HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

J. Biol. Chem., Vol. 277, Issue 30, 27360-27366, July 26, 2002

This Article Abstract Full Text (PDF) All Versions of this Article: 277/30/27360    most recent M203234200v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Walther, M. Articles by Kuhn, H. Search for Related Content PubMed PubMed Citation Articles by Walther, M. Articles by Kuhn, H. Social Bookmarking                      What's this?

The N-terminal Domain of the Reticulocyte-type 15-Lipoxygenase Is Not Essential for Enzymatic Activity but Contains Determinants for Membrane Binding^*

Matthias Walther, Monika Anton, Martin Wiedmann^§, Robert Fletterick^¶, and Hartmut Kuhn

From the  Institute of Biochemistry, University Clinics Charité, Humboldt University, Monbijoustrasse 2, D-10117 Berlin, Germany, the ^§ Cellular Biochemistry and Biophysics Program, Memorial Sloan-Kettering Cancer Center, New York, New York 10021, and the ^¶ Department of Biochemistry and Biophysics, University of California, San Francisco, California 94143-0448

Received for publication, April 4, 2002, and in revised form, May 7, 2002

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

The rabbit reticulocyte-type 15-lipoxygenase is capable of oxygenating biomembranes and lipoproteins without the preceding action of ester lipid cleaving enzymes. This reaction requires an efficient membrane binding, and the N-terminal -barrel domain of the enzyme has been implicated in this process. To obtain detailed information on the structural requirements for membrane oxygenation, we expressed the rabbit wild-type 15-lipoxygenase, its -barrel deletion mutant (catalytic domain), and several lipoxygenase point mutations as His-tagged fusion proteins in Escherichia coli and tested their membrane binding characteristics. We found that: (i) the -barrel deletion mutant was catalytically active and its enzymatic properties (K_M, V_max, pH optimum, substrate specificity) were similar to those of the wild-type enzyme; (ii) when compared with the wild-type lipoxygenase, the membrane binding properties of the N-terminal truncation mutant were impaired but not abolished, suggesting a role of the catalytic domain in membrane binding; and (iii) Phe-70 and Leu-71 (constituents of the -barrel domain) but also Trp-181, which is located in the catalytic domain, were identified as sequence determinants for membrane binding. Mutation of these amino acids to more polar residues (F70H, L71K, W181E) impaired the membrane binding capacity of the recombinant enzyme. These data indicate that the C-terminal catalytic domain of the rabbit 15-lipoxygenase is enzymatically active and that the membrane binding properties of the enzyme are determined by a concerted action of the N-terminal -barrel and the C-terminal catalytic domain.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Lipoxygenases (LOXs)¹ form a heterogeneous family (1-3) of lipid-peroxidizing enzymes that catalyze the dioxygenation of polyunsaturated fatty acids to their corresponding hydroperoxy derivatives. Plant and mammalian LOXs constitute single polypeptide chains that are folded into a two-domain structure (4-7). The large C-terminal domain may be considered as catalytic subunit because it contains the catalytically active non-heme iron and the substrate-binding cavity. In contrast, the function of the N-terminal -barrel domain has been elusive for several years. From the structural point of view, there is no obvious reason to believe that the N-terminal domain may interfere with the catalytic process except for modifying its efficiency. On the other hand, truncation studies on two mammalian LOX species revealed a loss of enzyme functionality when the N-terminal -barrel domain was deleted completely or in part (8, 9). In contrast, recent proteolysis studies in which the N-terminal -barrel domain of the soybean-LOX-1 was cleaved off by limited trypsinization led to a catalytically active mini-enzyme (10). The kinetic parameters of this truncated enzyme species were similar to those of the native LOX, indicating that the N-terminal domain may not be essential for the oxygenation of free polyenoic fatty acids and for membrane binding.

In cellular systems LOX isoforms are localized in different compartments. Mammalian 15-LOXs are cytosolic enzymes that translocate to various types of intracellular membranes following cell activation (11, 12). Mammalian 5-LOXs show a more complex distribution pattern (13). In resting cells they are located either in the cytosol (14) or in the nucleus (15). Cell stimulation that leads to an increased cytosolic calcium concentration induces translocation of the enzyme to the nuclear envelope (16), where it complexes with other constituents of the leukotriene synthesizing cascade (17). Experiments with detergent-solubilized inclusion bodies containing a fusion protein that consists of the human 5-LOX N-terminal -barrel domain and the glutathione S-transferase suggested that the -barrel domain is capable of binding calcium and, thus, may be important for calcium-dependent membrane association (18). When cells transfected with a green fluorescence protein construct containing the 5-LOX -barrel domain were stimulated with calcium ionophore, a strong fluorescence signal was observed at the nuclear envelope. Thus, the 5-LOX -barrel domain may drive the calcium-dependent membrane association observed in vivo (19). To determine the structural determinants for the translocation process, molecular modeling and site-directed mutagenesis was carried out on the human 5-LOX. These studies revealed that the first 130 amino acids might form a calcium-binding C2 domain (20), which is a structural motif that has been identified before in many proteins involved in membrane trafficking and transmembrane signaling (21-23). In addition the 5-LOX C2-domain contains three solvent-exposed tryptophans (Trp-13, Trp-75, Trp-102) that have been implicated in membrane binding (20).

