Eugenol Causes Oxidant-mediated Edema in Isolated Perfused Rabbit Lungs 1- 4

JOHN W. MCDONALD and JOHN E. HEFFNER

Introduction

Eugenol (4-allyl-2-methoxyphenol) is a phenolic compound that constitutes the major component of oil of cloves and exists to a lesser extent in extracts of other plants, such as cinnamon, basil, and nutmeg. Long used as a flavoring and fragrance agent in food products, the excellent antiseptic and analgesic properties of eugenol have provided medicinal application in dental and surgical pastes. Interest in the pulmonary effects of eugenol derives from anecdotal reports of acute pulmonary edema and hemoptysis in previously healthy young persons temporally related to smoking clove cigarettes (kreteks) (1-3). Imported into the United States since 1968,clovecigarettes contain tobacco and 35 to 400/0 clovematerial by weight added to impart flavor (4). Sales have increased to 150 million units in fiscal 1984, with the majority sold to young persons 17 to 30 years old (3). Combustion of each cigarette produces up to 15mg eugenol and eugenol pyrolysis products that are not present in pure tobacco cigarettes (5). Eugenol has demonstrated cytotoxic effects on pulmonary (6, 7) and nonpulmonary tissues (8). Infused intravenously in dogs (7) and instilled intratracheally in rats (6),eugenol causes acute hemorrhagic pulmonary edema. The mechanisms of eugenol-induced pulmonary edema, however, are unclear. Toxicity may result either directly from the effects of eugenol or eugenol metabolites on cell function or indirectly through the demonstrated capacity of eugenol to stimulate neutrophils to release toxic oxygen metabolites (9). The generation of phenoxyl radicals from eugenol in the presence of horseradish peroxidase (10) suggested to us the hypothesis that eugenol directly causes lung injury through oxidant-mediated mechanisms. To test this hypothesis we administered eugenol to isolated rabbit lungs perfused with a cell-free, albumin and salt solution ventilated with 95% oxygen and 5% carbon dioxide and observed the formation of 806

-

SUMMARY Eugenol, an extract of cloves, has been associated with pulmonary edema when inhaled from commercially available clove cigarettes. Wetested the hypothesis that eugenol directly causes lung edema through oxidant-mediated mechanisms by Infusing eugenol (0.1 and 1.0 mM) Into isolated rabbit lungs perfused with a cell-free albumin and physiologic salt solution. We observed lung edema (1.0 mM) as demonstrated by Increased lung weight gain and wet-to-dry lung weight ratios without alterations In mean pulmonary artery pressure. The oxygen metabolite scavengers catalase (1,000U/ml) and dlmethylthlourea (30 mM) attenuated lung edema. Instillation of dimethylurea, superoxlde dlsmutase, or heat-inactivated catalase did not prevent lung edema formation. Weconclude that eugenol causes lung edema in isolated lungs through oxidant-mediated mechanisms In the absence of circulating formed blood elements. Eugenol may be a valuable compound in the laboratory Investigation of edemogenlc disorders. AM REV RESPIR DIS 1991; 143:806-809

lung edema. Oxygen radical scavengers were coinfused in other experiments to evaluate the role of toxic oxygen metabolites on measured variables. These investigations may provide insight into the potential clinical toxicity of eugenol inhalation from clove cigarettes. More importantly, because eugenol can be administered to animals through the oral, intratracheal, and intravenous routes, demonstration that it causes lung edema through oxidantmediated mechanisms may provide an important new model for the study of oxidant-induced lung injury in whole animals. Methods Isolated Lung Preparation New Zealand white rabbits (2 to 2.5 kg) were used in these studies. Animals were anesthetized in their cages with intramuscular xylazine (10mg/kg; Cutter Laboratory, Shawnee, KS) before transport to the laboratory, where ketamine (25 to 50 mg/kg; Parke-Davis, Detroit, MI) wasinfusedintravenouslythrough an ear vein. A tracheotomy was performed through which a small-animal ventilator administered 95070 O 2 and 5% CO 2 at a respiratory rate of 24 breaths/min, tidal volume of 15 mIlkg, and positive end-expiratory pressure of 2 em H 2 0 . After a midsternotomy heparin (500 U) was injected into the right ventricle and the pulmonary artery and left ventricle were cannulated with plastic catheters. Formed blood elements were flushed with 500 ml Krebs-Henseleit (K-H) physiologic salt solution containing 3% bovine se-

rum albumin (BSA)(Sigma Chemical Co., St. Louis, MO). The heart-lung preparation was suspended in a humidification dome (38 0 C) from a force-displacement transducer (Grass Instruments, Quincy, MA) for continuous monitoring of changes in lung weight. Lung perfusion was then initiated with K-H solution and 3% BSA in a 300-ml closed, recirculating reservoir system at a constant flow rate of 100 mllmin by means of a peristaltic pump (model 1203;Harvard Apparatus Co., South Natick, MA). Mean pulmonary artery pressure (Pa) was monitored with a pressure transducer (type 4-327-0010; Bell and Howell,Pasadena, CA) and recorded continuously on a polygraph (model 7; Grass Instruments). At the end of the experiments lung preparations were removed, dissected free of mediastinal structures, weighed, and placed in an oven (60 0 C) for 72 h. Desiccated lungs

