding indicates that log-orderchanges in 5-LO action are not required to effect aclinical response and that changes on the order of 2-foldin 5-LO enzyme action are of clinical interest.

The regulation of 5-LO action is poorly understood,but appears to involve multiple control mechanisms.These mechanisms include regulation at the level of genetranscription, translation, enzyme translocation andenzyme inactivation; there is good reason to believethat all forms of 5-LO regulation are important andhighly interdependent.

5-LO action had classically been thought to besubstrate limited because prostaglandin synthesis wasthought to be substrate limited [17], but it is nowappreciated that the availability of free arachidonic acidis only one mechanism regulating 5-LO action [18]. For

# 1998 Blackwell Science Ltd, Clinical and Experimental Allergy, 28, Supplement 5, 164±170

Pharmacokinetics of 5-LO 165

Fig. 1. Schematic diagram showingthe biochemical pathways in theproduction of the cysteinyl

leukotrienes and LTB4.AA=arachidonic acid. Reprinted,with permission, from Annals ofInternal Medicine [35].

Paper 015 Disc

example in the presence of excess 15-HETE (15[S]-hydroxy-[5trans, 8trans, 11trans, 13cis] eicosatetraenoicacid), 15-HETE rather than arachidonic acid is asubstrate for 5-LO. As a result 5,12-diHETE is formedrather than LTA4 [19]. The production of LTA4 fromendogenous, but not exogenous, arachidonic acidrequires the presence of the 5-LO activating protein(FLAP) [20]. Therefore, the capacity to synthesize 5-LOproducts as a result of cellular activation per se isrestricted to cells containing both FLAP and 5-LO aswell as one of the other downstream enzymes (Fig. 1).Since FLAP is generally constitutively expressed, theavailability of FLAP serves as a regulatory mechanismamong various cell types but not within a given cell type[4]. With respect to cells containing both FLAP and 5-LO, there appears to be regulation at the level oftranslocation of 5-LO from soluble, probably bothnuclear and cytosolic fractions, to the perinuclearmembrane [21]. However, one of the most importantaspects of 5-LO regulation is inactivation of thefunctional enzyme by its primary catalytic product,LTA4 [22]. This mechanism has been shown to be activeboth with purified enzyme and in intact neutrophilicpolymorphonuclear leucocytes (PMNs) [22]. The inacti-vation that results from this mechanism, termed suicideinactivation occurs when 5-LO is catalytically active; thisinactivation is irreversible. Thus, for a cell withcatalytically active 5-LO to sustain product output,there must be a means by which to replace theinactivated enzyme. Since 5-LO output can be sustained,de novo production of the enzyme must take place;therefore it is critical to examine the factors regulating 5-LO gene transcription and translation.

With respect to translational regulation, there aredata indicating that 5-LO action is, in part, subject totranslational regulation. Pouliot and coworkers [23]have shown that granulocyte-macrophage colony sti-mulating factor (GM-CSF) up-regulates the expressionof 5-LO protein product without modifying the expres-sion of 5-LO mRNA in PMNs, as determined bynorthern analysis, or the stability of 5-LO mRNA.However, there are much more data indicating that 5-LO is subject to transcriptional regulation.

Several lines of evidence suggest that alterations in thetranscriptional regulation of the 5-LO gene are impor-tant for functional expression of 5-LO and thus mayhave clinical relevance. In response to differentiation andactivation by dimethyl sulfoxide (DMSO), phorbol 12-myristate-13-acetate (PMA), TGF-b, GM-CSF, IL-3,oxidized-LDL or Ca++ ionophore several leucocyte celllines exhibit increased steady-state 5-LO mRNA levels[24±26]. In some studies, at least part of this increase hasbeen attributable to augmented transcription, as demon-

