Metabolic activation of eugenol by myeloperoxidase in

Eva Kiorpelidou , Brian Foster , John Farrell , Monday O. Ogese , Lee Faulkner , Chris E. Goldring , B. Kevin Park , and Dean J. Naisbitt. Chemical Re...
10 downloads 0 Views 828KB Size
Chem. Res. Toricol. 1989, 2, 186-192

186

Metabolic Activation of Eugenol by Myeloperoxidase and Polymorphonuclear Leukocytes David Thompson,**+Despina Constantin-Teodosiu,t Kajsa Norbeck,? Bjorn Svensson,T and Peter Mold6ust Department of Toxicology, Karolinska Institutet, Box 60400, S-104 01 Stockholm, Sweden, and Research and Development Laboratories, Astra Lakemedel AB, Sodertalje, Sweden Received October 4, 1988

Eugenol has recently been associated with the toxic effects of clove cigarettes on human lungs. We have studied the metabolism and adverse effects of eugenol on human polymorphonuclear leukocytes (PMNs). Myeloperoxidase, isolated and purified from human PMNs, catalyzed the oxidation of eugenol to a reactive intermediate which is likely to be a quinone methide. Eosinophil peroxidase, lactoperoxidase, prostaglandin H synthase, horseradish peroxidase, and rat intestinal peroxidase also supported this hydrogen peroxide dependent reaction. Glutathione inhibited the formation of this metabolite, resulting in the formation of glutathione disulfide and a small amount of eugenol-glutathione conjugates. In cellular incubations, phorbol ester stimulated PMNs catalyzed the covalent binding of [3H]eugenol to cellular protein, which was partially inhibitable by azide. Intracellular glutathione levels decreased by 90% over a period of 30 min in phorbol ester stimulated PMNs exposed to 100 pM eugenol compared with decreases of 30% (phorbol ester alone) or 5 % (eugenol alone) in control incubations. In addition, eugenol was more cytotoxic to PMNs in the presence of phorbol ester than in its absence, and eugenol inhibited the phorbol ester stimulated oxidative burst in PMNs as reflected by a decrease in oxygen consumption, superoxide formation, and hydrogen peroxide formation. These results suggest that PMNs are capable of activating eugenol to a reactive intermediate and also suggest a mechanism whereby eugenol can potentially interfere with and adversely affect vital P M N functions.

Introduction Peroxidase enzymes are capable of activating xenobiotics to reactive intermediates which are potentially toxic to biological tissues ( I ) . Myeloperoxidase, a main component of granules in polymorphonuclear leukocytes (PMNs), is a mammalian peroxidase that metabolizes hydrogen peroxide in the presence of halide to form cytoxic products which can kill invading microorganisms, injure host cells, and inactivate humoral factors (2). In addition, myeloperoxidase (directly or by implication using PMNs) has been shown to metabolically activate several xenobiotic chemicals, including carcinogens, to potentially damaging reactive intermediates. These include phenol, l-naphthol, catechol, estradiol, bleomycin At, phenytoin, procainamide, N-methylaminoazobenzene, benzidine, aminofluorine, and benzo[a]pyrene-7,8-dihydrodiol(3-10). The activation of xenobiotics by PMNs has important toxicologic implications. For example, the reactive metabolites produced may be cytotoxic, mutagenic, or carcinogenic to the PMNs or surrounding cells or tissues and may provide a direct link in the relationship between inflammation and cancer (11). Kensler et al. (12) have recently demonstrated that a significant increase in the covalent binding of benzo[a]pyrene-7,8-dihydrodiolin epidermal DNA occurs in mouse skin in the presence of increased numbers of PMNs compared to mouse skin in which an inflammatory response was not invoked.

Eugenol is a naturally occurring phenolic compound that has enjoyed widespread use as a food flavor and fragrance agent (13). It is also used in dentistry as an analgesic component of various dental materials. Eugenol is a major component of clove cigarettes (60-70% tobacco and 30-40% cloves) and recently has been associated with severe, acute pulmonary illnesses in humans (14). The major symptoms included pulmonary edema, bronchospasm, and hemoptysis. Of particular interest is the mention of prodromal respiratory tract infections occurring with some of the more severe cases. LaVoie et al. (25) reported an acute toxic effect on the lungs of rats given eugenol intratracheally. Eugenol had an LD,, of 11mg/kg and elicited pulmonary edema and hemoptysis. Fotos et al. (16)have recently reported that eugenol inhibits human PMN cell migration and chemiluminescence while Suzuki et al. (17)reported cytotoxic effects of high concentrations of eugenol on guinea pig PMNs. We have recently observed that horseradish peroxidase can activate eugenol to a reactive intermediate that covalently binds to protein and glutathione and is also cytotoxic (18). These observations led us to examine the metabolism and possible toxic effects of eugenol on PMNs. We demonstrate in this report that myeloperoxidase and PMNs can metabolically activate eugenol to a potentially toxic reactive intermediate and that eugenol adversely affects vital PMN functions.