Although the human 5-LOX effectively binds to biomembranes in the presence of calcium, it does not oxygenate polyenoic fatty acids esterified to the membrane lipids. In contrast, the reticulocyte-type 15-LOXs are capable of oxygenating complex ester lipids (24, 25) even if they are bound in biomembranes (26, 27) or lipoproteins (28, 29). In fact, the oxygenation of membrane lipids was implicated in membrane degradation and/or remodeling during erythropoiesis (30, 31) and LOX-induced oxidation of low density lipoprotein was suggested to be involved in the pathogenesis of atherosclerosis (32). As 5-LOXs the mammalian 15-LOXs are capable of binding to biomembranes in the presence of micromolar calcium concentrations, and this membrane translocation can be reversed by calcium removal (12). However, in contrast to the 5-LOXs that preferentially translocate to the nuclear envelope, the reticulocyte-type 15-LOX associates with different types of biomembranes (endoplasmic membranes, mitochondrial membranes) and this membrane binding does not require a special docking protein (33). The x-ray coordinates of the rabbit 15-LOX suggested structural similarities of the N-terminal LOX domain with a C-terminal -barrel of mammalian lipases (6). Because this lipase domain was implicated in membrane association (34), a similar function has been proposed for the N-terminal -barrel domains of mammalian 15-LOXs, but for the time being there are no experimental data supporting this hypothesis.

To study the significance of the N-terminal -barrel domain of the rabbit 15-LOX for fatty acid oxygenation and membrane binding, we expressed the native enzyme and its -barrel truncation mutant as His-tagged fusion proteins and tested the purified enzyme species in membrane binding and activity assays. The results obtained indicate that the N-terminal -barrel domain is not essential for fatty acid oxygenation but contains at least two sequence determinants for membrane binding (Phe-70, Leu-71). In addition, mutagenesis studies identified a further membrane binding determinant (Trp-181) that, in contrast to Phe-70 and Leu-71, is located at the surface of the catalytic domain.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Chemicals-- The chemicals used were from the following sources: (5Z,8Z,11Z,14Z)-eicosa-5,8,11,14-tetraenoic acid (arachidonic acid), (9Z,12Z)-octadeca-9,12-dienoic acid (linoleic acid), EDTA, imidazole, and sodium borohydride from Serva (Heidelberg, Germany); ampicillin from Invitrogen (Eggenstein, Germany); kanamycin, glycerol, DTT (dithiothreitol), trichloroacetic acid, and sucrose from Sigma (Deisenhofen, Germany); (9Z,12Z,15Z)-octadeca-9,12,15-trienoic acid (-linolenic acid), (6Z,9Z,12Z)-octadeca-6,9,12-trienoic acid (-linolenic acid), and HPLC reference compounds ((12S,5Z,8Z,10E,14Z)-12-hydroxy-5,8,10, 14-eicosatetraenoic acid (12S-HETE), (15S,5Z,8Z,11Z,13E)-15-hydroxy-5,8,11,13-eicosatetraenoic acid (15S-HETE), (13S,9Z,11E)-13-hydroxyoctadeca-9,11-dienoic acid (13S-HODE)) from Cayman Chemical Co. (distributed by Alexis GmbH, Grünberg, Germany); and HPLC solvents from Baker (Deventer, The Netherlands). Restriction enzymes were purchased from New England Biolabs (Schwalbach, Germany). Phage T4 ligase, Pwo polymerase, and sequencing kits were obtained from Roche Molecular Biochemicals (Mannheim, Germany), and the Escherichia coli strain M15[prep4] was purchased from Qiagen (Hilden, Germany). Oligonucleotide synthesis was carried out by TiB-Molbiol (Berlin, Germany).

Bacterial Expression and Purification of the Recombinant Enzyme Species-- The recombinant wild-type 15-LOX and the various mutants were expressed in E. coli as His-tagged fusion proteins. For this purpose the 15-LOX cDNA was cloned into the pQE-9 expression plasmid (Qiagen) between the SalI and HindIII restriction sites. Bacteria were transformed with the recombinant plasmids, and, routinely, 3 liters of LB medium containing 100 mg/liter ampicillin and 25 mg/liter kanamycin were inoculated with a 15-ml overnight preculture. The bacteria were allowed to grow at 37 °C for 16 h, and then expression of the recombinant protein was induced by addition of 1 mM isopropyl--D-thiogalactopyranoside (final concentration). The cultures were kept for additional 2 h at 30 °C, and then the bacteria were pelleted, washed (PBS), and resuspended in 30 ml of PBS. Cells were disrupted by sonication with a Labsonic U tip-sonifier on ice (Braun, Melsungen, Germany), and the debris was spun down. The clear lysis supernatant was applied to a 0.5-ml nickel-agarose column (Qiagen). The column was washed twice with washing buffer (50 mM NaH₂PO₄, 300 mM NaCl, 10 mM imidazole, pH 8.0), and the adherent proteins were removed rinsing the column with elution buffer (50 mM NaH₂PO₄, 300 mM NaCl, 200 mM imidazole, pH 8.0). Five 0.25-ml fractions were collected, and the LOX activity was assayed in each of them. Routinely, more than 90% of the activity was recovered in fractions 2-4. These fractions were pooled, diluted 1:50 with 20 mM Tris-HCl buffer, pH 7.4, and loaded onto a Q-Sepharose column (gel bed volume, 250 µl; Amersham Biosciences, Freiburg, Germany) for further purification by anion exchange chromatography. After loading the column was washed twice with 1 ml of 20 mM Tris-HCl buffer, pH 7.4, and then the enzyme was eluted with 100 mM KCl dissolved in the same buffer. Fractions of 0.125 ml were collected, activity was assayed, and the active fractions were pooled. For storage the enzyme preparation was supplemented with 10% glycerol and stored at 80 °C. Starting from a 3-liter culture, this procedure yielded approximately 1 mg of electrophoretically pure protein.

