Pharmaceutical Agents Known to Produce Disulfiram-Like Reaction: Effects on Hepatic Ethanol Metabolism and Brain Monoamines

Abstract

Several pharmaceutical agents produce ethanol intolerance, which is often depicted as disulfiram-like reaction. As in the case with disulfiram, the underlying mechanism is believed to be the accumulation of acetaldehyde in the blood, due to inhibition of the hepatic aldehyde dehydrogenases. In the present study, chloramphenicol, furazolidone, metronidazole, and quinacrine, which are reported to produce a disulfiram-like reaction, as well as disulfiram, were administered to Wistar rats and the hepatic activities of alcohol and aldehyde dehydrogenases (1A1 and 2) were determined. The expression of aldehyde dehydrogenase 2 was further assessed by Western blot analysis, while the levels of brain monoamines were also analyzed. Finally, blood acetaldehyde was evaluated after ethanol administration in rats pretreated with disulfiram, chloramphenicol, or quinacrine. The activity of aldehyde dehydrogenase 2 was inhibited by disulfiram, chloramphenicol, and furazolidone, but not by metronidazole or quinacrine. In addition, although well known for metronidazole, quinacrine also did not increase blood acetaldehyde after ethanol administration. The protein expression of aldehyde dehydrogenase 2 was not affected at all. Interestingly, all substances used, except disulfiram, increased the levels of brain serotonin. According to our findings, metronidazole and quinacrine do not produce a typical disulfiram-like reaction, because they do not inhibit hepatic aldehyde dehydrogenase nor increase blood acetaldehyde. Moreover, all tested agents share the common property to enhance brain serotonin, whereas a respective effect of ethanol is well established. Therefore, the ethanol intolerance produced by these agents, either aldehyde dehydrogenase is inhibited or not, could be the result of a “toxic serotonin syndrome,” as in the case of the concomitant use of serotonin-active medications.

The disulfiram-ethanol reaction (DER) refers to the unpleasant symptoms experienced upon drinking alcoholic beverages after having previously ingested disulfiram. After Hald and Jacobsen (Hald and Jacobsen 1948a, 1948b) investigated the DER in 1948. disulfiram was introduced as an adjunct in the treatment of chronic alcoholism (Martensen-Larsen 1948). Blurred vision, nausea, vertigo, anxiety, and cardiovascular effects such as hypotension, palpitations, tachycardia, and flushing of the face and neck (Kitson 1977) are some of the symptoms observed in the course of a DER, whereas some isolated cases of fatal reactions have been reported (Cahill 1972). Classically, the DER symptoms have been attributed to the inhibition of the hepatic aldehyde dehydrogenase (ALDH) (Deitrich and Hellerman 1963; Kitson 1975) and the subsequent accumulation of toxic amounts of acetaldehyde in the blood (Hald and Jacobsen 1948b; Larsen 1948).

Acetaldehyde is normally produced as a result of the initial oxidation of ethanol mainly by the alcohol dehydrogenase (ADH) of the liver, although it does not accumulate in the tissues, because it is further oxidized, almost as soon as it is formed, by the hepatic ALDH. In mammals, liver ALDH exists as several isozymes that differ in their subcellular location, electrophoretic mobility, isoelectric point, kinetic properties, molecular size, and substrate specificity. So far, the NAD-dependent ALDH systems, located in mitochondria and cytosol, are the principal metabolic pathways in which acetaldehyde is converted to acetate (Eriksson, Marselos, and Koivula 1975; Marjanen 1973; Tottmar, Peterson, and Kiessling 1973). Weiner (Weiner 1987) has shown by incubating rat liver slices with acetaldehyde, that a low- K _M mitochondrial ALDH (ALDH2) might be responsible for 60% of the metabolism of acetaldehyde, whereas high-K _M cytosolic ALDH (ALDH1A1) metabolized an additional 20%, and the remaining 20% was caused by an undetermined system, possibly the cytochrome P-450 (CYP) 2E1, as it was proposed later by Terelius et al. (Terelius et al. 1991).