strated by nuclear run-off analysis [26,27]. The mechan-isms by which an inflammatory microenvironmentaugments 5-LO gene transcription have not beenestablished. However, since the 5-LO gene promotercontains numerous consensus binding sites for manyknown transcription factors, including Sp1, Sp3, Egr-1,Egr-2, NF-kB, GATA, Myb, and AP family members, itis not unreasonable to suggest that one or more of thesefactors could be involved. To identify important cis-promoter elements, Hoshiko and associates [28] createda panel of chloramphenicol acetyl transferase (CAT)reporter constructs consisting of 5'-deletions of thegenomic DNA upstream from the translational start sitefor the 5-LO gene and transfected these constructs intoHeLa and HL-60 cells. The resulting data indicated thatthe 5' flanking region of the 5-LO gene contained DNAsequences with both positive and negative regulatoryinfluences on reporter gene transcription (Fig. 2). Withinthe region designated by Hoshiko et al. as the transcrip-tion factor-binding region, a series of five tandem bindingmotifs for the transcription factor Sp1 were identifiedand found to be necessary for pormoter-reporterconstruct activity in HeLa and HL-60 cells. Althoughthis point was not noted by Hoshiko and coworkers, 5-LO is the only sequence in the GENBANK database inwhich there are five tandem Sp1 binding motifs; thisregion of the promoter can also bind the transcriptionfactor known as Egr-1.


166 E. Silverman et al.

Fig. 2. Schematic diagram of the 5' flanking region of thehuman 5-LO gene. Areas were identified by Hoshiko et al. bydeletional analysis with CAT reporter constructs [28]. The

presence of the region from 56 to 179 bp upstream from thetranslational start site was found to be essential for fullexpression of reporter gene activity. This region contains a

number of consensus binding motifs for various transcriptionfactors including five tandem Sp1 consensus binding motifs,i.e. GGGCGG. in tandem. This is the only gene in the

GENBANK database in which five tandem Sp1 binding motifsare found.

Paper 015 Disc

Sp1 is a ubiquitous zinc-finger protein expressed innearly all cell types and is required for the expression ofmany essential genes. Sp1 binds to G+C-rich sitescontaining the consensus sequence (GGGCGG) andsimilar variants. In general, levels of Sp1 expression arehighest in cells undergoing differentiation, and thesehigh levels of Sp1 may be required for the subsequentinduction of tissue-specific genes [29]. Egr-1 and relatedfamily members, namely Egr-2 and Egr-3, are also zinc-finger transcription factors that bind to similar G+C-rich sequences containing the consensus sequence(GCG(T/G)GGGCG). Egr family members are exam-ples of `immediate-early response' proteins and arerapidly and transiently induced by a large number ofgrowth factors, cytokines, and injurious stimuli. Egr-1can displace Sp1 bound to promoter regions of severalgenes and increase transcription above basal levels [30].Indeed, this mechanism of transcriptional activation hasbeen linked with the induction of platelet-derivedgrowth factor (PDGF) in several models of vascularinjury [31]. For example, Egr-1 and Egr-2 levels increaseduring granulocyte differentiation and after cytokine(M-CSF) activation concurrent with 5-LO induction.Moreover, it can be shown that, dexamethasone blocksEgr-1 and Egr-2 expression and `stabilizes' leukocyteactivation [32]. Based on these data it is speculated that5-LO induction during leucocyte differentiation andcytokine activation is mediated by increased Sp1 and/orEgr-1; a Sp11Egr-1 displacement mechanism may beinvolved. If so, it is possible that the naturally occurring5-LO promoter mutations discovered by us, as describedbelow, alter 5-LO functional expression by interferingwith this trans-activation process.

The perceived importance of differences in 5-LO geneexpression as a mechanism regulating 5-LO activity ledus to look for mutations within the 5-LO gene thatwould cause alterations in enzyme production or

structure. It was reasoned that such mutations wouldlikely be present in both non-asthmatic subjects andpatients with asthma but that they would probably befunctionally silent in non-asthmatic individuals becausethe 5-LO pathway would not be activated in this group.Using single-stranded conformational polymorphism(SSCP) analysis, genomic DNA was examined from 25non-asthmatic subjects and 31 patients with asthma formutations within the 5-LO gene that would causealtered amino acid sequence, mRNA splicing, ortranscriptional regulation. Three mutations were identi-fied within the protein encoding region that would notmodify amino acid sequence or likely have any effect onmRNA splicing and therefore were not pursued further.Of greater significance were a family of mutations foundwithin the functionally important G+C-rich core pro-moter region. These mutations consisted of the additionof one or the deletion of one or two Sp1 consensusbinding motifs (Fig. 3). Because of the close similaritybetween Sp1 and Egr-1 binding motifs, the changes alsoresulted in the addition of one or the deletion of one ortwo Egr-1 binding sites. Approximately 35% of thepopulation carries at least one mutant allele at thislocus; thus, if the presence of a mutant allele modifies 5-LO gene expression such modifications will occur withreasonably high frequency in the population.