Site-directed Mutagenesis-- Site directed mutagenesis was performed by the PCR-overlap extension technique using mismatching synthetic oligonucleotides. The PCR products containing the nucleotide exchanges were digested with appropriate restriction enzymes and inserted into the pQE-9-expression plasmid containing the wild-type 15-LOX as His-tagged fusion protein. For each mutant, 10-20 clones were screened by restriction mapping and activity assays to identify LOX-positive clones, and all mutants were confirmed by sequencing. For deletion of the N-terminal -barrel domain, a SalI restriction site was introduced in front of Cys-115 by PCR. The PCR fragment was digested with SalI and KpnI, ligated into the pQE-9 expression plasmid containing the wild-type 15-LOX, and E. coli (M15[prep4]) were transformed with this plasmid. This procedure led to a N-terminal truncation mutant that lacked the first 114 amino acids (-barrel truncation mutant). Expression and purification of this truncation mutant was performed as described for the other mutants. In addition to this -barrel truncation, further truncation mutants were created starting with the following amino acids: Gly-155, Glu-165, Phe-175, and Pro-197. The Pro-197 mutant lacked the entire helix 2 (6) that contains Trp-181 (see Fig. 7). Unfortunately, all these mutants turned out to be enzymatically inactive.

Fatty Acid Oxygenase Activity-- The fatty acid oxygenase activity of the various enzyme species was assayed either spectrophotometrically measuring the increase in absorbance at 235 nm or by HPLC quantification of the oxygenation products. For the spectrophotometric measurements, the assay mixture consisted of a 0.1 M sodium phosphate buffer, pH 7.4, containing 0.1 mM fatty acid substrate. Before the reaction was started by enzyme addition, the mixture was sonicated in an ultrasonic bath for 5 min. The reaction was stopped by the addition of 1 ml of ice-cold methanol, hydroperoxides formed were reduced to the more stable hydroxy compounds with sodium borohydride, and the sample was acidified to pH 3 (acetic acid). After removing the precipitate by centrifugation, aliquots of the clear supernatant were injected to RP-HPLC for product analysis. For calculation of the kinetic parameters (K_M and V_max), the substrate concentration was varied in the range of 5-50 µM. For Lineweaver-Burk analysis, the linear parts of the kinetic progress curves were used.

To test the enzymatic activity of the mutants created, 5 ml of bacterial liquid cultures were grown and expression of the recombinant enzyme species was induced as described above. The bacteria were pelleted, resuspended in 0.5 ml of PBS, and after cell lysis the homogenates were incubated with 0.1 mM arachidonic acid for 10 min at room temperature. The reaction was stopped by addition of 0.5 ml of ice-cold methanol, and hydroperoxy fatty acids formed were reduced with sodium borohydride. After acidification to pH 3, precipitates were spun down and aliquots were injected to RP-HPLC.

Membrane Oxygenase Activity-- To determine the membrane oxygenase capacity of wild-type and mutant 15-LOX species, we used beef heart submitochondrial particles (SMP) or dog pancreas EDTA-high salt-stripped rough microsomes (EKRM) as model membranes. For the measurements of the membrane oxygenase activity, the enzyme species were incubated at room temperature for 10 min in 0.5 ml of PBS containing 0.5 mg of SMP protein. The reaction was stopped by the addition of 500 µl of methanol, and hydroperoxides formed were reduced with sodium borohydride. Then 0.2 ml of 40% KOH were added, and the ester lipids were hydrolyzed at 60 °C for 30 min under argon atmosphere. The mixture was acidified to pH 3, precipitates were removed by centrifugation, and aliquots of the clear supernatant were injected to RP-HPLC for quantification of the lipid oxygenation products.