Various therapeutic agents have been found to produce untoward effects when combined with ethanol, which often are reported as disulfiram-like reaction. Among other drugs, chloramphenicol, metronidazole, furazolidone, and quinacrine have been demonstrated to bring about such a reaction (Edwards, Fink, and Van-Dyke 1986; Fried 1980; Uri and Parks 1983). In accordance with the DER, disulfiram-like reaction has been also traditionally attributed to the decrease of acetaldehyde elimination from the body, due to inhibition of the hepatic ALDH. However, it is well established, that at least metronidazole, unlike disulfiram, neither inhibits ALDH of the liver nor increases blood acetaldehyde (Tillonen et al. 2000; Vasiliou, Malamas, and Marselos 1986; Visapaa et al. 2002). Some other mechanisms have been proposed to explain the toxic response of this drug to ethyl alcohol, including inhibition of the hepatic ADH (Gupta, Woodley, and Fried 1970), increased intracolonic acetaldehyde due to the replacement of intestinal anaerobes by ADH-containing aerobes (Tillonen et al. 2000), or a central action predisposing to increased sedation by ethanol (Vasiliou. Malamas. and Marselos 1986).

In order to investigate the effects of the above-mentioned pharmaceutical agents on the hepatic ethanol metabolism, as well as on the brain monoaminergic systems, disulfiram, chloramphenicol, furazolidone. metronidazole, and quinacrine were administered to Wistar rats and the hepatic enzyme activities of ADH and ALDHs (ALDH1A1 and ALDH2) were measured. The expression of ALDH2, which plays a key role in acetaldehyde metabolism, was further assessed by Western blot analysis. Moreover, in four brain subregions (hypothalamus, striatum, midbrain, and frontal cortex), the levels of brain biogenic monoamines and their metabolites were determined. Finally, after coadministration of ethanol with disulfiram, chloramphenicol, or quinacrine, the concentration of acetaldehyde was evaluated in whole blood samples.

MATERIALS AND METHODS

Animals

Male albino rats (Wistar/Af/Han/Mol/Io/RR), 4 months old and weighing 300 to 350 g, were used in this study. All animals were housed in groups of two to three in plastic cages (Macrolon) with a wood-chip bedding (Populus sp.) and had free access to tap water and pellet chow (Biozoe, Greece). The animals were maintained on a 12h/12h light/dark cycle with lights on at 07:00 h and a constant temperature at 22 ± 2°C.

Experiments on animals were handled with human care in accordance with the National Institutes of Health guidelines and the European Union directive for the care and the use of laboratory animals (Greek presidential decree No. 160 1991).

Drugs

Disulfiram (Sigma Chemicals, St. Louis, MO) and quinacrine (Sigma Chemicals) were dissolved in olive oil and 0.9% NaCl (saline), respectively. Chloramphenicol (Sigma Chemicals) and metronidazole (Caps. Flagyl, 500 mg; Rhone-Poulenc Rorer AEBE) were suspended in saline, whereas furazolidone (Sigma Chemicals) was suspended in olive oil. Ethanol (99.99%; Riedel de Haen, Germany) was administered as a 20% (v/v) solution in saline. All drugs and ethanol were administered in a volume of 1 ml/100 g body weight.

Treatment Protocol

Thirty-six animals were divided into six groups of six. Five groups of animals were treated for 4 consecutive days with disulfiram (75 mg/kg body weight), chloramphenicol (200 mg/kg body weight), furazolidone (100 mg/kg body weight), metronidazole (200 mg/kg body weight), or quinacrine (50 mg/kg body weight), respectively, with an additional injection on the 5th day, about 3 h before killing. The last group of animals served as control and received only olive oil (1 ml/100 g body weight) following the above mentioned treatment protocol.