*To whom correspondenceshould be addressed at the Laboratory

Materials. Myeloperoxidase and eosinophil peroxidase were isolated from human blood and purified as previously described (19). Prostaglandin H synthase was obtained from Oxford Research Laboratory (Oxford, MI). Rat intestinal peroxidase was prepared as described by Kimura and Jellinck (20). The following

of Molecular Biophysics, National Institute of Environmental Health Sciences, P.O. Box 12233, Research Triangle Park, NC 27709.

Karolinska Institutet.

* Astra Lakemedel AB.

Materials and Methods

0893-228~/S9/2~02-0186$01.50/0 0 1989 American Chemical Society

Oxidation of Eugenol by PMNs chemicals and enzymes were obtained from Sigma Chemical Co. (St. Louis, MO): glutathione (oxidized and reduced), hydrogen peroxide (30% solution), eugenol, cytochrome c, sodium azide, phorbol myristate acetate (PMA), ferroammonium sulfate, potassium thiocyanate, silver(1) oxide, lactoperoxidase, superoxide dismutase, and horseradish peroxidase (type VI). Monobromobimane was ordered from Calbiochem-Behring (La Jolla, CA). Acetylated cytochrome c was prepared as described by Azzi et al. (21). [SH]Eugenol (3.4 gCi/gmol), labeled on the methoxy group, was synthesized as previously described (18) while [3H]glutathione (58 Ci/mmol) was purchased from New England Nuclear (Boston, MA). All other chemicals were of the highest grade available commercially. Peroxidase Assays. The peroxidase-dependent oxidation of eugenol was monitored spectrophotometrically as described previously (18)by measuring the change in absorbance at 350 nm. The increase in absorbance a t this wavelength is thought to be due to the formation of a quinone methide metabolite. Incubations generally contained 0.5 mM eugenol, 0.25 mM hydrogen peroxide, and various concentrations of peroxidase in a total volume of 1 mL of 0.1 M phosphate buffer, p H 7. In some incubations with myeloperoxidase, various concentrations of glutathione were added at the beginning of the reaction or during the course of the reaction. The fate of glutathione in these reactions was followed by measuring the loss of glutathione (22) and the formation of glutathione disulfide (23). The formation of eugenol-glutathione conjugates was measured by HPLC. Chromatography was carried out on a Waters system (Milford, MA) with a 990 photodiode array detector. Samples were analyzed by using a Beckman Ultrasphere ODS column (5 gm) and a methanol gradient. Buffer A consisted of 0.25% perchloric acid and 0.25% acetic acid adjusted to pH 3.5 with 10 M sodium hydroxide. Buffer B was 100% methanol. Initial conditions were 80% A. After 20 min, a linear gradient was used to change from 80% A to 50% A over the next 15 min. Fifty percent A was maintained from 35 until 50 min, when the initial conditions were restored. The flow rate was 2 mL/min throughout the entire run. Eugenol-Glutathione Conjugates. Synthetic eugenol-glutathione conjugates were prepared by reacting eugenol quinone methide with glutathione. Eugenol quinone methide was prepared according to Zanarotti (24). Silver(1) oxide (1.625 mmol) was added to 10 mL of carbon tetrachloride containing eugenol (0.375 mmol). The mixture was stirred vigorously at 65 "C for 10 min. The reaction was filtered and let cool. The quinone methide was reacted with glutathione (0.375mmol in 10 mL of 0.1 M phosphate buffer, pH 8.0) for 3 h at 37 "C with vigorous stirriig. The aqueous phase was washed with 10 mL of carbon tetrachloride. An aliquot was then washed on a PrepSep C18 extraction column (Fisher Scientific, Fair Lawn, NJ) and eluted with 1mL of methanol. This solution was analyzed by HPLC for eugenol-glutathione conjugates. In a separate incubation, 1 mM eugenol, 25 gg of myeloperoxidase, and 0.25 mM hydrogen peroxide were incubated in a volume of 10 mL of phosphate buffer (pH 7.0). After approximately 3 min (when the solution was yellow colored), the reaction mixture was extracted with 10 mL of carbon tetrachloride. The carbon tetrachloride extract was added to an equal volume of phosphate buffer (pH 8.0) containing 100 mg of glutathione and incubated at 37 "C with vigorous stirring for 1h. The reaction product was concentrated and analyzed the same way as the synthetic reaction products. PMN Isolation. PMNs were isolated from blood from healthy human volunteers by using a procedure described by Trush e t al. (11). Blood was centrifuged a t 150g for 10 min, and the plasma and buffy coat were discarded. The leukocyte- and erythrocyte-rich fraction was mixed with an equal volume of 6% dextran and incubated in inverted syringes for 45 min at 37 "C. The upper leukocyte-containing fraction was collected, pooled, and spun a t 150g for 5 min a t 4 "C. Contaminating erythrocytes were lysed by adding cold 0.155 M NH,Cl, 0.01 M KHC03, and 0.1 mM EDTA buffer (pH 7.4). PMNs were washed, resuspended in phosphate-buffered saline with 0.1% glucose, and counted on a hemocytometer. This procedure yielded a final preparation of cells that were >95% PMNs and that were 90-95% viable as determined by leakage of lactate dehydrogenase (25). Effects of Eugenol on Oxidative Burst. The effects of eugenol on the PMA-stimulated oxidative burst in PMNs were