Next it was ascertained if these mutations had afunctional effect on transcription in chloramphenicolacetyl-transferase reporter assays. Studies comparing300 bp of 5' flanking sequence consisting of either thewild-type or one of the mutant 5-LO promoters revealedsignificant differences among their ability to drive trans-cription in HeLa cells (Fig. 4). In these cells, all theconstructs containing mutant alleles were less effectiveat directing transcription than were constructs contain-ing the wild-type sequence. The magnitude of the de-crease in CAT reporter activity was small (on the order



Fig. 3. Location of the polymorphisms identified in the transcription factor binding region of the 5-LO gene. Numbering appliesonly to the normal sequence and is registered such that the A of the ATG start codon is + 1. Sp1 binding motifs (GGGCGG) are

shown as single underlined alternating with double underlined. Also shown below sequence 1 are the five interlocking Egr-1binding motifs; these are numbered 1 through 5. (No. 1) Normal sequence of the tandem Sp1 binding motifs in the transcriptionfactor binding region of the 5-LO gene. (No. 2) Sequence with the 12-bp deletion. (No. 3) Sequence with the 6-bp deletion. (No. 4)

Sequence with the 6-bp addition. Because the gene contains five Sp1 binding motifs in tandem, assignment of the specific deletionswithin the region of tandem repeats is arbitrary. Sequence numbering is from the GENBANK accession #M38191.Seq Human 5-Lipoxygenase Gene Exon #1±1/91. Reprinted, with permission, from the Journal of Clinical Investigation [34].

Paper 015 Disc

of 20±35%); however; clinical data for zileuton obtainedin asthma treatment trials [14] indicate the potential fora clinically significant benefit if this magnitude of effectwas translated into a 20±35% decrease in the availabilityof end product. These data indicate that in HeLa cells allmutant forms of the promoter are less effective inbinding and directing transcription of the 5-LO than thewild type form of the promoter.

Because these mutant alleles differed in the number ofSp1 and Egr-1 binding motifs, the question was raised ifthis change in CAT reporter activity was reflected in analtered capacity of oligonucleotides corresponding tothe mutant forms of the transcription factor-bindingregion to bind nuclear extracts from cells known toproduce Sp1 and Egr-1 when cultured under variousconditions. For this analysis, mobility shift assays(EMSA) were performed using nuclear extracts fromhuman umbilical-vein endothelial cells (HUVECs) in theabsence and presence of phorbol myristate acetate(PMA) (Fig. 5a). When HUVECs are cultured in theabsence of PMA, Sp1, but not Egr-1, is a knownconstituent of their nuclear extract; in contrast, whenHUVECs are cultured in the presence of PMA, their

nuclear extracts contain both Sp1 and Egr-1 [33].Synthetic oligonucleotides corresponding to the wild-type 5-LO transcription factor-binding region were sub-jected to electrophoresis in the absence of nuclearextract (indicated in the figure as `-'), in the presenceof nuclear extract from HUVECs cultured withoutPMA (indicated in the figure as `no PMA'), and with



Fig. 4. Relative CAT activity, corrected for transfectionefficiency, of HeLa cells transfected with pCAT containing

either the wild-type 5-LO promoter (assigned a value of 100) ormutant promoter isolated by PCR from subjects harbouringthe 12-bp deletion, the 6-bp deletion and the 6-bp addition.The capacity of the mutant forms of DNA to promote

transcription of the CAT reporter constructs was significantlyless than that of the wild type DNA. Results are the mean offive experiments each performed in triplicate. Reprinted, with

permission, from the Journal of Clinical Investigation [34].

Fig. 5. Electrophoretic mobility shift assay (EMSA) and super-shift analysis of nuclear proteins from HUVEC with radiola-belled wild-type (WT) oligonucleotide. (A) No bands are seen in

the absence of nuclear extract, and the radiolabelledWTmigratesto the bottom of the gel. Unstimulated cells (No PMA) produceessentially one intense band that supershifts with the addition of

anti-Sp1 antibodies. PMA-stimulated cells produce an additionallower band that supershifts with anti-Egr-1 antibodies. Controlantibodies to Sp4 and ETS proteins produced no change in theintensity or position of bands. (B) All three mutant oligonucleo-

tides are capable of binding Sp1 and inducible Egr-1. Specificactivity and quantity of nuclear extract were equivalent for eacholigonucleotide. The bands produced by WT are more intense

than those produced by the mutant forms. Reprinted, withpermission, from the Journal of Clinical Investigation [34].