(Received in original form June 5, 1990 and in revised form September 24, 1990) 1 From the Division of Pulmonary and Critical Care Medicine, Medical Universityof South Carolina, Charleston, South Carolina, and the Department of Internal Medicine, St. Joseph's Hospital and Medical Center, Phoenix, Arizona. 2 Supported in part by Grant No. HL-oI992 from the National Heart, Lung, and Blood Institute. 3 Presented in part at the Annual Meeting of the American Thoracic Society, Boston, Massachusetts, May 22, 1990and published in abstract form [Am Rev Respir Dis 1990; (Suppl 139:A532»). 4 Correspondence and requests for reprints should be addressed to John E. Heffner, M.D., Director of Medical Critical Care, Department of Internal Medicine, 350 West Thomas Road, St. Joseph's Hospital and Medical Center, Phoenix, AZ 85001.

807

EUGENOL-INDUCED PULMONARY EDEMA

werethen reweighedfor determination of wetto-dry lung weight ratios.

Eugenol Experimental Protocols After initiation of lung perfusion in the recirculating reservoir system the lungs were observed for 10 min to assure that Pa and lung weights werestable. Volumes of eugenol (Sigma) of 5 or 50 III were then instilled into the reservoir to produce perfusate concentrations of 0.1 or 1.0 mM. The lungs were then observed for an additional 45 min (55 min of total lung perfusion) for changes in lung weight and Pa. Control studies were observed for 55 min without the instillation of eugenol. Eugenol-antioxidant Experimental Protocols In other studies the following antioxidants were infused into the reservoir after 7 min of lung perfusion followed by eugenol (1.0mM) at 10min to evaluate the role of oxidant generation in eugenol-induced lung edema formation: 1,000Vlml of catalase (Worthington Diagnostic Systems, Freehold, NJ), 30 mM dimethylthiourea (DMTV) (Alpha Products, Danvers, MA), and 233 Vlml of superoxide dismutase (SOD) (Sigma). Dimethylurea (DMU) (Sigma) served as a control for the osmotic effect of DMTV and was added in a concentration of 30 mM. To control for a nonspecific protein effect of catalase, heatinactivated catalase (prepared by autoclaving stock solutions of catalase at 120° C for 1 h) was instilled into the reservoir at a concentration of 1,000 Vlml. Statistics Data are presented as mean ± standard error of the mean (SEM). A microcomputer (Macintosh" SE; Apple Computer, Cupertino, CA) and a statistical package (Statview'" 512+; Brainpower, Calabasas, CA) performed the statistical evaluations. One-way analysis of variance with Fisher's protected least significant difference test for multiple comparisons was used to determine differences between groups. P values less than 0.05 were considered significant.

TABLE 1 EFFECTS OF EXPERIMENTAL CONDITIONS ON ISOLATED PERFUSED RABBIT LUNGS· Experimental Condition

n

Control EUG, 0.1 EUG, 1.0 EUG, 1.0 EUG, 1.0 EUG, 1.0 EUG, 1.0 EUG, 1.0

6 6 6 6 3 6 6 6

mM mM mM mM mM mM mM

+ + + + +

CAT HI/CAT DMTU DMU SOD

Pa (mmHg)

8.9 8.7 9.7 8.2 8.1 7.2 8.8 8.5

± ± ± ± ± ± ± ±

-0.3 5.3 33.8 8.6 35.1 5.1 31.5 36.8

± ± ± ± ± ± ± ±

0.2 4.1t 4.6l 2.5t 2.8l 1At 5.7l 2.7l

Wet-to-Dry Weight Ratio

7.7 7.0 13.8 8.2 14.0 5.8 13.2 12.5

± ± ± ± ± ± ± ±

0.2 0.6t 1.8l 1.2t 2.1l OAt 1.8l 1.5l

Definition of abbreviations: Pa = mean pulmonary arterial pressure at 55 min; lung weight gain = increase in lung weight at 55 min; EUG = eugenol; CAT = catalase; HI/CAT = heat-inactivated catalase; DMTU = dimethylthiourea; DMU = dimethylurea; SOD = superoxide dismutase. • Values are mean ± SEM; n = number of lungs. t p < 0.05 compared with respective values in 1.0 mM eugenol experiments. :j: p < 0.05 compared with respective values in control experiments.

figure 1). There was no change in Pa (8.7 ± 0.8 mm Hg) from baseline values (table 1) or differences in wet-to-dry lung weight ratios (7.0 ± 0.6) compared to control studies (figure 2). Instillation of 1.0mM eugenol (n = 6) caused an immediate and progressive lung weight gain (33.8 ± 4.6 g) that was significantly increased compared to controllungs (figure 1;p < 0.05). These lungs maintained stable baseline Pa (9.7 ± 1.1 mm Hg) that were similar to control values (table 1). Wet-to-dry lung weight ratios were increased at the end of the experiments (13.8 ± 1.8)compared with control (p < 0.05; figure 2).