Chem. Res. Toxicol., Vol. 2, No. 3, 1989 187 determined by measuring superoxide formation, hydrogen peroxide generation, and oxygen consumption. Superoxide formation was measured by following the reduction of acetylated cytochrome c using an Aminco DW-2 spectrophotometer operating in double-beam mode (mono 1,550 nm; mono 2,540nm). Incubations contained various concentrations of eugenol, 1 X lo6 PMNs, and 0.2mg of acetylated cytochrome c in a total volume of 1 mL of phosphate-buffered saline, pH 7.4. Incubations were carried out a t 37 "C. In some incubations 0.2 mg of superoxide dismutase was included as a control. The rate of superoxide formation was determined after addition of 100 nM PMA by using a molar extinction coefficient of 21 000. Hydrogen peroxide formation was measured by using the ferroammonium sulfate/potassium thiocyanate method as described by Hildebrandt et al. (26). Incubations contained various concentrations of eugenol, 2 X lo6 PMNs, and 1 mM azide in a total of 1 mL of phosphate-buffered saline, pH 7.4. Reactions were initiated with 100 nM PMA and incubated a t 37 "C for 10 min. The reactions were stopped by the addition of 1 mL of ice-cold 10% trichloroacetic acid and spun a t 3000 rpm to pellet protein. One-milliliter aliquots of the supernatant were used for hydrogen peroxide measurements. The effect of eugenol on oxygen consumption in PMNs was measured by using a Yellow Springs Model 53 oxygen monitoring system with a Clark-type oxygen electrode. Incubations contained various concentrations of eugenol and 5 X lo6 PMNs in a total volume of 3 mL of phosphate-buffered saline, pH 7.4, and were conducted at 37 "C. When a stable base line for oxygen consumption was achieved, 100 nM PMA was added to initiate the oxidative burst. Covalent Binding of [3H]Eugenol to PMN Protein. Reactions contained 100 gM eugenol (0.25 gCi), 1 X lo6 PMNs, 100 nM PMA (where indicated), and 10 mM azide (where indicated) in a total volume of 1 mL of 0.1 M phosphate-buffered saline, pH 7.4, and were incubated a t 37 "C for 30 min. Reactions were stopped by the addition of 4 mL of methanol and spun a t 3000 rpm to pellet the protein. The pellets were repeatedly washed with methanol to remove unbound eugenol (five to eight washes). The protein pellets were then dissolved in 1N NaOH with heating a t 60 "C for 1 h. The solutions were neutralized with HC1, and an aliquot (0.5 mL) was counted for radioactivity and another aliquot (50 gL) used for protein determination (27). Glutathione Depletion and Cytotoxicity in PMNs. Glutathione in PMNs was measured by using the thiol reagent monobromobimane as described by Cotgreave and Mold6us (28). Incubations contained 100 gM eugenol (where indicated), 100 nM PMA (where indicated), and 1 X lo6 PMNs in a total of 1 mL of 0.1 M phosphate-buffered saline (pH 7.4). Incubations were conducted a t 37 "C for various time periods up to 30 min. Cytotoxicity experiments were carried out by using various concentrations of eugenol in 30-min incubations identical with those described above. Cytotoxicity was assessed by using lactate dehydrogenase leakage as indicator (25).

Results Eugenol was oxidized by myeloperoxidase, isolated from

human PMNs, t o a yellow-colored metabolite with an absorbance m a x i m u m at 350 nm. The formation of t h i s metabolite occurred i n an enzyme-dependent manner (Figure 1A). The presence of reduced glutathione at the beginning of these incubations resulted in the inhibition of the formation of t h i s metabolite (Figure 1B). In addition, glutathione added after the reaction had started (Figure 1C) was capable of reacting directly with the metabolite. The presence of up t o 100 mM sodium chloride had n o effect o n the rate of oxidation of eugenol (not shown). The reactivity and absorption m a x i m u m of the eugenol metabolite suggest that i t is a quinone methide (18,29). W e attempted t o confirm the identity of t h i s metabolite by separation and isolation of the metabolite using HPLC and then comparing i t with synthetic quinone methide (24). However, we were unable to detect a n y product that

188 Chem. Res. Toxicol., Vol. 2, No. 3. 1989

Thompson et al.