In a parallel series of experiments, 24 animals were divided into four groups of six. Three groups received disulfiram (75 mg/kg body weight), chloramphenicol (200 mg/kg body weight), or quinacrine (50 mg/kg body weight), respectively for 4 consecutive days, whereas the remaining group was used as respective control and received only olive oil (1 ml/100 g body weight). On the 5th day, 1 h after the last injection of the drug or vehicle and 2 h before sacrifice, all animals received intragastrically 2 g/kg body weight ethanol.

All drugs were administered intraperitoneally in a single dose each day. The doses of the drugs used in the present study-are pharmacological, which means that they are (per kg body weight) about 10 times higher than the doses given to humans in therapeutics, which is a common practice in experiments in pharmacology and toxicology.

Tissue Preparation

Rats were sacrificed by decapitation and the hypothalamus, striatum, midbrain, and frontal cortex were microdissected immediately after killing, according to a conventional and well established technique (Cabrera-Vera et al. 2000). Samples were kept at −80°C until neurotransmitter analyses. The frozen tissues were thawed, weighed, and homogenized for 20 s with a sonicator in ice-cold 0.2 N perchloric acid. The homogenate was centrifuged at 10,000 × g or 15 min at 4°C, and then the supernatant was divided into two portions. The first aliquot was used for high-performance liquid chromatographic analyses (HPLC) of serotonin (5-HT), 5-hydroxyindole-3-acetic acid (5-HIAA), and homovanillic acid (HVA), whereas the second 0.2-ml aliquot was transferred to an Eppendorf tube containing 20 mg activated alumina in order to extract noradrenaline (NA), dopamine (DA), and 3,4-dihydroxyphenylacetic acid (DOPAC) from the homogenate prior to HPLC detection.

The liver of each animal was also dissected and a small amount (3 g) was homogenized in a mechanical homogenizer, in 3 volumes (w/v) ice-cold 0.25 M sucrose solution. The homogenate was then centrifuged for 10 min at 700 × g. The pellet, which represented the nuclear fraction, was discarded and the mitochondria were isolated by centrifugation at 10,000 × g for 15 min. The pellet was resuspended in 3 ml sucrose medium and washed at 10,000 × g for 15 min. The final pellet was resuspended in 1 ml sucrose medium containing 1% sodium deoxycholate and designated the mitochondrial fraction. The supernatant fraction of the first 10,000 × g centrifugation was subjected to a new 60-min centrifugation at 105,000 × g to obtain the cytoplasmic fraction. All steps were carried out from 0°C to 4°C and both fractions were stored in small aliquots at −80°C.

In the experiments with ethanol, heart blood samples were collected via cardiac puncture immediately after decapitation and 500 μl were added in 2 ml of ice-cold 0.6 M perchloric acid prepared in saline. After centrifugation at 4000 × g for 10 min at 4°C, supernatants (500/μl) were collected and acetaldehyde levels were determined within 3 h by head-space gas chromatography (hs-GC). No corrections have been made for the nonenzymatic formation of acetaldehyde from ethanol during the analytical procedures, because the levels of artifactual acetaldehyde were under the limit of detection.

Determination of Brain Biogenic Monoamines

Ion-pair reverse-phase chromatography was performed with an HPLC system (LC-9A: Shimadzu. Japan) equipped with an analytical column of 250 mm × 4.6 mm (Jones-Apex. ODS. C-18), 5-μm particle size, coupled to an electrochemical detector (L-ECD-6A: Shimadzu. Japan) maintained at 0.75 V. The mobile phase was a mixture of 0.1 M citric acid. 0.1 M sodium acetate, and 0.27 mM octyl sulfate with 25% methanol (v/v). The separations were performed isocratically at a flow rate of 0.6 ml/min (Mefford 1981; Mefford, Gilberg, and Barchas 1980). Samples were injected manually (20 μl) and the compounds under investigation were identified by comparison with the retention times of the authentic standards. The order of elution of the standard solutions was NA (5.6 min), DA (7.1 min), DOPAC (8.4 min), 5-HIAA (10.5 min), 5-HT (12.1 min), and HVA (14.2 min).