Paper 015 Disc

nuclear extract from HUVECs cultured in the presenceof PMA (indicated in the figure as `PMA'). In addition,supershift analyses were performed with antibodies toSp1, Egr-1, Sp4 and ETS. The EMSA showed a singleband when nuclear extract from HUVECs culturedwithout PMA was used and two bands when extractfrom HUVECs cultured with PMA was used. Further-more, the band with the lower mobility supershiftedwith the antibody to Sp1 and the band with the highermobility supershifted with the antibody to Egr-1. WhenEMSAs were performed with oligonucleotides corre-sponding to the wild-type and various mutant forms ofthe 5-LO promoter, the same qualitative pattern emerg-ed but the bands from the mutant forms of the promoter(Fig. 5b) were 47±54% less intense than those from thewild-type for Sp1 and 42±67% less intense for Egr-1[34]. The decreased binding observed with the mutantpromoters may offer an explanation for the decreasedtranscription levels in our promoter-reporter assays.

To determine the effect of each candidate transscripi-tion factor (Sp1 or Egr-1) in isolation, we performedcotransfection studies in Drosophila SL2 cells (Schneidercells). Schneider cells, unlike most mature mammaliancells, do not naturally express Sp1 or Egr-1 and provide alow background of basal transcription. In this system,overexpression of either Egr-1 or Sp1 by cotransfectionwith the respective expression construct increased CATactivity of all promoter-reporter constructs (Fig. 6; leftpanel for Sp1, right panel for Egr-1). However, not allpromoter mutants were equally effective at drivingtranscription. Mutant Add(+6) was the most activepromoter, with wild-type, Del(±6) andDel(±12) constructsshowing progressively less responsiveness to Egr-1. Sp1overexpression resulted in a similar pattern of reporter

gene transcription among the mutant forms of thepromoter, however, the relative changes were lessimpressive than those observed with Egr-1. Together,these studies suggest that 5-LO transcription may bedifferentially regulated among the wild-type and mutantgenes, and that the ability of these promoters to mediatetranscription reflects the number of intact Sp1/Egr-1-binding sites. These results differ from those reported forHeLa cells in that the addition alleles weremore responsivethan the deletion alleles. The results for Schneider cellsindicate that the functional effect of these mutationsdepends substantially on the cell type used (e.g. HeLa vs.Scheinder) and the activating conditions studied.

It is speculated that increasing the number of Sp1/Egr-1 consensus binding sites in the promoter regionincreases the affinity for and therefore the trans-activa-tion potential of these transcription factors for the 5-LOgene. These promoter changes are likely to have a moreprofound effect on gene transcription and expression indifferentiated or cytokine-activated leukocytes whenlevels of Sp1 and Egr-1 are high. It is unknown whetherleukocytes from individuals with these promoter muta-tions have altered 5-LO activity and whether this altered5-LO activity is reflected in asthma phenotype.

Although it has been known for some time that 5-LO isa critical enzyme in the metabolic pathways leading to theproduction of the leukotrienes, only in the past 3±5 yearshas evidence accrued from clinical studies indicating that5-LO action plays a pivotal role in the biology of humandisease, especially asthma [35]. Using SSCP a family ofnaturally occurring promoter mutations has been identi-fied within the 5-LO gene which consist of a variablenumber of tandem Sp1/Egr-1 consensus binding sites.These mutations alter transcription factor binding andreport construct activity suggesting that they may have invivo consequences. Indeed, in the light of these observa-tions, it is reasonable to conclude that the regulation ofthe action of 5-LO will have important implications forour understanding of the biology of asthma.

References

1 Samuelsson B. Leukotrienes: mediators of immediatehypersensitivity reactions and inflammation. Science1983; 220:568±75.