Membrane Binding Assay-- In our standard membrane binding assay, 200 µg of SMP were incubated at room temperature with 2.5 µg of recombinant 15-LOX (wild-type or mutant enzyme species) in 50 mM HEPES buffer containing 150 mM NaCl, 5 mM MgCl₂, and 1 mM DTT, pH 7.4 (total assay volume 25 µl). After a 5-min incubation period, the samples were underlaid with a 0.1-ml sucrose cushion (500 mM sucrose in the same buffer) and centrifuged for 15 min at 100,000 × g in a Beckmann tabletop centrifuge (4 °C). The pellet of this centrifugation step represented LOX-loaded SMPs. The supernatant was carefully removed and transferred to a separate tube, and ovalbumin was added to reach a final concentration of 60 µg/ml. The supernatant proteins were then precipitated with trichloroacetic acid, reaching a final concentration of 10%, and the precipitate was centrifuged for 20 min at 20,000 × g. The supernatant of the centrifugation step was discarded, and the two pellets (100,000 × g pellet and 20,000 × g pellet) were reconstituted in 25 µl of 2-fold concentrated electrophoresis loading buffer (0.58 M sucrose, 280 mM Tris base, 211 mM Tris-HCl, 139 mM SDS, 1 mM EDTA, 0.44 mM Serva Blue G250, and 200 mM DTT). After heating to 95 °C for 5 min, aliquots (10 µl) were loaded onto SDS-PAGE. After electrophoresis, the proteins were blotted to a nitrocellulose membrane by a semi-dry blotting technique and the blots were probed with a mouse anti-RGS-His tag antibody (Qiagen). As secondary antibody a peroxidase-conjugated anti-mouse IgG antibody (Sigma) was used. The blots were visualized by treating them with the Western Lightning Chemiluminescence Plus reagent (PerkinElmer Life Sciences). The intensity of the immunoreactive bands was quantified densitometrically using the Phoretix 1D software package (Phoretix International, Newcastle, UK).

HPLC Analysis-- HPLC was performed on a Shimadzu system connected to a Hewlett Packard diode array detector 1040A. Reverse phase-HPLC was carried out on a Nucleosil C-18 column (Macherey-Nagel, KS system, 250 × 4 mm, 5-µm particle size) coupled with an appropriate guard column (30 × 4 mm, 5-µm particle size). For analysis of the hydroxylated fatty acids, a solvent system of methanol/water/acetic acid (80/20/0.1, v/v/v) was used at a flow rate of 1 ml/min. The chromatographic scale was calibrated for conjugated dienes by injecting known amounts of 15-HETE (5-point calibrations). The chromatogram was followed simultaneously at 235 nm (quantification of hydroxy fatty acids) and 210 nm (quantification of non-oxygenated polyenoic fatty acids). To judge the degree of membrane oxygenation, we calculated the hydroxy fatty acid/polyenoic fatty acid ratio (OH-FA/FA ratio), which relates the lipid oxygenation products to the parent fatty acids. Reference compounds (mixture of 15S-, 12S-, and 5S-HETE) were injected and chromatographed under our experimental conditions with retention times between 8 and 13 min. Straight phase-HPLC (SP-HPLC) was performed on a Zorbax-SIL column (250 × 4.6 mm, 5-µm particle size) using a solvent system of n-hexane/2-propanol/acetic acid (100/2/0.1, v/v/v) at a flow rate of 1 ml/min. For enantiomer separation of 15-HETE and 13-HODE, chiral phase-HPLC was performed on a Chiralcel OD column (250 × 4 mm, 5-µm particle size) with the solvent system n-hexane/2-propanol/acetic acid (100/5/0.1, v/v/v) and a flow rate of 1 ml/min. 9S-HODE(E,Z) and 13S-HODE(Z,E) were assigned by co-injections with authentic standards. The chemical identity of the all-E isomers was concluded from their UV spectra, which exhibited a maximal absorbance at 231 nm (the Z,E-diene chromophore of 13S- and 9S-HODE shows a _max of 235 nm).

Miscellaneous Methods-- Protein concentration was determined with the Roti-Quant detection system (Roth, Karlsruhe, Germany), which is based on the Bradford method. SMPs, which mainly constitute inside-out vesicles of the inner mitochondrial membrane, were prepared as described previously (35). The missing amino acids in the crystal structure of the rabbit reticulocyte 15-LOX were modeled in using the Swiss-PdbViewer (www.expasy.ch/spdbv; Ref. 40), followed by energy minimization of the side chains.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