Enzyme Assays

All the assays of dehydrogenase enzymes were performed spectrophotometrically by following the formation of NADH at 340 nm in a Shimadzu UV1601 spectrophotometer. ADH as well as ALDH2 was measured at 25°C, whereas the cytoplasmic isozyme of ALDH (ALDH1A1) at 37°C.

ADH was assayed according to Koivula et al. (Koivula, Koivusalo, and Lindros 1975) in a mixture containing 70 mM NaOH-glycine buffer (pH 9.6), 0.67 mM NAD, and 10 mM ethanol, whereas for the determination of ALDH1 Al, the assay mixture contained 75 mM sodium pyrophosphate buffer (pH 8.0), 1 mM pyrazole (for the inhibition of alcohol dehydrogenase), 1 mM NAD, and 5 mM propionaldehyde as substrate (Marselos, Strom, and Michalopoulos 1986).

The mitochondrial enzyme ALDH2 was assayed by using 50/μM acetaldehyde as substrate. The assay mixture contained 75 mM sodium pyrophosphate buffer (pH 8.0), 1 mM NAD, 1 mM pyrazole (for the inhibition of alcohol dehydrogenase), and 2 μM rotenone (for the inhibition of NAD oxidase) (Vasiliou, Malamas, and Marselos 1986).

Protein determination was carried out by the method of Lowry and coworkers (Lowry et al. 1951).

Immunoblotting

Lysates (10/μg each lane) were mixed with sample buffer, boiled for 5 min, and subsequently run on 10% sodium dedecyl sulfate–polyacrylamide gel electrophoresis (SDS-PAGE) gels. Separated proteins were transferred to nitrocellulose (Hybond-C Extra; Amersham International) overnight. The nitrocellulose membranes were blocked with 3% nonfat dry milk in Trisbuffered saline (TBS) (10mM Tris-HCl, pH 7.5, 150mM NaCl), 0.05% Tween-20 overnight. Subsequently, the membranes were incubated with the appropriate dilution of primary antibodies, 1:10,000 for rabbit anti-ALDH2 (generously provided by Dr. Henry Weiner, Department of Biochemistry, Purdue University) and 1:20.000 for mouse anti-β-actin (A5441: Sigma) in 1% nonfat dry milk in TBS-0.05% Tween-20 for 2 h. The blots were washed for 15 min with TBS-0.05% Tween-20, and then incubated for 45 min with secondary antibody conjugated to horseradish peroxidase goat anti-mouse (1:6000) (Pierce, Rock-ford, IL) or goat anti-rabbit (1:7000) (Santa Cruz Biotechnology, Santa Cruz, CA). Proteins were detected by chemiluminescence using the Supersignal (Pierce).

Determination of Blood Acetaldehyde

Quantitation of acetaldehyde was performed by hs-GC using acetonitrile as internal standard, on a Shimadzu GC-17A gas chromatograph, equipped with a SUPELCOWAX fused silica capillary column (30 m × 0.25 mm, film thickness 0.25 μm), isothermally held at 60°C, and a flame ionization detector (260°C). Sample pretreatment (50°C for 3.5 min) and injection of a sample volume of 500 μl (injector at 115°C) was performed automatically by an AOC-5000 hs-GC injection system. The carrier gas was helium at a flow rate of 0.7 ml/min (Sarkola et al. 2002).

Statistical Analysis

Data are expressed as the group mean ± SD (n = 6). Statistical analysis was performed with Student’s t test. The limit for statistical significance was at p values less than .05.

RESULTS

Enzyme Activities

Disulfiram, chloramphenicol, and furazolidone inhibited ALDH2 activity (p <.001), which on the contrary was not significantly affected by the administration of metronidazole or quinacrine (Figure 1a ).

ALDH1A1 activity was also inhibited by disulfiram and furazolidone (p <.001), whereas the application of the other drugs of our study had no significant effect on this cytoplasmic isozyme of ALDH (Figure 1b ).