Antioxidant Effects on Eugenol-induced Lung Edema To determine the role of toxic oxygen metabolites in the generation of eugenolinduced lung edema formation, oxygen radical scavengers wereinfused 3 min before the instillation of eugenol at a concentration of 1.0 mM into the perfusate 40

Results

Effects of Eugenol on Pa and Lung Edema Formation Control lungs perfused with 3070 BSA and Krebs-Henseleit solution alone (n = 6) maintained stable Pa (table 1) and lung weights (figure 1)throughout the 55 min of perfusion. Lung wet-to-dry weight ratios were 7.7 ± 0.2 at the end of the experiments (figure 2). Infusion of eugenol into isolated lungs caused a dose-dependent weight gain during the 55 min of lung perfusion. Mean weight gain from 0.1 mM concentration of eugenol (n = 6) was 5.33 ± 4.14 g, which was not significantly increased compared to control (P < 0.05;

0.9 0.8 1.1 004 0.5 004 0.9 0.6

Lung Weight Gain (g)

30

§ z

20

:ci:

CJ

~

CJ 10 Z

:3

-10

+---.---,----.----.---...---,-..-----,---.----.----.-----, o 10 20 30 40 50 60 TIME

(min)

Fig. 1. Effect of 0.1 mM eugenol (large closed circles, n = 6) and 1.0 mM eugenol (open circles, n = 6) on lung weight in isolated perfused rabbit lungs. Parallel time controls (small closed circles, n = 6) are also included. Asterisk denotes p < 0.05 compared with control values.

reservoir. Infusion of catalase (1,000 U/ml perfusate) before eugenol (n = 6) attenuated but did not entirely prevent lung weight gain (8.6 ± 2.5 g; figure 3). Wet-to-dry lung weight ratios in experiments infused with catalase and eugenol (8.2 ± 1.2) were significantly less than those observed in experiments infused with eugenol alone (p < 0.05) and similar to control values (p = NS; figure 2). There were no changes in Pa throughout experiments infused with catalase and eugenol (table 1). Heat-inactivated catalase did not prevent lung weight gain (35.1 ± 2.8 g; figure 3) or the observed increase in wet-to-dry lung weight ratio (14.0 ± 2.1; figure 2). Pa remained stable throughout the experiments (8.1 ± 0.5 mm Hg; table 1). The infusion of DMTU (30 mM) before eugenol (1.0 mM; n = 6) similarly attenuated lung weight gain (5.1 ± 1.4 g) to values less than those observed in experiments infused with eugenol alone but greater than control values (figure 4). Wet-to-dry lung weight ratios (5.8 ± 0.4) were similar to those in control studies (figure 2; p = NS). There were no changes in Pa from baseline values (7.2 ± 0.4 mm Hg; table 1). In other experiments (n = 6) DMU (30 mM) was infused before eugenol to control for the osmotic protective effects of DMTU. DMU did not prevent lung weight gain (31.5 ± 5.7 g; figure 4) or the observed increase in wet-to-dry lung weight ratio (13.2 ± 1.8; figure 2) observed with the infusion of eugenol (1.0 mM). Pa remained stable throughout the experiments (8.8 ± 0.9 mm Hg; table 1). Infusion of SOD (233 U/ml perfusate; n = 6) had no significant effect on attenuating eugenol-induced lung weight gain (36.8 ± 2.7 g; figure 5) or the wetto-dry lung weight ratio at the end of the experiments (12.5 ± 1.5; figure 2). Pa re-

MCDONALD AND HEFFNER

808

*

*

14

o

*

~12

II:

~10 CJ

Z 8

::;) ...J

t

>- 6

II:

C

~

4

W ~

2

I-

CONTROL

EUG

EUG

EUG

EUG

EUG

CAT

HI/CAT

olhu

mained stable throughout the 55-min period of lung perfusion (8.5 ± 0.6 mm Hg; table 1). Discussion

In the isolated perfused rabbit lung (IPL) model we have shown that 1.0 mM eugenol causes immediate toxicity characterized by increased lung edema formation. The rapid time course indicates that the pulmonary edema occurs directly from eugenol or a rapidly formed metabolite. The absence of circulating cellular elements in the perfusate demonstrates that lung injury occurs independent of eugenol effects on neutrophil function. Attenuation of lung weight gain by the coinfusion of catalase or DMTU suggests that lung edema is mediated through oxidant-dependent mechanisms. The absence of a pressor response after infusion of eugenol is interesting considering that other compounds that cause oxidant-mediated lung edema in IPL,

*

Fig. 2. Effects of 0.1 mM (n = 6) and 1.0 mM (n = 6) eugenol (EUG) on wetto-dry lung weight ratios after 55 min of isolated lung perfusion. Addition of eugenol with catalase (EUG + CAT, n = 6 or dimethylthiourea (EUG + DMTU, n = 6) maintained wet-to-dry lung weight ratios similar to control values. Addition of eugenol with superoxide dismutase (EUG + SOD, n = 6), dimethylurea (EUG + DMU, n = 6), or heatinactivated catalase (EUG + HI/CAT, n = 3) caused similar increases in wetto-dry lung weight ratios compared to the infusion of 1.0 mM eugenol. Asterisk denotes p