A

C

Tim Imn) T m (min.) Time (mn.) Figure 1. Myeloperoxidase-dependent oxidation of eugenol. Oxidation was monitored by following the absorbance change at 350 nm, which is thought to represent the formation of eugenol quinone methide. Reactions contained 500 pM eugenol, 0.5 pg of peroxidase, and 250 pM hydrogen peroxide in 1mL of 0.1 M phosphate buffer, pH 7.0, unless indicated otherwise. (Panel A) Effect of peroxidase concentration on eugenol oxidation: (a) 0.1 pg of peroxidase, (b) 0.2 pg, and (c) 0.5 pg/mL. (Panel B) Effect of various concentrations of glutathione on eugenol oxidation. Numbers on the figure refer to micromolar glutathione present at start of reaction. (Panel C) Reaction of glutathione with eugenol oxidation product. Glutathione (50 pM) was added to the reaction at the arrow. '

I

'

'

'

+ I '

0.4

'

'

I

3

0.6

0)

V

0.3

0.2

0.1

c

-

0.4

0

e0 v)

1

.a

a

0

1 4

1

4

I

L 0

0.2

10

20

30

40

50

Time (min) Figure 2. Chromatogram of the products obtained from the reaction of glutathione with an extract from a eugenol/myeloperoxidase reaction. Reaction conditions are described under Materials and Methods. Peaks 1-3 are eugenol-glutathione conjugates while peak 4 represents parent eugenol. Absorbance was measured a t 280 nm. adsorbed at 350 n m using either normal-phase or reverse-phase columns, even when injecting t h e synthetic quinone methide. W e were successful in trapping t h e reactive metabolite with glutathione. A reaction containing eugenol, hydrogen peroxide, a n d myeloperoxidase was extracted with carbon tetrachloride a n d subsequently incubated with glutathione. T h e resulting a d d u c t s were analyzed by HPLC (Figure 2). Three peaks were obtained which were d e p e n d e n t o n t h e presence of glutathione (peaks 1-3). T h e same t h r e e peaks were also obtained from a reaction of synthetic quinone methide with glutathione (not shown). T h e UV spectrum of each peak from t h e enzymatic reaction was identical with its counterpart

250

300

350

400

450

500

Wavelength Figure 3. UV spectra of the major eugenol-glutathione conjugate (peak 3). (Trace A) Metabolite from synthetic reaction; (trace B) metabolite from the myeloperoxidase reaction. Details are described in the text and under Materials and Methods. Table I. Fate of Glutathione in Myeloperoxidase-Catalyzed Eugenol Oxidation" reaction GSH GSSG coniueates complete system 65 f 4 447 f 33 1.6 f 0.1 aIncubations contained 500 nmol of GSH at the start of reactions. Each reaction also contained 0.5 mM eugenol, 500 gM Hz02, and 0.2 pg/mL HRP in a total of 1 mL of 0.1 M phosphate buffer, pH 7. Values represent nmol f SE of product formed in 10-min incubations. Values for GSSG are expressed as GSH equivalents. In control incubations without eugenol, GSH loss was less than 10%.

from t h e synthetic reaction. Figure 3 compares t h e UV spectra of peak 3 from both t h e synthetic reaction (trace A) a n d t h e enzymatic reaction (trace B). T h e three peaks from t h e synthetic reaction have been identified as eugenol-glutathione conjugates by fast a t o m b o m b a r d m e n t mass spectrometry ([M - 13-, 468) a n d proton NMR a n d will be reported in a subsequent paper (30). T h e reactivity of t h e intermediate, absorption maximum, a n d a d d u c t

Chem. Res. Toxicol., Vol. 2, No. 3, 1989 189

Oxidation of Eugenol by PMNs

/

Contrd

Table 11. Effect of Eugenol on Hydrogen Peroxide Formation in PMA-StimulatedPMNs4 eugenol concn, % of eugenol concn, % of PM

0 (control) 100 250

control

PM

control

100 66 56

500 1000

35 f 4 21 f 3

*4

*2

nControl rate was 6.2 nmol of hydrogen peroxide formed/(lO min.106 cella). Incubations contained 2 X lo6 PMNs, 1 mM sodi. um azide, and various concentrations of eugenol in a total of 1 mL of 0.1 M phosphate buffer, pH 7. Reactions were initiated with 100 nM PMA and incubated at 37 "C for 10 min. Values represent mean f SE of triplicate determinations.