Finally, ADH activity was not significantly influenced by any of the drugs tested, with the exception of metronidazole, which produced a slight inhibition (p <.025) of this enzyme (Figure 1c ).

Western Blot Analyses (ALDH2)

As shown in Figure 2 and despite any differences observed in the activity of ALDH2 by the use of the tested drugs, the expression of ALDH2 protein in the rat liver remained unaffected after treatment with these pharmaceutical substances.

Brain Biogenic Monoamines

NA and DA

As expected, according to the known inhibitory effect of disulfiram on dopamine β-hydroxylase, the concentration of NA was found to be decreased in hypothalamus (p <.01; Figure 3a ), whereas DA content was increased in striatum (p <.001; Figure 3b ) and midbrain (p <.05; Figure 3c ) after disulfiram treatment.

The only statistically significant effect of chloramphenicol concerned a reduction of the levels of DA in striatum (p < .05; Figure 3b ). On the contrary, furazolidone elicited a pronounced elevation of NA and DA levels in hypothalamus (NA: p < .025: DA: p < .001; Figure 3a ), striatum (p < .025; Figure 3b ), and midbrain (p < .001; Figure 3c ). Metronidazole, on the other hand, failed to produce any significant changes on the levels of brain catecholamines. Finally, quinacrine reduced the levels of DA in frontal cortex (p < .005; Figure 3d ) as well as the levels of NA in hypothalamus (p < .005; Figure 3a ).

DOPAC, HVA, and DA Turnover

Disulfiram treatment resulted in a reduction of the levels of HVA in frontal cortex (p < .025; Table 1), with a parallel reduction of the turnover of DA in striatum (p < .001; Table 1) and frontal cortex (p < .005; Table 1). DA turnover was determined by using the ratio of HVA plus DOPAC divided by DA levels.

Chloramphenicol decreased the levels of DOPAC in striatum (p < .005; Table 1) and frontal cortex (p < .025; Table 1), whereas furazolidone diminished both the levels of DOPAC and HVA, as well as the turnover of DA in all brain regions determined (for hypothalamus DOPAC, HVA: p < .001, DA turnover: p < .0025; for striatum DOPAC, HVA, DA turnover: p < .001; for midbrain DOPAC: p < .005, HVA: p < .001, DA turnover: p < .0025; for frontal cortex DOPAC, HVA: p < .001, DA turnover: p < .005; Table 1). Metronidazole was also found to decrease the levels of DOPAC in all brain regions measured (hypothalamus, striatum, and frontal cortex: p < .05; midbrain: p < .001; Table 1); in addition, it reduced the levels of HVA in striatum (p < .0025; Table 1) and midbrain (p < .01; Table 1) and the turnover of DA in striatum (p < .001; Table 1). Regarding the effects of quinacrine in the metabolites of DA, the only significant effect observed was a reduction of the levels of DOPAC in frontal cortex (p < .05; Table 1).

5-HT, 5-HIAA, and 5-HT Turnover

The only influence of disulfiram on the brain serotonergic system concerned an increment of the levels of 5-HIAA, as well as of the 5-HT turnover in midbrain and frontal cortex (p < .05; Figure 4c, d ). The turnover of 5-HT was determined by using the ratio of 5-HIAA/5-HT levels.

Chloramphenicol increased the levels of 5-HT in striatum (p < .05; Figure 4b ) and midbrain (p < .025; Figure 4c ), whereas it decreased the levels of 5-HIAA and the turnover of 5-HT in striatum (p < .025; Figure 4b ).

As shown in Figure 4a–d , furazolidone treatment resulted in a profound increase of 5-HT (p < .001), as well as in a decrease of 5-HIAA (hypothalamus and frontal cortex: p < .001; striatum: p < .025; midbrain: p < .05) and 5-HT turnover (p < .001) in all brain regions measured.

