a -Tocopherol Inhibits the Respiratory Burst in Human Monocytes ATTENUATION OF p47 phox MEMBRANE TRANSLOCATION AND PHOSPHORYLATION*

Vitamin E ( a -tocopherol), one of the most important natural antioxidants, is assumed to be beneficial in the prevention of cardiovascular diseases. a -Tocopherol ex-hibits acyl-peroxyl-radical scavenger properties and ex-erts cell-mediated actions in the hemovascular compart-ment, such as inhibition of superoxide anion (O 2 .) production by leukocytes. The aim of this study was to examine the mechanism underlying the inhibitory effect of a -tocopherol on O 2 . production by human monocytes. In activated monocytes O 2 . is produced by the NADPH-oxidase enzyme complex. The oxidase activation elicited by phorbol myristate acetate (PMA) requires membrane translocation of several cytosolic factors. We found that in human PMA-stimulated adherent monocytes, a -to-copherol (but not b -tocopherol) inhibited O 2 . production in intact cells but had no effect on a membrane preparation containing activated NADPH-oxidase, suggesting that a -tocopherol impairs the assembly process of the enzyme complex. We showed that translocation and phosphorylation of the cytosolic factor p47 phox were re-duced in monocytes preincubated with a -tocopherol. We


In activated monocytes O 2
. is produced by the NADPHoxidase enzyme complex. The oxidase activation elicited by phorbol myristate acetate (PMA) requires membrane translocation of several cytosolic factors. We found that in human PMA-stimulated adherent monocytes, ␣-tocopherol (but not ␤-tocopherol) inhibited O 2 . production in intact cells but had no effect on a membrane preparation containing activated NADPH-oxidase, suggesting that ␣-tocopherol impairs the assembly process of the enzyme complex. We showed that translocation and phosphorylation of the cytosolic factor p47 phox were reduced in monocytes preincubated with ␣-tocopherol. We verified that the tryptic phosphopeptide map of monocyte p47 phox was similar to that of neutrophil p47 phox , indicating that several serine residues were phosphorylated. Peptides whose phosphorylation is dependent on protein kinase C (PKC) were phosphorylated to a lesser degree when p47 phox was immunoprecipitated from ␣-tocopherol-treated monocytes. In vitro, the activity of PKC from monocytes was inhibited by ␣-tocopherol in a specific manner compared with that of ␤-tocopherol or Trolox®. Membrane translocation of PKC was not affected. These results show that ␣-tocopherol inhibits O 2 . production by human adherent monocytes by impairing the assembly of the NADPH-oxidase and suggest that the inhibition of phosphorylation and translocation of the cytosolic factor p47 phox results from a decrease in PKC activity.
Oxidative modification of low density lipoprotein (LDL) 1 ap-pears to be a key event in the early stages of atherogenesis (1). Monocyte release of superoxide anion (O 2 . ), a reactive oxygen species (ROS), induces LDL oxidation (2). Vitamin E (␣-tocopherol), a lipophilic molecule present in plasma lipoproteins, is one of the most important natural antioxidants (3). It is able to scavenge acyl-peroxyl radicals in membrane structures and may prevent or delay cardiovascular diseases (4,5), possibly through its capacity to increase LDL resistance to oxidative modification (6). The possibility that ␣-tocopherol has additional antioxidant effects by inhibiting ROS generation has received little attention. We (7) and others (8 -10) have shown that ␣-tocopherol inhibits O 2 . production by human monocytes (7), rat macrophages, and neutrophils (8 -10), but the mechanism(s) underlying this inhibitory effect are unknown. The system responsible for O 2 . production in phagocytic cells is the multicomponent enzyme NADPH-oxidase. This complex includes membrane-bound cytochrome b 558 and cytosolic proteins (p47 phox , p67 phox , Rac1/2, and p40 phox ) (11) that translocate to the membrane during stimulation to form a catalytically active oxidase (12). During NADPH-oxidase activation, p47 phox is phosphorylated on several serine residues (13). Protein kinase C (PKC) is involved in NADPH-oxidase activation and can phosphorylate p47 phox (14,15). PKC is a family of Ca 2ϩ -and phospholipid-dependent serine/threonine kinases that play a pivotal role in agonist-stimulated cell functions (16). Several PKC isoforms have been described (17). The conventional PKCs ␣PKC and ␤PKC are present in monocytes and are activated by phorbol esters such as phorbol 12-myristate 13-acetate (PMA) (18). In human monocytes and neutrophils, PMA induces the production of O 2 . and the phosphorylation of several proteins, one of which is p47 phox (14). The purpose of this study was to explore the underlying mechanism of ␣-tocopherol-induced inhibition of O 2 . production by monocytes. We found that ␣-tocopherol treatment of human adherent monocytes inhibited PMA-induced translocation and phosphorylation of the cytosolic oxidase component p47 phox .