2 Stene DO, Murphy RC. Metabolism of leukotriene E4 inisolated rat hepatocytes. Identification of beta-oxidationproducts of sulfidopeptide leukotrienes. J Biol Chem 1988;

263:2773±8.3 Dennis EA, Rhee SG, Billah MM, Hannun YA. Role of

phospholipase in generating lipid second messengers insignal transduction. Faseb J 1991; 5:2068±77.

4 Peters-Golden M, McNish RW. Redistribution of 5-lipoxygenase and cytosolic phospholipase A2 to the nuclear



Fig. 6. Transient transfection analysis comparing the WT andmutant 5-LO promoters for Sp1 and Egr-1 inducibility. CATactivity was directly proportional to the number of SP1/Egr-1

consensus binding sites. The CAT reporter construct contain-ing 5-LO promoter Add (+6) was most responsive, whereasconstructs containing promoters WT, Del (76) and Del (712)

showed progressively less inducibility. Relative CAT activitiesand standard deviations are shown. Egr-1 over expressionresulted in greater differences compared with Sp1.

Paper 015 Disc

fraction upon macrophage activation. Biochem Biophys

Res Commun 1993; 196:147±53.5 Hammarberg T, Zhang YY, Lind B, Radmark O,Samuelsson B. Mutations at the c-terminal isoleucine and

other potential iron ligands of 5-lipoxygenase. Eur JBiochem 1995; 230:401±7.

6 Funk CD, Hoshiko S, Matsumoto T, Radmark O,Samuelsson B. Characterization of the human 5-lipoxy-

genase gene. Proc Natl Acad Sci USA 1989; 86:2587±91.7 Rola-Pleszczynski M. LTB4 and PAF in the cytokinenetwork. Adv Exp Med Biol 1991; 314:205±21.

8 Drazen JM. Inhalation challenge with sulfidopeptideleukotrienes in human subjects. Chest 1986; 89:414±9.

9 Holgate ST, Bradding P, Sampson AP. Leukotriene

antagonists and synthesis inhibitors: new directions inasthma therapy. J Allergy Clin Immunol 1996; 98:1±13.

10 Drazen JM, Lewis RA, Wasserman SI, Orange RP, AustenKF. Differential effects of a partially purified preparation of

slow-reacting substance of anaphylaxis on guinea pig trachealspirals and parenchymal strips. J Clin Invest 1979; 63:1±5.

11 DahleÂ n S-E, Hedqvist P, Hammarstrom S, Samuelsson B.

Leukotrienes are potent constrictors of human bronchi.Nature 1980; 288:484±6.

12 Weiss JW, Drazen JM, Coles N, McFadden ER Jr, Weller

PF, Corey EJ et al. Bronchoconstrictor effects of leuko-triene C in humans. Science 1982; 216:196±8.

13 Israel E, Dermarkarian R, Rosenberg M, Sperling R,

Taylor G, Rubin P et al. The effects of a 5-lipoxygenaseinhibitor on asthma induced by cold, dry air. N Engl J Med1990; 323:1740±4.

14 Israel E, Rubin P, Kemp JP, Grossman J, PiersonWE, Siegel

SC et al. The effect of inhibition of 5-lipoxygenase by zileutonin mild to moderate asthma. Ann IntMed 1993; 119:1059±66.

15 Spector SL, Smith LJ, Glass M, Birmingham BK, Bronsky

EA, Dunn KD et al. Effects of 6 weeks of therapy with oraldoses of ICI 204, 219, a leukotriene D4 receptor antagonist,in subjects with bronchial asthma. Am J Respir Crit Care

Med 1994; 150:618±23.16 Reiss TF, Sorkness CA, Stricker W, Botto A, Busse WW,

Kundu S et al. Effects of montelukast (MK-0476), a potent

cysteinyl leukotriene receptor antagonist, on bronchodila-tion in asthmatic subjects treated with and without inhaledcorticosteroids. Thorax 1997; 52:45±8.

17 Bergstrom S. Prostaglandins: members of a hormonal

system. These physiologically very potent compounds ofubiquitous occurrence are formed from essential fattyacids. Science 1967; 157:382±91.