Figure 4. Inhibition of PMA-stimulated superoxide formation

in PMNs by eugenol. Various concentrations of eugenol (pM, as indicated on the figure) were preincubated with PMNs for 1 min prior to the addition of PMA. Control rates of superoxide formation varied from 0.8 to 1.2 nmol/(min.106cells). I

formation with glutathione are all consistent with the hypothesis that the reactive metabolite is a quinone methide. In myeloperoxidase-catalyzed oxidations of eugenol containing glutathione at the start of the reaction, a large percentage of glutathione was oxidized to glutathione disulfide (ca. 9070,see Table I). A very small, but detectable (using [3H]glutathione),amount of eugenol-glutathione conjugate(s) was also formed. These results suggest that glutathione reacts primarily with the eugenol phenoxy1 radical, reducing it back to the parent compound while glutathione subsequently dimerizes to form glutathione disulfide. The small amount of quinone methide like metabolite formed in these reactions may react directly with glutathione to form conjugates or polymerize to form other products. These results parallel those previously seen with horseradish peroxidase (18). We tested several other mammalian peroxidases for their ability to oxidize eugenol. Lactoperoxidase, eosinophil peroxidase, prostaglandin H synthase, and rat intestinal peroxidase were also capable of oxidizing eugenol (not shown). Therefore, eugenol oxidation is not a peroxidase-specific reaction. It should be noted that although the peroxidase component of prostglandin H synthase is capable of oxidizing eugenol with hydrogen peroxide as substrate, eugenol has also been shown to be a potent inhibitor of the cyclooxygenase component of the enzyme when arachidonic acid is used as substrate (31, 32). The effects of eugenol on various cellular functions of PMNs were also investigated. The stimulation of PMNs by phorbol ester initiates a burst in oxygen consumption characterized by activation of a membrane-bound NADPH oxidase and subsequent formation of superoxide and hydrogen peroxide (33). Hydrogen peroxide then serves as substrate for myeloperoxidase. Eugenol-dependent inhibition of superoxide formation in PMA-stimulated PMNs is shown in Figure 4. PMNs were preincubated for 1 min with various concentrations of eugenol before PMA was added to stimulate the oxidative burst. A concentration of 100 pM eugenol was found to inhibit the rate of superoxide formation by approximately 50%. Higher

10

20

30

TIME (min)

Figure 5. Loss of glutathione during incubation of PMA-stim-

ulated PMNs with eugenol. Levels of reduced glutathione in PMNs were measured at various time points in the presence or absence of 100 pM eugenol and 100 nM PMA. Control level of glutathione was 1.9 nmol/lOs cells. Key: (0)100 pM eugenol alone; ( 0 ) 100 nM PMA alone; (A) 100 nM PMA + 100 pM eugenol.

concentrations inhibited superoxide formation almost completely. In an effort to ensure that eugenol was not simply interfering with the superoxide measurement, rather than the oxidative burst, we measured the effect of eugenol on superoxide formation using the xanthine/ xanthine oxidase system. Eugenol inhibited superoxide formation in this system also, but it was found to be due to the direct inhibition of the xanthine oxidase rather than interference with the superoxide measurement because eugenol also inhibited oxygen consumption in this system (results not shown). A separate experiment showed that eugenol was not reactive toward potassium superoxide (not shown). Therefore, we concluded that eugenol was not interfering with the superoxide measurement. To further test this hypothesis, we measured the effect of eugenol on two separate parameters associated with the oxidative burst: hydrogen peroxide generation and oxygen consumption. Eugenol inhibited hydrogen peroxide formation at concentrations similar to those that inhibited superoxide formation in PMNs (Table 11). In oxygen consumption experiments, a concentration of 1 mM eugenol inhibited the rate of oxygen consumption [6.9nmol of oxygen consumed/(min.106cells) in control cells] by 53 f 5 % . Thus, by using three separate end points, we established that eugenol inhibited events associated with the oxidative burst of PMNs. It also appears that the effect of eugenol on the oxidative burst at least partially involves a direct inhibition of NADPH oxidase since oxygen consumption was inhibited at a high concentration of eugenol. The metabolism and toxic effects of eugenol on PMNs were also investigated. We observed that, in the presence of 100 pM eugenol, intracellular levels of glutathione in PMA-stimulated PMNs decreased rapidly to 10% of

190 Chem. Res. Toxicol., Vol. 2, No. 3, 1989 Table 111. Covalent Binding of [SH]Eugenol to PMN Protein" reaction nmol bound/mg of protein control (-PMA) 0.16 f 0.03 complete system 1.73 f 0.37 complete system + azide 1.02 f 0.23

"Reactions contained 100 pM eugenol (0.2 p C i ) , 100 nM PMA (except where indicated), 10 mM sodium azide (where indicated), and 1 X lo6 PMNs/mL in phosphate-buffered saline (pH 7.4) and were incubated at 37 "C for 30 min. Values represent mean f SE of triplicate replications.

OL6

200

460 660 8i0 lob0

EUGENOL (pM)

Figure 6. Cytotoxicity of eugenol to PMNs in the presence and absence of PMA. PMNs were incubated with various concentrations of eugenol for 30 min in either the presence ( 0 )or absence (A)of 100 nM PMA. Cytotoxicitywas measured by using lactate

dehydrogenase leakage.

control levels after incubation for 30 min (Figure 5). Incubations with eugenol alone did not significantly affect glutathione levels while incubations with PMA alone caused a decrease of approximately 30%. Using [3H]eugenol, we observed that PMA-stimulated PMNs catalyzed the covalent binding of eugenol to cellular protein (Table 111). This covalent binding was partially inhibitable by azide, indicating that at least part of the covalent binding was attributable to myeloperoxidase-catalyzed activation of eugenol. Finally, we measured the cytotoxicity of various concentrations of eugenol to PMNs in the presence and absence of phorbol ester (Figure 6). In the absence of PMA, no significant change in cytotoxicity occurred after 30-min incubation of PMNs with a eugenol concentration range of 0-1 mM. In PMA-stimulated PMNs, however, increasing concentrations of eugenol elicited an increase in cytotoxicity as reflected by leakage of lactate dehydrogenase into the incubation medium. A concentration of 1 mM eugenol caused 50% cell death in PMA-stimulated PMNs as opposed to 23% cell death in unstimulated cells.