EXPERIMENTAL PROCEDURES
Materials-dl-␣-Tocopherol, dl-␤-tocopherol, and PMA were from Sigma; SDS-PAGE reagents were from Bio-Rad; and [ 32 P]orthophosphate and [␥-32 P]ATP were from NEN Life Science Products. Anti-p47 phox antibody was a kind gift from Dr. Babior (Scripps Research Institute). Anti-PKC antibody was from Santa Cruz Biotechnology.
Monocyte Preparation-Mononuclear cells were obtained from healthy subjects by dextran sedimentation and Ficoll-Paque fractionation of freshly drawn blood. The cells were resuspended in RPMI medium ϩ 10% fetal calf serum and then plated in Petri dishes for 2 h at 37°C (20 ϫ 10 6 cells/10 ml), in humidified air with 5% CO 2 . The adherent monocytes remaining after washing with RPMI (about 5 ϫ 10 6 cells/dish) were incubated with ␣-tocopherol or ethanol in basal conditions for 30 min and then stimulated with PMA (30 ng/ml) for 10 min. For the phosphorylation experiments, cells were resuspended in phosphate-free Dulbecco's modified Eagle's medium, treated with 2.7 mM * The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
§ Recipient of a grant from Association pour la Recherche Contre le Cancer.
Superoxide Production Measurement-The effect of ␣-tocopherol on O 2 . production was measured in a lucigenin-enhanced chemiluminescence assay. Adherent monocytes purified in a 48-well plate were preincubated with various amounts of ␣-tocopherol or its ethanol vehicle for 30 min before stimulation with 60 ng/ml PMA in the presence of 100 M lucigenin. Luminescence was measured by an ultrasensitive photoncounting imaging camera monitored by a computer-assisted image processor (Argus 100, Hamamatsu Photonics, Japan). Photon emission was recorded for 60 min after PMA stimulation and was proportional to O 2 . production (7). We also tested the effect of ␣-tocopherol on an acellular system producing O 2 . , by using the superoxide dismutaseinhibitable cytochrome c reduction method (19). The membrane fraction containing active NADPH-oxidase was obtained by sonication of stimulated human neutrophils or monocytes (1 g/ml PMA) in 0.34 M sucrose ϩ 1 mM PMSF and ultracentrifugation (100,000 ϫ g for 30 min at 4°C in a Beckman TL 100). The pellet (membrane fraction) was solubilized in 0.34 M sucrose. The production of O 2 . was measured (20) in the presence of ␣-tocopherol or ethanol. p47 phox and PKC Translocation-Resting and stimulated cells were scraped from the Petri dishes (5 ϫ 10 6 cells/ml), sonicated on ice (2 ϫ 10 s) in relaxation buffer (100 mM KCl, 3 mM NaCl, 3.5 mM MgCl 2 , 10 mM PIPES, 1 mM EGTA, 10 g/ml pepstatin, 10 g/ml leupeptin, 0.5 mM PMSF), and centrifuged (600 ϫ g for 10 min at 4°C) to remove nuclei and unbroken cells; the supernatant was then ultracentrifuged (100,000 ϫ g for 30 min at 4°C). Pellets were washed in relaxation buffer, dissolved in Laemmli sample buffer, and submitted to 10% SDS-PAGE (21) and then blotted onto nitrocellulose using the Towbin protocol (22). The upper part of the gel was revealed with anti-PKC antibody (1/1000 dilution), and the lower part with anti-p47 phox (1/5000) and labeling by secondary antibodies (1/2000), followed by ECL detection.