The His-tagged N-terminal Truncation Mutant Exhibits 15-LOX Activity-- The rabbit reticulocyte 15-LOX consists of a N-terminal -barrel domain with a molecular mass of approximately 12 kDa and a C-terminal catalytic domain of approximately 63 kDa. Both domains share a 1600-Ų interface, and the C-terminal domain contains the catalytic non-heme iron and the entire substrate-binding pocket. These structural properties suggested that the C-terminal domain by its own may be catalytically active, but detailed truncation studies on mammalian LOXs (9) led to inactive enzyme species. To confirm these data and to test the relevance of the N-terminal -barrel domain for membrane binding, we constructed His tag bacterial expression vectors for the wild-type rabbit 15-LOX and its -barrel truncation mutant (C-terminal domain). The recombinant proteins expressed in E. coli were purified to apparent homogeneity (Fig. 1) from bacterial lysis supernatants and tested in appropriate assay systems. Surprisingly, we found that the -barrel truncation mutant was enzymatically active and converted arachidonic acid predominantly to (15S,5Z,8Z,11Z,13E)-15-hydro(pero)xyeicosa-5,8,11,13-tetraenoic acid (15S-H(p)ETE). This product pattern was almost identical to that of the wild-type enzyme (Fig. 2) and did not reveal major differences when compared with the native LOX (36). However, we observed some differences when we compared the substrate specificity and the kinetic properties of the -barrel truncation mutant with the recombinant wild-type enzyme (see below). The progress curves of arachidonic acid oxygenation (Fig. 3) indicate two kinetic peculiarities of the -barrel truncation mutant (C terminus), which were observed consistently in all measurements. (i) The -barrel truncation mutant does not exhibit a kinetic lag phase, indicating that peroxide activation kinetics may be different. (ii) It rapidly loses activity during arachidonic acid oxygenation. Because the activity could not be restored by substrate addition, one may conclude that the mutant may undergo suicidal inactivation more rapidly than the wild-type enzyme. Comparison of the basic kinetic parameters indicated that the catalytic activity (V_max) of the truncation mutant was approximately 5-fold lower than that of the wild-type enzyme (Table I), suggesting that the rate-limiting step of the catalytic process, the initial hydrogen abstraction, may be hindered. On the other hand, the comparable K_M values suggested that substrate binding is only minimally impacted by truncation. These data and the highly specific product pattern of arachidonic acid and linoleic acid oxygenation suggest that the N-terminal -barrel of mammalian 15-LOXs may not be of major importance for substrate binding if free fatty acids are offered as oxidizable substrates in the absence of biomembranes.

View larger version (29K): [in this window] [in a new window]   Fig. 1.   SDS-electrophoresis of purified rabbit 15-LOX species. The wild-type recombinant 15-LOX and various mutants were expressed as His-tagged fusion proteins in E. coli (3 liters of culture). The cells were pelleted, washed, and disrupted by sonication. The His-tagged proteins were purified by sequential open bed chromatography on nickel-agarose and Q-Sepharose (see "Materials and Methods"). Aliquots of the enzyme preparation (5 µg) were applied to electrophoresis, and proteins were visualized by Coomassie staining.

View larger version (34K): [in this window] [in a new window]   Fig. 2.   HPLC analysis of arachidonic acid oxygenation products. The wild-type rabbit 15-LOX and its -barrel truncation mutant were expressed as His-tagged fusion proteins and purified as described under "Materials and Methods." Aliquots (2 µg of the wild-type enzyme and 10 µg of the truncation mutant) were assayed for fatty acid oxygenase activity in the standard photometric system (0.5-ml reaction volume, 100 µM substrate concentration). The reaction was stopped by addition of 0.5 ml of ice-cold methanol, the hydroperoxy fatty acids formed were reduced with sodium borohydride, and aliquots were injected to RP-HPLC (see "Materials and Methods"). Inset, the 15-HETE peak in RP-HPLC was collected and the solvent was evaporated. The residue was reconstituted in hexane and analyzed for enantiomer composition by chiral phase HPLC (see "Materials and Methods"). This experiment was carried out three times, and a representative set of data is shown.

View larger version (14K): [in this window] [in a new window]   Fig. 3.   Progress curves of arachidonic acid oxygenation. The wild-type rabbit 15-LOX and its -barrel truncation mutant were expressed as His-tagged fusion protein and purified as described under "Materials and Methods." Aliquots (0.5 µg of the wild-type enzyme and 10 µg of the truncation mutant) were used to assay the arachidonic acid oxygenase activity with a Shimadzu UV2100 spectrophotometer. This experiment was carried out three times with different enzyme preparations, and representative progress curves are shown.

Deletion of the -barrel domain altered the substrate specificity of the enzyme (Table II). Among the fatty acids tested in this study, (9Z,12Z)-octadeca-9,12-dienoic acid (linoleic acid) and (8Z,11Z,14Z)-eicosa-8,11,14-trienoic acid turned out to be the best substrates for the recombinant wild-type enzyme. (5Z,8Z,11Z,14Z)-Eicosa-5,8,11,14-tetraenoic acid (arachidonic acid) and the two linolenic acid isomers ((6Z,9Z,12Z)-octadeca-6,9,12-trienoic acid and (9Z,12Z,15Z)-octadeca-9,12,15-trienoic acid) were less effectively oxygenated. In contrast, (9Z, 12Z, 15Z)-octadeca-9,12,15 trienoic acid (linolenic acid) was the preferred substrate for the -barrel truncation mutant and linoleic acid was an even worse substrate than arachidonic acid. It should, however, be stressed that the alterations in substrate specificity induced by -barrel truncation were rather quantitative, and we did not observe major qualitative differences. The pH optima of both enzyme species were also comparable (Table II).

The N-terminal -Barrel Is Not Required for Membrane Binding-- The native 15-LOX effectively interacts with biomembranes and lipoproteins (27, 28), and here we confirmed this finding for the His-tagged wild-type enzyme. This enzyme species exhibited a membrane oxygenase activity of 0.066 s¹ using submitochondrial particles as model membrane. Assuming linear reaction kinetics over the entire incubation period, a membrane oxygenase activity of 0.003 s¹ was calculated for the -barrel truncation mutant. Despite these quantitative differences, the patterns of oxygenation products were very similar, with esterified 13-H(p)ODE being the major product (Fig. 4). Although these data are compatible with the assumption that the N-terminal -barrel domain may be involved in membrane binding (6), they also indicate that this structural element may not be essential for this enzyme property.