Discussion Polymorphonuclear leukocyte dependent activation of a broad spectrum of chemicals has been described (3-11). Attempts have been made to link some of the clinical effects of these chemicals to their neutrophil-dependent metabolism to reactive intermediates. In this report we studied the enzymatic (myeloperoxidase) and cellular (PMN) dependent metabolism of eugenol to determine if potentially harmful reactive intermediates are produced. We observed that myeloperoxidase catalyzed the oxidation of eugenol to a reactive intermediate which is likely a quinone methide. This oxidation of eugenol was not unique to myeloperoxidase but also occurred with several

Thompson et al.

other mammalian peroxidase enzymes. We have previously observed that horseradish peroxidase catalyzed oxidation of eugenol proceeds through the formation of a phenoxy1 radical (18). Myeloperoxidase-dependent oxidation of eugenol occurs presumably via a similar mechanism, since both of these enzymes are known to catalyze one-electron oxidations of xenobiotics. Our present experiments with glutathione also suggest this type of mechanism. The myeloperoxidase-dependent oxidation of eugenol in the presence of glutathione led to the formation of glutathione disulfide and a small amount of eugenol-glutathione conjugateb) as well as inhibition of the formation of the quinone methide like metabolite. The interaction of reduced glutathione with the eugenol phenoxyl radical would account for the formation of glutathione disulfide (Table I) and the inhibition of the formation of subsequent oxidation products (Figure lb). Phenoxy1 radicals that did not interact with glutathione could then further react to form polymeric products or form the quinone methide like metabolite, which could subsequently form conjugates with glutathione (Table I). In incubations with PMNs, we found evidence which suggested that peroxidase-dependent oxidation of eugenol also occurred intracellularly. Intracellular glutathione decreased rapidly in PMA-stimulated cells exposed to eugenol, and covalent binding of [3H]eugenol to cellular protein was observed in stimulated cells. The intracellular formation of reactive metabolites of eugenol appears to occur through the ability of eugenol to serve as a reducing cofactor for the hydrogen peroxidelmyeloperoxidase reaction rather than a direct oxidation by superoxide as has been described for some phenols (3). We observed that potassium superoxide was unable, by itself, to oxidize eugenol to the quinone methide. Other oxidants may be involved, however, since azide did not fully inhibit the covalent binding of eugenol to PMN protein. The effects of eugenol on PMNs were dependent on concentration. At lower concentrations (ca. 100 pM) eugenol was metabolized to a covalently bound product and a metabolite that depeleted intracellular glutathione. No cytotoxic effects were seen at these concentrations although there was some inhibition of events associated with the oxidative burst (superoxide and hydrogen peroxide formation). At higher eugenol concentrations, greater inhibition of the oxidative burst was apparent and eugenol became cytotoxic to the PMNs themselves. Thompson et al. (18) demonstrated recently that horseradish peroxidase derived metabolites of eugenol are highly cytotoxic to isolated rat hepatocytes used as a target cell. Similarly, reactive intermediates from myeloperoxidase-catalyzed oxidation of eugenol may be responsible for the cellular toxicity observed in PMNs. However, since eugenol also inhibits the oxidative burst, high concentrations of this compound may actually inhibit the formation of hydrogen peroxide necessary to sustain the myeloperoxidasereaction. Thus, high concentrations of eugenol may inhibit its own metabolism, and therefore, we cannot rule out the possibility that some of the cytotoxic effects may be due to factors other than the formation of oxidized eugenol metabolites. At concentrations of eugenol around M, for example, Cotmore et al. (34) found inhibition of mitochondrial respiration and Hume (35)reported that eugenol inhibited cellular respiration in several cell types. Suzuki (17) found that very high concentrations of eugenol (ca. 5 mM) were cytotoxic to guinea pig MNS and affected cell membranes, actually leading to a stimulation of superoxide formation.