Two-dimensional Tryptic Phosphopeptide Mapping-The nitrocellulose area containing 32 P-labeled p47 phox was incubated for 30 min at 37°C with polyvinylpyrrolidone, washed five times with water and twice with 50 mM ammonium bicarbonate, and then incubated overnight with trypsin (15). After washing, peptides were dissolved in electrophoresis buffer and applied to a cellulose thin layer plate. Electrophoresis and chromatography were carried out as described previously (13). The plate was analyzed in an Instant Imager apparatus equipped with Instant Imager software (Packard).
Assay of PKC Activity-Resting adherent monocytes were scraped in lysis buffer (20 mM Tris-HCl, pH 7.5, 2 mM EDTA, 10 mM EGTA, 0.3 M sucrose, 2 mM dithiothreitol, 2 mM PMSF, 10 g/ml leupeptin, 10 g/ml pepstatin) at 3 ϫ 10 7 cells/ml. They were then sonicated and ultracentrifuged to obtain the cytosolic fraction as described above. PKC activity in aliquots of cytosolic fraction was measured by PS/DG-stimulated histone III-S phosphorylation. Activating liposomes containing ␣or ␤-tocopherol were prepared as follows. PS and DG were dissolved in CHCl 3 and then mixed with tocopherol in ethanol; the mixture was dried under N 2 , resuspended in 20 mM Tris-HCl, pH 7.5, and sonicated for 2 min on ice under N 2 . Histone III-S (200 g/ml) was incubated with the PKC-containing cell fraction in reaction buffer (20 mM Tris-HCl, pH 7.5, 10 mM MgCl 2 ) with 1 mM CaCl 2 and lipids (final concentrations 200 g/ml PS, 10 g/ml DG, 20 g/ml tocopherol). Phosphorylation was performed by adding 10 M ATP containing 0.1 Ci of [␥-32 P]ATP/ml for 10 min at 30°C and stopped on P81 filter paper. After extensive washing in 0.8% phosphoric acid, the amount of remaining 32 P-histone was measured by scintillation counting. Specific activity was the difference between activity in the presence and absence of Ca 2ϩ , PS, and DG. When Trolox was used it was added to the reaction mixture at the same molarity as ␣-tocopherol (46 M).

␣-Tocopherol Inhibits O 2 . Production in Intact Human
Adherent Monocytes-PMA-stimulated production of O 2 . by monocytes was depressed in the presence of ␣-tocopherol in a concentration-dependent manner (Fig. 1). The inhibition was not complete, reaching a maximum of 50% at 45 g/ml ␣-tocopherol. It is noteworthy that the basal activity of unstimulated monocytes (representing 15 Ϯ 5% (n ϭ 4) of the activity of stimulated cells) was not inhibited by ␣-tocopherol. O 2 . production by xanthine/xanthine oxidase was not decreased by ␣-tocopherol (data not shown), suggesting that inhibition did not result from scavenging of O 2 . . This lack of O 2 . scavenging was previously found even in the presence of LDL (7), showing that it is not due to the absence of a lipid phase in the medium. ␣-Tocopherol also failed to decrease O 2 . production in a membrane preparation containing NADPH-oxidase (Fig. 1). This showed that ␣-tocopherol did not directly interfere with activated NADPH-oxidase but may instead have inhibited an event preceding oxidase activation. In contrast to this depressive effect of ␣-tocopherol, ␤-tocopherol enhanced O 2 . production by monocytes. 2 These results provide additional evidence for an action of ␣-tocopherol that is independent of its scavenging properties and suggest that ␣-tocopherol has a specific inhibitory effect. ␣-Tocopherol Inhibits p47 phox Translocation-As p47 phox translocation is a key event in NADPH-oxidase activation (23), we investigated the effect of ␣-tocopherol on this process by using Western blot analysis of the membrane fractions. We chose a concentration of ␣-tocopherol that clearly inhibited O 2 . production (Fig. 1). When monocytes were stimulated with PMA, pretreatment with 20 g/ml ␣-tocopherol inhibited membrane translocation of p47 phox (Fig. 2). In resting monocytes, p47 phox was barely detectable in the membrane fraction, and ␣-tocopherol alone did not induce detectable movement of p47 phox between the cytosolic and membrane compartments (Fig. 2). Western blots were scanned, and the combined densitometry data indicated that in the presence of ␣-tocopherol the level of p47 phox translocation was 70.9 Ϯ 4.3% of control (n ϭ 4)(p Ͻ 0.05).