View larger version (28K): [in this window] [in a new window]   Fig. 4.   Products formed during the oxygenation of biomembranes by the wild-type rabbit 15-LOX and its -barrel truncation mutant. The purified 15-LOX species (1.5 µg of the wild-type enzyme and 15 µg of the truncation mutant) were incubated for 10 min with SMP (500 µg of membrane proteins) in 0.5 ml of PBS at room temperature. The reaction was terminated by addition of 0.5 ml of ice-cold methanol and the hydroperoxy lipids were reduced with sodium borohydride. After alkaline hydrolysis and acidification, the precipitate was spun down and aliquots of the clear supernatants were injected to RP-HPLC (see "Materials and Methods"). The fractions containing the hydroxy fatty acids were pooled, and the solvent was evaporated. The residue was reconstituted in 0.2 ml of hexane and analyzed by SP-HPLC (see "Materials and Methods"). Upper trace, wild-type enzyme; lower trace, -barrel truncation mutant.

To compare directly the membrane binding capacities of the His-tagged wild-type 15-LOX and its -barrel truncation mutant, we established a membrane binding assay based on immunological quantification of the membrane-bound share of the two LOX-species. For this purpose the enzymes were incubated with SMPs or EKRM as model membranes. SMPs were preferred in most experiments because they have been shown previously to serve as excellent substrate for the native rabbit 15-LOX (37). After a 5-min incubation period, the particles were spun down (100,000 × g) and immunological quantification of the membrane-bound (pellet) and unbound (supernatant) shares of the enzyme was carried out by immunoblotting. From Fig. 5 it can be seen that the majority of the wild-type 15-LOX was found in the pellet fraction, indicating an efficient membrane binding. In contrast, the -barrel truncation mutant was less effectively bound. In fact, we recovered approximately 56% of the enzyme from the 100,000 × g supernatant. These data, which are in line with the results of the membrane oxygenase measurements, suggest that the N-terminal -barrel domain appears to be involved in membrane binding but may not be essential for this process. Thus, there appear to be structural determinants for membrane binding in both the N-terminal -barrel domain and in the C-terminal catalytic domain.

View larger version (85K): [in this window] [in a new window]   Fig. 5.   Membrane binding of the 15-LOX and its -barrel truncation mutant. The wild-type rabbit 15-LOX and its -barrel truncation mutant were expressed as His-tagged fusion protein and purified as described under "Materials and Methods." The membrane binding assay (see "Materials and Methods") was carried out five times with different enzyme preparations, and a representative set of experimental data is shown.

Sequence Determinants for Membrane Binding-- Because of thermodynamic reasons, most cytosolic proteins have only a limited number of hydrophobic side chains displayed on their surface. However, the reticulocyte-type 15-LOX is somewhat unusual in this respect. Taking 35% or greater as the fraction of surface for an amino acid that is solvent-exposed, we found 85 of the 663 amino acids that position their side chains toward the aqueous medium. As expected, more than 80% of them carry polar or charged residues. Surprisingly, 17 amino acids behave unusually because they present their hydrophobic side chains to the protein surface. In water solutions these side chains will get in contact with the solvent but under in vivo conditions they may possibly interact with biomembranes or other hydrophobic surfaces. Of these 17 residues, 10 (Tyr-15, Phe-70, Leu-71 (-barrel domain), Trp-181, Leu-192, Val-194, Leu-195, Leu-291, Tyr-292, Phe-412 (catalytic domain)) are distributed on a face that also contains the opening of the active site. Among them Phe-70, Leu-71 (6), and Trp-181 are most prominently solvent-exposed; thus, they constitute suitable candidates for site-directed mutagenesis. The remaining seven amino acids do not belong to this face and are scattered over the entire protein molecule. Leu-71 aligns with Trp-75 of the human 5-LOX, and this amino acid has recently been implicated in translocation of the enzyme to the nuclear envelope following cell stimulation (20). In the present study we mutated Phe-70 and Leu-71 to charged amino acids with similar steric properties (F70H, L71K, F70H/L71K) and compared the membrane binding activities of the mutants with that of the wild-type enzyme. From Table III it can be seen that the mutants exhibit a reduced membrane binding activity. However, the effects were not additive because binding activity of the double mutant was not significantly impaired when compared with the single point mutations. It is important for the interpretation of these data that the mutant enzyme species exhibited similar arachidonic acid oxygenase and biomembrane oxygenase activities as the wild-type enzyme (Table III).