Chem. Res. Toxicol., Vol. 2, No. 3, 1989 191

Oxidation of Eugenol by PMNs

The intracellular formation of eugenol phenoxy1 radical and quinone methide like metabolites would lower levels of reduced glutathione and other protective thiols, including perhaps protein thiols as well. This could occur either through oxidation or conjugation. Modulation of cellular thiol status by incubation of PMNs with thioldepleting agents such as reactive aldehydes has been shown to inhibit the activity of the plasma membrane NADPH oxidase responsible for superoxide generation (36). Our results thus provide a molecular basis for a mechanism explaining the inhibitory effects of eugenol on PMNs reported in this paper and elsewhere (16). Similarly, these results suggest that inhalation of eugenol may result in the increased formation of cytotoxic metabolites of eugenol by peroxidase-containing phagocytes and/or the depressed function of these cells in their antibacterial and antiviral capabilities. This may result in increased susceptibility to infection as has been reported with exposure of several animal models to ozone and other oxidants (37). The toxic effects of eugenol may thus be involved in the reported association of clove cigarette mediated acute lung damage in humans and prodromal respiratory tract infections (14).

Acknowledgment. This project was supported by funds from Karolinska Institutet and the Swedish Medical Research Council. Registry No. Peroxidase, 9003-99-0; eugenol, 97-53-0; glutathione, 70-18-8; prostaglandin H synthetase, 59763-19-8; eugenol-glutathione conjugate, 120882-70-4.

References (1) O’Brien, P. J. (1988) Radical formation during the peroxidase catalyzed metabolism of carcinogens and xenobiotics: The reactivity of these radicals with GSH, DNA, and unsaturated lipid. Free Radical Biol. Med. 4, 169-183. (2) Klebanoff, S. J. (1980) Oxygen metabolism and the toxic properties of phagocytes. Ann. Intern. Med. 93, 480-489. (3) Eastmond, D. A., French, R. C., Ross, D., and Smith, M. T. (1987) Metabolic activation of 1-naphthol and phenol by a simple superoxide-generating system and human leukocytes. Chem.-Biol. Interact. 63, 47-62. (4) Sadler, A,, Subrahmanyam, V. V., and Ross, D. (1988) Oxidation of catechol by horseradish peroxidase and human leukocyte peroxidase: Reactions of o-benzoquinone and o-benzosemiquinone. Toxicol. Appl. Pharmacol. 93, 62-71. (5) Uetrecht, J., Zahid, N., and Rubin, R. (1988) Metabolism of procainamide to a hydroxylamine by human neutrophils and mononuclear leukocytes. Chem. Res. Toxicol. 1, 74-78. (6) Uetrecht, J., and Zahid, N. (1988) N-chlorination of phenytoin by myeloperoxidase to a reactive metabolite. Chem. Res. Toxicol. 1, 148-151. (7) Tsuruta, Y., Subrahmanyam, V. V., Marshall, W., and O’Brien, P. J. (1985) Peroxidase-mediated irreversible binding of arylamine carcinogens to DNA in intact polymorphonuclear leukocytes activated by a tumor promoter. Chem.-Bid. Interact. 53, 25-35. (8) Trush, M. A., Kensler, T. W., and Seed, J. L. (1986) Activation of xenobiotics by human polymorphonuclear leukocytes via reactive oxygen-dependent reactions. In Biological Reactive Intermediates (Kocsis, J. J., Jollow, D. J., Witmer, C. M., Nelson, J. O., and Synder, R., Eds.) Vol. 111, pp 311-321, Plenum Publishing Corp., New York. (9) Geuskins, M., Burglen, M. J., and Uriel, J. (1977) In vitro binding of 3H-estradiol to eosinophil and neutrophil granulocytes in various tissues (normal and neoplastic) of newborn and adult rats. Virchows Arch., B. 24, 67-78. (10) Trush, M. A. (1984) Activation of bleomycin A2 to a DNAdamaging intermediate by phorbol ester-stimulated human polymorphonuclear leukocytes. Toxicol. Lett. 20, 297-302. (11) Trush, M. A., Seed. J. L., and Kensler, T. W. (1985) Oxidantdependent metabolic activation of polycyclic aromatic hydrocarbons by phorbol ester-stimulated human polymorphonuclear leukocytes: Possible link between inflammation and cancer. Proc. Natl. Acad. Sci. U.S.A. 82, 5194-5198.