Metronidazole also increased 5-HT levels in hypothalamus (p < .005; Figure 4a ), striatum (p < .001; Figure 4b ), and midbrain (p < .025; Figure 4c ), whereas it suppressed the levels of 5-HIAA in hypothalamus (p < .001; Figure 4a ). Additionally, metronidazole decreased 5-HT turnover in all brain regions examined (hypothalamus and striatum: p < .001; midbrain: p < .025; frontal cortex: p < .05; Figure 4a–d ).

Finally, quinacrine significantly elevated 5-HT in hypothalamus (p < .0025; Figure 4a ), midbrain (p < .05; Figure 4c ), and frontal cortex (p < .005; Figure 4d ), whereas it diminished 5-HIAA levels in striatum (p < .025; Figure 4b ), as well as the turnover of 5-HT in all related brain regions (hypothalamus, striatum, and frontal cortex: p < .05; midbrain: p < .005; Figure 4a–d ).

Blood Acetaldehyde Levels

As expected, disulfiram and chloramphenicol resulted in elevated blood acetaldehyde levels after ethanol administration (p < .001 and p < .0025, respectively; Figure 5). On the contrary, coadministration of quinacrine with ethanol had no effect on blood acetaldehyde (Figure 5).

DISCUSSION

The ingestion of alcohol by individuals previously treated with disulfiram brings about marked signs and symptoms, which are attributed to an increase in the concentration of acetaldehyde in the body, due to inhibition of the hepatic ALDH by disulfiram (Richie 1985). Besides disulfiram, disulfiram-like reactions may also occur in patients who have ingested alcohol after treatment with a number of pharmaceutical agents, like chloramphenicol, metronidazole, furazolidone, and quinacrine (Edwards, Fink, and Van-Dyke 1986; Fried 1980; Uri and Parks 1983; Lau et al. 1992).

As in the case with the DER, a disulfiram-like reaction is believed to be basically an “acetaldehyde syndrome.” The inhibition of the metabolism of acetaldehyde by these drugs, which theoretically leads to increased blood acetaldehyde levels, has been suggested as the main explanation for this syndrome, albeit it has not been confirmed in all cases by blood acetaldehyde measurements or protein expression of ALDH2. However, at least for metronidazole, it has been clearly shown that it does not inhibit the hepatic ALDHs nor increases blood acetaldehyde levels in rats (Kalant et al. 1972; Tillonen et al. 2000; Vasiliou, Malamas, and Marselos 1986). In the present study, we also confirmed that metronidazole has no effect on the activities of the hepatic ALDH1A1 and ALDH2. Thus, it seems that there is no proper explanation for the reported disulfiram-like reaction of this antimicrobial agent after alcohol intake. On the other hand, quinacrine, in a parallel way with metronidazole, did not exert any detectable influence on ALDH2 or ALDH1A1, although it did not increase the levels of blood acetaldehyde after ethanol ingestion. This suggests that also in the case of quinacrine, an alternative mechanism, other than the inhibition of the hepatic ALDHs, should account for its reported ethanol intolerance (Uri and Parks 1983). Regarding the effects of the other drugs tested in this study on the activity of the hepatic ALDHs, we demonstrated that disulfiram, chloramphenicol, and furazolidone inhibited ALDH2. Moreover, as expected, disulfiram and chloramphenicol significantly increased blood acetaldehyde levels after ethanol administration. ALDH1A1 was also inhibited by disulfiram and furazolidone, whereas chloramphenicol had no effect on the activity of this cytoplasmic isozyme of ALDH. Furthermore, Western blot analysis, even where ALDH2 specific activity was repressed, did not reveal respective changes in the quantity of the protein band. The decreased in vivo activity, observed in our experiments after treatment with disulfiram, chloramphenicol, or furazolidone may be due to post-translational modifications of ALDH2 protein, or to interference with processes related with substrate and coenzyme binding, as has been already suggested (Deitrich and Hellerman 1963; Deitrich and Erwin 1971).