PMA is a direct agonist of conventional and novel isoforms of PKC (17). The previous results with PMA suggested that PKC is involved in p47 phox translocation and phosphorylation. In addition, PKC phosphorylates p47 phox at several sites (13). The PKC pathway was therefore a possible target for the action of ␣-tocopherol. To determine whether ␣-tocopherol inhibited specific phosphorylation sites, we carried out two-dimensional phosphopeptide mapping of p47 phox . The pattern of tryptic p47 phox phosphopeptides obtained in human adherent monocytes under PMA stimulation (Fig. 4) was close to that of human neutrophils (15). ␣-Tocopherol inhibited the phosphorylation of the major PKC-dependent sites (shown by arrows), whereas two phosphopeptides that are phosphorylated in resting cells (broken arrows) were not affected by ␣-tocopherol. This provides further evidence that the inhibitory effect of ␣-tocopherol is mediated by PKC.
PKC Translocation Is Insensitive to ␣-Tocopherol-To assess whether ␣-tocopherol inhibits p47 phox translocation and phosphorylation by modulating PKC activation, we used Western blot analysis to visualize the translocation of PKC from cytosol to membrane. The antibody used to reveal PKC cross-reacted with ␣ and ␤ isoforms of PKC present in monocytes (18) . Fig. 5 shows the massive translocation of PKC from cytosol to membrane after PMA stimulation, with only a trace of PKC remaining in the cytosolic fraction. ␣-Tocopherol did not modify this process (100.5 Ϯ 4.5% of the control (n ϭ 3) as determined by densitometry analysis). It did, however, inhibit p47 phox membrane translocation in the same cell extracts. Western blot analysis of p47 phox in the cytosols confirmed that equivalent amounts of cells were used (Fig. 5). Taken together, these results suggest that ␣-tocopherol interferes with the activity of membrane-bound conventional PKC rather than with its membrane translocation but do not exclude an action of ␣-tocopherol on other PKC isoforms.
␣-Tocopherol Inhibits PKC Activity in Vitro-We then tested the effect of ␣-tocopherol on PKC activity. A PKC preparation from the cytosolic fraction was incubated in the presence of PS/DG liposomes, with or without incorporated ␣-tocopherol. Fig. 6 shows that ␣-tocopherol strongly inhibited PKC activity, whereas ␤-tocopherol was less effective and Trolox had a weak effect; this is consistent with a specific action of ␣-tocopherol.

DISCUSSION
In this study ␣-tocopherol depressed the PMA-induced respiratory burst of human adherent monocytes without affecting the assembled oxidase activity of membrane preparations originating from PMA-activated monocytes or neutrophils. This strongly suggests that the ␣-tocopherol inhibition of O 2 . production is at least in part related to a functional impairment of the NADPH-oxidase assembly. Indeed, we found that ␣-tocopherol inhibited p47 phox translocation to the membrane and impaired p47 phox phosphorylation. In human neutrophils, oxidase activation by PMA is believed to be mediated by PKC (15,24). The results described here in human monocytes suggest that the PKC pathway is involved in O 2 . production, as PMA-induced p47 phox phosphorylation was inhibited by GF109203X (Fig. 3A), a potent inhibitor of PKC (25). We further showed the following: (i) ␣-tocopherol inhibited PMA-induced but not basal O 2 . production; (ii) ␣-tocopherol inhibited PMA-induced p47 phox translocation and phosphoryl- FIG. 4. Effect of ␣-tocopherol (␣-toc) on the phosphorylation of p47 phox as analyzed by tryptic phosphopeptide mapping. p47 phox was labeled with 32 P and isolated as described in Fig. 3. The protein bound to the nitrocellulose membrane was submitted to tryptic digestion. Phosphopeptides were analyzed by two-dimensional separation as described under "Experimental Procedures" and detected by an Instant Imager apparatus. See "Results" for arrow legends. Repeated three times.

␣-Tocopherol Depresses p47 phox Translocation and Phosphorylation
ation; and (iii) the peptides whose phosphorylation was inhibited by ␣-tocopherol are known PKC substrates (15). These results suggest that the PKC pathway is a target of ␣-tocopherol in human monocytes. The lack of effect of ␣-tocopherol on basal activity of unstimulated cells suggests that PKC isoforms that can be activated by PMA are sensitive to ␣-tocopherol.