Kensler, T. W., Egner, P. A., Moore, K. G., Taffe, B. G., Twerdok, L. E. and Trush, M. A. (1987) Role of inflammatory cells in the metabolic activation of polycyclic aromatic hydrocarbons in mouse skin. Toxicol. Appl. Pharmacol. 90,337-346. 3) IARC Monographs (1985) Eugenol. In Eualuation of the Carcinogenic Risk of Chemicals to Humans, Vol. 36, pp 75-97, International Agency for Research on Cancer, Lyon France. 4) Morbidity and Mortality Weekly Report (1985) Illnesses possibly associated with smoking clove cigarettes. Vol. 34, pp 297-299. (15) LaVoie, E. J., Adams, J. D., Reinhardt, J., Rivenson, A., and Hoffman, D. (1986) Toxicity studies on clove cigarette smoke and constituents of clove: determination of the LDWof eugenol by intratracheal instillation in rats and hamsters. Arch. Toxicol. 59, 78-81. (16) Fotos, P. G., Woolverton, C. J., Van Dyke, K., and Powell, R. L. (1987) Effects of eugenol on polymorphonuclear cell migration and chemiluminescence. J. Dent. Res. 66, 774-777. (17) Suzuki, Y., Sugiyama, K., and Furuta, H. (1985) Eugenol-mediated superoxide generation and cytotoxicity in guinea pig neutrophils. Jpn. J . Pharmacol. 39, 381-386. (18) Thompson, D., Norbeck, K., Olsson, L., Constantin-Teodosiu, D., Van der Zee, J., and MoldBus, P. (1989) Peroxidase-catalyzed oxidation of eugenol: formation of a cytotoxic metabolite(& J . Biol. Chem. 264, 1016-1021. (19) Svensson, B. E., Domeij, K., Lindvall, S., and Rydell, G. (1987) Peroxidase and peroxidase-oxidase activities of isolated human myeoloperoxidases. Biochem. J. 242, 673-680. (20) Kimura, S., and Jellinck, P. H. (1982) Studies on mammalian intestinal peroxidase. Biochem. J. 205, 271-279. (21) Azzi, A., Montecucco, C., and Richter, C. (1975) The use of acetylated ferricytochrome C for the detection of superoxide radicals produced in biological membranes. Biochem. Biophys. Res. Commun. 65,597-603. (22) Ellman, G. L. (1959) Tissue sulfhydryl groups. Arch. Biochem. Biophys. 82, 70-77. (23) Sachetta, P., Di Cola, D., and Federici, G. (1986) Alkaline hydrolysis of N-ethylmaleimide allows a rapid assay of glutathione disulfide in biological samples. Anal. Biochem. 154, 205-208. (24) Zanarotti, A. (1985) Synthesis and reactivity of vinyl quinone methides. J. Org. Chem. 50, 941-945. (25) MoldBus, P., Hogberg, J., and Orrenius, S. (1978) Isolation of liver cells. Methods Enzymol. 52, 60-71. (26) Hildebrandt, A. G., Roots, I., Tjoe, M., and Heinemeyer, G. (1978) Hydrogen peroxide in hepatic microsomes. Methods Enzymol. 52, 342-350. (27) Peterson, G. L. (1977) A simplification of the protein assay method of Lowry which is more generally applicable. Anal. Biochem. 83, 346-356. (28) Cotgreave, I. A., and MoldBus, P. (1986) Methodologies for the application of monobromobimane to the simultaneous analysis of soluble and protein thiol components of biological systems. J. Biochem. Biophys. Methods 13, 231-249. (29) Pew, J. C., Connors, W. J., and Kunishi, A. (1965) Enzymatic dehydrogenation of lignin model phenols. In Chimie et Biochemie de la Lignine de la Cellulose et des Hemicelluloses, Actes du Symposium International de Grenoble, France, Julliet 1964, pp 229-245, Impimeries Reunies de Chambery, Grenoble. (30) Thompson, D., Constantin-Teodosiu, Egestad, B., Mickos, H., and MoldBus, P. (1989) Formation of glutathione conjugates during oxidation of eugenol by microsomal fractions of rat liver and lung. Biochem. Pharmacol. (submitted for publication). (31) Dewhirst, F. (1980) Structure-activity relationships for inhibition of prostaglandin cyclooxygenase by phenolic compounds. Prostaglandins 20, 209-222. (32) Rasheed, A., Laekeman, G . M., Vlietinck, A. J., Janssens, J., Hatfield, G., Totte, J., and Herman, A. G. (1984) Pharmacological influence of nutmeg and nutmeg constituents on rabbit platelet function. Planta Med. 50, 222-226. (33) Rossi, F. (1986) The 02--forming NADPH oxidase of the phagocytes: nature, mechanisms of action and function. Biochim. Biophys. Acta 853, 65-89. (34) Cotmore, J. M., Burke, A., Lee, N. H., and Shapiro, I. M. (1979) Respiratory inhibition of isolated rat liver mitochondria by eugenol. Arch. Oral Biol. 24, 565-568. 2)

192 Chem. Res. Toxicol., Vol. 2, No. 3, 1989

Thompson et al.

(35) Hume, W. R. (1984) Effect of eugenol on respiration and di-

and pulmonary alveolar macrophages. Biochem. Pharmacol. 36,

vision in human pulp, mouse fibroblasts, and liver cells in vitro. J. Dent. Res. 63,1262-1265. (36) Witz, G., Lawrie, N. J., Amoruso, M. A,, and Goldstein, B. D. (1987) Inhibition by reactive aldehydes of superoxide anion radical production from stimulated polymorphonuclear leukocytes

721-726. (37) Gardner, D. E. (1982) Effects of gases and airborne particles on lung infections. In Air Pollution-Physiological Effects (McGrath, J. J., and Barnes, E. B., Eds.) pp 47-79, Academic

Press, New York.