According to the results of our experiments, ADH was not affected by any of the drugs tested, with the exception of metronidazole, which slightly inhibited the activity of the enzyme. This inhibitory effect of metronidazole on ADH has been also pointed out by Gupta et al. (Gupta, Woodley, and Fried 1970). In accordance, a toxicological analysis of a fatal case of a metronidazole-ethanol interaction, yielded elevated concentrations of serum ethanol (Cina, Russell, and Conradi 1996). On the other hand, a possible inhibitory effect on the activity of ADH and the subsequent increment in blood ethanol levels have been implicated by some investigators in the mechanism of the DER (Carper, Dorey, and Beber 1987; Sharkawi 1980). Further support to this proposition can be drawn from the fact that at least some of the symptoms of the DER, such as nausea, vertigo, and blurred vision, resemble certain symptoms encountered in acute ethanol intoxication where high concentrations of ethanol are attained (Sharkawi 1980). Accordingly, in the case of metronidazole, our findings suggest that the inhibition of ADH, even marginal, and the possible reduction of ethanol elimination should also be considered as an explanation for some of the side effects observed when the use of this agent is combined with ethanol.

In addition to the effects on the activity of the hepatic ethanol-metabolizing enzymes, the drugs tested produced a number of significant changes on the levels of brain biogenic monoamines, among which, those concerning the noradrenergic and the serotonergic system appear to be more important as a possible explanation of their disulfiram-like reactions. In contrast, the central dopaminergic system was affected in an inconsistent way, which can hardly stand as a possible mechanistic explanation of the clinical syndrome arising when these drugs are coingested with alcohol.

It is already well established that disulfiram inhibits dopamine β-hydroxylase (Goldstein et al. 1964), the enzyme that catalyzes the conversion of DA to NA in the catecholamine storage vesicles (Richie 1985). As a consequence, accumulation of DA with a respective decrease of the levels of NA in the brain, heart, and several other organs have been reported (Goldstein and Nakajima 1967; Musacchio et al. 1966). This diminution of NA, which is evident after disulfiram administration in peripheral tissues as well as in the brain, has been implicated in the hypotension observed during the DER (Kitson 1977). However, the depression of central NA is not only related with hypotension, but also with a generalized sedation (Rossetti et al. 1992). In support of this interpretation, disulfiram administration was shown to be effective in controlling psychomotor excitement in manic patients, an action attributed to its blocking effect on the synthesis of brain NA (Auriol, Bardou, and Lambic 1980). Likewise, ethanol also reduces the content of NA in the brain, which is incriminated for its well-known sedative effects (Rossetti et al. 1992). Thus, it is believed that the response to alcohol observed in patients under therapy with disulfiram could also be potentiated by a mechanism of true synergism, as in the case of concomitant use of other sedative drugs (Vasiliou, Malamas, and Marselos 1986). In the present study, disulfiram and quinacrine administration led to decreased NA levels in hypothalamus. According to these findings, it seems that also in the case of quinacrine, which does not inhibit ALDH2, the diminution of NA in the brain could, at least partly, explain some aspects of the interaction of this drug with ethanol.

Concerning the brain serotonergic system, we demonstrated that all pharmaceutical products tested had an enhancing effect on brain 5-HT levels, with the exception of disulfiram. A parallel decrease of the levels of 5-HIAA as well as of the turnover of 5-HT are consistent with the well known inhibitory action of these drugs on the activity of monoamine oxygenase (MAO) (Banerjee and Basu 1978; Befani et al. 2001; Planz, Quiring, and Palm 1972). Alcohol, on the other hand, interacts with serotonergic synaptic transmission in the brain in several ways. Even a single dose of ethanol alters various aspects of the serotonergic system. In humans, for example, the levels of 5-HT metabolites in the urine and blood increase after a single drinking session, indicating increased 5-HT release in the nervous system (LeMarquand, Pihl, and Benkelfat 1994a). In addition, animal studies also have concluded that acute ethanol administration elevates 5-HT levels within the brain (LeMarquand, Pihl, and Benkelfat 1994b; Thielen, Morzorati, and McBride 2001; Yoshimoto et al. 1992).