The mechanism by which ␣-tocopherol interacts with PKC is unclear. In our model ␣-tocopherol failed to inhibit PKC translocation to the membrane (Fig. 5), suggesting that it modulates only the activity of membrane-bound enzyme. In vitro, ␣-tocopherol inhibited monocyte PKC activity only when ␣-tocopherol was incorporated in PS/DG liposomes (Fig. 6). There was no action when ␣-tocopherol was supplied as a solution in ethanol (data not shown) or when it was replaced by Trolox, a hydrophilic analog of ␣-tocopherol that is unable to penetrate liposomes but has a preserved antioxidant moiety (Fig. 6). These results suggest that ␣-tocopherol incorporation into the lamellar (pseudo-membrane) structure of liposomes is a prerequisite for its inhibitory action on PKC. Tocopherol compounds have been shown to alter the structural organization of liposomes (26,27) and would thus be potentially able to modify PKC activity. Several lipids known to incorporate into membrane structures are able to modulate PKC activity (28) and O 2 . production in human adherent monocytes stimulated by PMA (29). The present results support a membrane location for the interaction between PKC and ␣-tocopherol, as ␣-tocopherol (9,30) and activated PKC (Fig. 5) are both located in the membrane fraction. Our results further suggest an additional specific effect of ␣-tocopherol on PKC, as ␣-tocopherol inhibited PKC activity much more strongly than did ␤-tocopherol (Fig.  6). Taken together these results suggest that the inhibition of PKC activity (i) is not directly due to the antioxidant capacity of ␣-tocopherol, (ii) requires the integration of ␣-tocopherol in a pseudo-membrane structure, and (iii) is due to an interaction between the ␣-tocopherol molecular structure and PKC. In addition, as ␣-tocopherol inhibited both PKC activity and superoxide production whereas ␤-tocopherol and Trolox did not (this report and Ref, 10), these data support a link between the effect of ␣-tocopherol on PKC activity and O 2 . production.
Vitamin E is a natural lipophilic antioxidant compound that protects lipid structures from peroxidation. It is the first antioxidant to be consumed during oxidative modification of LDL (31). Epidemiological studies and intervention trials have suggested a potential effect of vitamin E in the prevention of atherosclerosis (5,32). This beneficial role has been defined as the ability of vitamin E to increase LDL resistance to oxidation (6), LDL oxidation being a key event in atherogenesis (33). Vitamin E may also protect LDL by attenuating the respiratory burst, as NADPH-oxidase activity and O 2 . production by monocytes and macrophages are able to induce LDL oxidation (2,34). Monocytes from volunteers taking oral vitamin E showed lower O 2 . production together with an impaired capacity to oxidize LDL (8). Similarly, J774 macrophages supplemented in vitro with vitamin E lost their power to oxidize LDL (35).
Our results provide a model for a monocyte-mediated effect of vitamin E in preventing LDL oxidation; following PKC inhibition, p47 phox translocation and phosphorylation are inhibited, thereby decreasing superoxide production. As well as elucidating the way in which vitamin E delays and/or decreases oxygen-radical injury of LDL, this cellular model provides new insights into the potential beneficial effect of vitamin E on atherogenesis.
FIG. 5. ␣-Tocopherol does not modify PKC translocation in PMA-treated human monocytes. Cells were treated as described in Fig. 1. After SDS-PAGE and blotting of cytosol and membrane fractions, PKC and p47 phox were detected with their respective antibodies and revealed with the ECL technique. Concentrations used are as follows: PMA 30 ng/ml; ␣-tocopherol (␣-toc) 20 g/ml. Data are representative of five experiments.
FIG. 6. Effect of ␣-tocopherol, ␤-tocopherol, and Trolox on PKC activity of monocyte cytosol. Cytosol fraction was separated from resting monocytes by sonication and ultracentrifugation. PKC activity was measured by PS/DG-and Ca 2ϩ -dependent transfer of [ 32 P]ATP to histone III-S. ␣and ␤-tocopherol were integrated in PS/DG liposomes, with a final concentration in the assay mixture of 20 g/ml. Trolox was added at 11.6 g/ml (46 M). Activity was measured on 6.5 g of cytosolic protein. Values are expressed as the percentage of control activity (mean Ϯ S.E., n ϭ 3). Control (100%) corresponds to 6080.35 Ϯ 290.4 pmol of 32 P/mg of protein/min.