View this table: [in this window] [in a new window]   Table III Membrane oxygenase activity and membrane binding of wild-type and mutant rabbit 15-LOX Wild-type and mutant rabbit 15-LOX species were purified as described under "Materials and Methods." Aliquots of the enzyme preparations were measured in the standard fatty acid oxygenase assay and membrane oxygenase assay systems (see "Materials and Methods"). Activities are given in turnover numbers (nmol of product formation/nmol of enzyme × s), the membrane oxygenase activities were calculated from a raw data obtained from a 10-min incubation period, and the wild-type LOX activity (0.066 s1) was set 100%. For the membrane-binding studies, 2.5 µg of the enzyme preparations were incubated with submitochondrial particles as described under "Materials and Methods." The membrane particles were spun down, and the shares of bound and unbound LOX species were quantified by immunoblotting using an anti-His tag antibody. The membrane binding assay was carried out three times with different enzyme preparations.

Our observations that the -barrel truncation mutant exhibits both membrane binding and membrane oxygenase activities prompted us to search for potential binding determinants in the catalytic domain, that may be able to act in concert with Phe-70 and Leu-71. Inspecting the surface of the rabbit 15-LOX for exposed hydrophobic amino acids, which may interact with these two residues, we identified Trp-181 as a potential candidate, which is a prominent part of the face of hydrophobic amino acids surrounding the substrate binding site opening. Unfortunately, the electron density for residues 177-188 was ambiguous and thus the exact position of Trp-181 is not well defined. However, structural modeling suggested a helical structure for these residues (6) with Trp-181 being exposed at the protein surface. To find out whether this residue may be important for membrane binding, this space-filling hydrophobic amino acid was mutated to a bulky but charged glutamate (W181E) as well as to a small but uncharged alanine (W181A). When expressed as His-tagged fusion proteins, both mutants were enzymatically active. However, under identical experimental conditions, the catalytic activity (arachidonic acid turnover) of the W181E mutant was 1 order of magnitude lower than that of the wild-type enzyme and of the W181A mutant (Table IV). Both mutants converted arachidonic acid to 15S-H(p)ETE, and the major product of biomembrane oxygenation was esterified 13S-H(p)ODE (data not shown). As predicted, the W181E mutant exhibited a strongly impaired membrane binding activity because approximately 80% of the enzyme was recovered from the supernatant in our membrane binding assay. In contrast, the majority (>80%) of the wild-type enzyme was membrane-bound (Fig. 6). With the W181A mutant, the alterations were not as dramatic (Fig. 6), but here again we observed a reduced share of membrane association (60%).

View larger version (102K): [in this window] [in a new window]   Fig. 6.   Membrane binding of various 15-LOX point mutants. The wild-type rabbit 15-LOX and the different point mutations were expressed as His-tagged fusion protein and purified as described under "Materials and Methods." The membrane binding assay (see "Materials and Methods") was carried out three times, with each mutant yielding similar results. A representative set of data is shown.

Summarizing these data one may conclude that Phe-70 and Leu-71 localized in the -barrel domain and Trp-181 of the catalytic domain contribute to the membrane binding activity of the enzyme. To find out whether both regions may act in concert, we created the F70H/W181A double mutant and compared their membrane binding activity with that of the wild-type enzyme and the two single mutants. For this comparison the W181A exchange was selected because the W181E mutation virtually wiped out the membrane binding activity of the enzyme and, thus, a possible contribution of Phe-70 could have been masked. The F70H/W181A double mutant exhibited an impaired membrane binding activity when compared with the two single mutants and with the wild-type enzyme (Table V). These data suggest that the effects of F70H and W181A exchange appear to be additive and that both determinants may act in concert.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Plant and mammalian LOXs are two-domain enzymes consisting of a large C-terminal domain, which is mainly helical, and a small N-terminal -barrel domain. The C-terminal domain contains catalytically important constituents and, thus, may be considered the catalytic subunit. However, previous attempts failed to create functional -barrel truncation mutants. To judge these findings one must consider that N-terminal deletions are always critical because there is a high risk for protein misfolding. To reduce this risk we expressed the C-terminal domain of the rabbit 15-LOX as a His-tagged fusion protein and found it to be enzymatically active. Except for the 5-fold impaired catalytic activity, the -barrel truncation mutant exhibited similar enzymatic properties as the wild-type enzyme. These data indicate for the first time that the N-terminal -barrel domain is not essential for the catalytic activity of mammalian LOXs and that most of the important enzymatic properties are retained in -barrel truncation mutants. Similar conclusions have recently been drawn from proteolytic studies carried out on the soybean LOX-1 (10).

Sequence alignments of the rabbit 15-LOX with structurally related proteins revealed similarities of its -barrel domain with a C-terminal -barrel of the human pancreatic lipase (34). These modeling studies suggested that the N-terminal domain of LOXs might be responsible for membrane binding. However, we found that the -barrel truncation mutant retained some membrane binding activity, indicating that the -barrel may not be essential for this enzyme property. When compared with the wild-type enzyme, membrane binding was impaired for the -barrel truncation mutant and, thus, one may conclude that with the wild-type enzyme the -barrel domain may contribute to membrane binding.