18 Rosenthal MD, Rzigalinski BA, Blackmore PF, FransonRC. Cellular regulation of arachidonate mobilization andmetabolism. Prostaglandins Leukot Essent Fatty Acids

1995; 52:93±8.19 Petrich K, Ludwig P, Kuhn H, Schewe T. The suppression

of 5-lipoxygenation of arachidonic acid in human poly-morphonuclear leucocytes by the 15-lipoxygenase product

(15S)-hydroxy-(5Z,8Z,11Z,13E) -eicosatetraenoic acid:structure-activity relationship and mechanism of action.Biochem J 1996; 314:911±6.

20 Dixon RA, Diehl RE, Opas E, Rands E, Vickers PJ, Evans

JF et al. Requirement of a 5-lipoxygenase-activating

protein for leukotriene synthesis. Nature 1990; 343:282±4.21 Lepley RA, Fitzpatrick FA. Inhibition of mitogen-acti-

vated protein kinase kinase blocks activation and redis-

tribution of 5-lipoxygenase in HL-60 cells. Arch BiochemBiophys 1996; 331:141±4.

22 Lepley RA, Fitzpatrick FA. Irreversible inactivation of 5-lipoxygenase by leukotriene A4. Characterization of

product inactivation with purified enzyme and intactleukocytes. J Biol Chem 1994; 269:2627±31.

23 Pouliot M, Mcdonald PP, Khamzina L, Borgeat P, McColl

SR. Granulocyte-macrophage colony-stimulating factorenhances 5-lipoxygenase levels in human polymorpho-nuclear leukocytes. J Immunol 1994; 152:851±8.

24 Jakobsson PJ, Steinhilber D, Odlander B, Radmark O,Claesson HE, Samuelsson B. On the expression andregulation of 5-lipoxygenase in human lymphocytes. ProcNatl Acad Sci USA 1992; 89:3521±5.

25 SteinhilberD,HoshikoS,GrunewaldJ,RadmarkO,Samuels-son B. Serum factors regulate 5-lipoxygenase activity inmaturatingHL60 cells. BiochimBiophysActa 1993; 1178:1±8.

26 Stankova J, Rolapleszczynski M, Dubois CM. Granulo-cyte-macrophage colony-stimulating factor increases 5-lipoxygenase gene transcription and protein expression in

human neutrophils. Blood 1995; 85:3719±26.27 Ponton A, Thirion JP, Sirois P. Repression of the 5-

lipoxygenase gene by c-myb overexpression in differen-

tiated HL-60 cells. Prostaglandins 1997; 53:49±58.28 Hoshiko S, Radmark O, Samuelsson B. Characterization

of the human 5-lipoxygenase gene promoter. Proc NatlAcad Sci USA 1990; 87:9073±7.

29 Courey AJ, Tjian R. Mechanisms of transcriptional controlas revealed by studies of human transcription factor Sp1.In: McKnight SL, Yamamoto KR. (eds) Transcriptional

Regulation. Cold Spring Harbor, NY. Cold Spring HarborPress 1992: 743±769.

30 Khachigian LM, Linder V, Williams AJ, Collins T. Egr-1-

induced endothelial gene expression: a common theme invascular injury. Science 1996; 271:1427±31.

31 Silverman ES, Khachigian LM, Lindner V, Williams AJ,

Collins T. Inducible PDGF a-chain transcription in smoothmuscle cells is mediated by Egr-1 displacement of Sp1 andSp3. Am J Physiol-Heart Circ Phy 1997; 42:H1415±H1426.

32 Kharbanda S, Nakamura T, Stone R, Hass R, Bernstein S,

Datta R et al. Expression of the early growth response 1and 2 zinc finger genes during induction of monocyticdifferentiation. J Clin Invest 1991; 88:571±7.

33 Khachigian LM, Williams AJ, Collins T. Interplay of Sp1and Egr-1 in the proximal platelet-derived growth factor a-chain promoter in cultured vascular endothelial cells. J Biol

Chem 1995; 270:27679±86.34 In KH, Asano K, Beier D, Grobholz J, Finn PW,

Silverman EK et al. Naturally occurring mutations in thehuman 5-lipoxygenase gene promoter that modify tran-

scription factor binding and reporter gene transcription. JClin Invest 1997; 99:1130±7.

35 O'Byrne PM, Israel E, Drazen JM. Antileukotrienes in the

treatment of asthma. Ann Intern Med 1997; 127:472±80.



Paper 015 Disc

Documents

Pharmacogenetics of the 5-lipoxygenase pathway in asthma