The concomitant administration of two or more agents that elevate synaptic levels of brain 5-HT is implicated for the induction of a toxic syndrome, known as the “serotonin syndrome” (Sternbach 1991). The most serious cases are usually caused by unintended synergism between a selective serotonin-reuptake inhibitor (SSRI) and one or more other medications, such as MAO inhibitors (Bijl 2004). Moreover, the combination of ethanol with a MAO inhibitor or a SSRI has also been reported to produce a serotonin syndrome (Holman et al. 1977; Velez et al. 2004). The clinical manifestations of this hyper-serotonergic condition are a triad of altered conscious state, autonomic dysfunction, and neuromuscular excitability (Jones and Story 2005). However, the clinical picture of the serotonin syndrome may be highly variable and a wide range of symptoms have been reported during its course (Radomski et al. 2000). Surprisingly, in a retrospective study (Radomski et al. 2000), where the full symptom profile of 24 cases of serotonin syndrome was reviewed and analyzed, one can realize that, in fact, all the symptoms observed during a “disulfiram reaction” are included in the detailed list of symptoms provided by the above-mentioned study. On the grounds of this observation, and because both ethanol and the group of agents tested in the present study increase brain 5-HT, it is tempting to speculate that the ethanol intolerance produced by these drugs could be due to enhanced 5-HT concentration in the central nervous system, within the scope of a serotonin syndrome. This may equally concern substances with the ability to inhibit ALDH2, such as chloramphenicol and furazolidone, or substances lacking this ability, such as metronidazole and quinacrine, provided that they can increase the central levels of 5-HT. In a more general concept, we suggest that clinicians should be aware of the serotonergic properties of these pharmaceutical agents, which should be prescribed very cautiously in patients under therapy with MAO inhibitors, SSRIs, or other medications that have a high probability of inducing serotonin syndrome. Due to the potentially serious nature of this condition, it seems prudent that physicians always monitor patients receiving combinations of serotonergic drugs and be alert to the possibility of “serotonergic duplication.”

On the basis of our data, it can be concluded that the ethanol intolerance produced by a number of drugs cannot be always attributed to the inhibition of the hepatic ALDH2 under the title “disulfiram-like reaction.” as has been defined until now, because some of these agents do not exhibit any influence on this enzyme. Actually, the current study is the first to show that quinacrine does not inhibit hepatic ALDHs. nor increases blood acetaldehyde. Similar findings concerning metronidazole were shown in earlier rat experiments and were confirmed in the present work. To our knowledge this is also the first report of the effects of these drugs on the protein expression of ALDH2, which, however, remained unaffected. On the other hand, the pharmaceutical agents tested, except disulfiram, produced an increase of brain 5-HT. Metronidazole also inhibited ADH activity, whereas quinacrine decreased the levels of brain NA. Therefore, it becomes apparent that an unpleasant reaction to ethanol may originate from biochemical or physiological changes other than the mere inhibition of ALDH.

The present study clearly shows that, at least in the case of metronidazole and quinacrine, the mechanism behind their reported ethanol intolerance is not located in the liver, because they do not inhibit hepatic ALDH nor increase blood acetaldehyde. Consequently, these two pharmaceutical agents do not produce a typical disulfiram-like reaction, suggesting that some adverse drug reactions after alcohol ingestion have been inappropriately characterized as such. On the other hand, regardless of the inhibition of the hepatic ALDHs, the substances tested share the common property to increase the levels of brain 5-HT. While the enhancing effects of ethanol on the central levels of 5-HT are well established. Accordingly, it could be assumed that the reaction to ethanol exhibited by these drugs may be the result of an interaction in the context of a type of a serotonin syndrome, as in the case of the concomitant administration of agents possessing serotonergic activity.