The -barrel domain of the human 5-LOX has recently been implicated in calcium-dependent membrane binding (20), and a search for amino acids involved in membrane-translocation identified three surface-exposed tryptophans (Trp-13, Trp-75, Trp-102) as possible sequence determinants of membrane binding. Site-directed mutagenesis of these amino acids impaired the membrane binding activity of this protein to phospholipid vesicles (20). To find out whether these residues may also be involved in membrane binding of the rabbit 15-LOX, we carried out an amino acid alignment and obtained the following results. Trp-13 of the human 5-LOX aligns with Ile-14 of the rabbit 15-LOX, Trp-75 aligns with Leu-71, and Trp-102 aligns with Trp-100. If one determines the position of these residues in the crystal structure of the rabbit 15-LOX (no x-ray data available for the human 5-LOX), one may conclude that Ile-14 and Trp-100 are located at the interdomain interface of the N-terminal -barrel. Thus, these residues may be important for interdomain interaction but are unlikely to impact membrane binding of the rabbit enzyme. In contrast, Leu-71 and the adjacent Phe-70 are clearly surface-exposed (Fig. 7) and, thus, may play a role for membrane binding. We mutated Phe-70 and Leu-71 to charged counterparts with spatial properties similar to those of the original residues and observed impaired membrane binding properties. These data suggest that these residues may be important for membrane binding.

View larger version (167K): [in this window] [in a new window]   Fig. 7.   Localization of the sequence determinants for membrane binding in the crystal structure of the rabbit 15-LOX. The crystal structure of the rabbit reticulocyte 15-LOX is shown. The set of x-ray coordinates is incomplete. No data are available for Cys-210, Gly-211, Glu-601, Pro-602, and residues 177-187 (6). Structural modeling indicated that the latter residues form a helix, and this structural element as well as the side chains for other lacking amino acids were modeled into the x-ray structure. The sequence determinants of membrane binding are represented as CPK models.

The -barrel truncation mutant of the rabbit 15-LOX also exhibits a membrane-oxygenase activity, but it was unclear which parts of the catalytic subunit may be involved in enzyme-membrane interaction. Potentially important amino acids were assumed to be located in proximity to the -barrel domain (in particular to Phe-70, Leu-71) but also close to the entry into the substrate-binding pocket. Searching the three-dimensional structure of this enzyme for possible secondary-structural elements that fulfill these requirements, we conclude that helix 2 may be a suitable candidate. Unfortunately, the x-ray coordinates for this helix (amino acids 177-188) have not been determined crystallographically, but structural modeling suggested a surface-exposed Trp-181. Because solvent-exposed tryptophans are quite unusual in most proteins, we hypothesized that this hydrophobic amino acid may act in concert with Phe-70 and Leu-71 (Fig. 7) and possible other hydrophobic residues to anchor the enzyme at biomembranes via hydrophobic interactions. Our finding that mutation of this apolar amino acid to a charged glutamate strongly impaired the membrane binding activity indicates that Trp-181 is likely to be involved in the membrane binding.

As indicated above, other LOX isoforms, such as the human 5-LOX (13, 20), the cucumber lipid body LOX (38), and the soybean 15-LOX-1 (10, 39), are also capable of binding to biomembranes. For the 5-LOX (20) and the cucumber (38) linoleate 13-LOX, it has been suggested that their N-terminal -barrel domains are responsible for the membrane binding activities. In contrast, proteolytic -barrel truncation of the soybean LOX-1 led to an increased membrane binding affinity, indicating that the N-terminal domain may even impair the membrane binding capacity (10). We found that, for optimal membrane binding of the rabbit reticulocyte-type 15-LOX, a concerted action of the N-terminal -barrel domain and the C-terminal catalytic domain is required. These data suggest that the relative contribution of the N-terminal -barrel domain to membrane binding capacity may be variable and may depend on the LOX isoform.

	FOOTNOTES

^* This work was supported by Deutsche Forschungsgemeinschaft Grant Ku 961/7-1, by National Institutes of Health Grant HL60889 (to M. W.), and by a grant from the DeWitt Wallace Fund.The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

To whom correspondence may be addressed. Tel.: 49-30-450528040; Fax: 49-30-450528905; E-mail: hartmut.kuehn@charite.de.

Published, JBC Papers in Press, May 9, 2002, DOI 10.1074/jbc.M203234200

	ABBREVIATIONS

The abbreviations used are: LOX, lipoxygenase; 15S-H(p)ETE, (15S,5Z,8Z,11Z,13E)-15-hydro(pero)xyeicosa-5,8,11,13-tetraenoic acid; 12S-H(p)ETE, (12S,5Z,8Z,10E,14Z)-12-hydro(pero)xyeicosa-5,8,10,14-tetraenoic acid; 13S-H(p)ODE, (13S,9Z,11E)-13-hydro(pero)xyoctadeca-9,11-dienoic acid; RP, reverse phase; SP, straight phase; HPLC, high performance liquid chromatography; DTT, dithiothreitol; SMP, submitochondrial particle; PBS, phosphate-buffered saline; EKRM, EDTA-high salt-stripped rough microsome.

	REFERENCES

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				Y. Zheng, W. E. Boeglin, C. Schneider, and A. R. Brash A 49-kDa Mini-lipoxygenase from Anabaena sp. PCC 7120 Retains Catalytically Complete Functionality J. Biol. Chem., February 22, 2008; 283(8): 5138 - 5147. [Abstract] [Full Text] [PDF]