- Email: [email protected]
Hepatotoxicity of Flunitrazepam and Alcohol In Vitro
Toxicology in Vitro 13 (1999) 393±401 www.elsevier.com/locate/toxinvit
Hepatotoxicity of Flunitrazepam and Alcohol In Vitro M. S. ASSAF and M. S. ABDEL-RAHMAN* University of Medicine & Dentistry of New Jersey, New Jersey Medical School, Department of Pharmacology and Physiology, Newark, NJ 07103-2714, USA (Accepted 2 December 1998) AbstractÐFlunitrazepam (FNZ) is a benzodiazepine derivative more potent than diazepam. FNZ abuse in the US has emerged in the last few years and has a growing popularity among young people and drug abusing populations. Ethanol (EtOH) consumption with FNZ enhances euphoria and onset of action. It is postulated that FNZ and EtOH cause liver cell injury. In this study, hepatocytes are employed to study the hepatotoxicity of FNZ, EtOH and their combination (FNZ-EtOH). Hepatocytes (2 106 cells/ml) isolated from male Sprague±Dawley rats were exposed to saline, FNZ, EtOH or FNZ-EtOH in combination. The uptake of 0.4% trypan blue and the leakage of alanine aminotransferase (ALT) and aspartate aminotransferase (AST) enzymes into the incubation media were used to assess cell membrane damage of hepatocytes. Where metabolism of FNZ is nearly complete through several hepatic pathways, animal pretreatment with Phenobarbital was used to study the eect of microsomal enzyme induction on cellular injury. FNZ (0.16 mM), EtOH (32.56 mM) or their combination caused a signi®cant (P < 0.05) decrease in cell viability. Compared with control, FNZ and FNZ-EtOH in combination caused signi®cant AST leakage over the 2-hour incubation period. EtOH alone caused signi®cant AST leakage after 2 hours of incubation. The leakage of ALT enzyme was signi®cant for FNZ, EtOH and FNZ-EtOH over the 2-hour incubation period. While FNZ alone did not produce any signi®cant enzymatic leakage in the Phenobarbital pretreated groups, the leakage of ALT and AST were signi®cant for FNZ-EtOH in combination as early as 30 minutes of incubation. A signi®cant depletion (P < 0.05) of glutathione (GSH) was observed for EtOH and FNZ-EtOH in combination treated samples. This investigation suggests that FNZ and EtOH cause hepatotoxicity, and their combinations have an additive eect in increasing liver toxicity. Induction of microsomal enzymes revealed that FNZ is more hepatotoxic than the metabolites. And FNZ alone has no eect on GSH content. # 1999 Elsevier Science Ltd. All rights reserved Abbreviations: ALT = alanine aminotransferase; AST = aspartate aminotransferase; EtOH = ethanol; FNZ = ¯unitrazepam; GSH = glutathione; NADH = nicotinamide adenine dinucleotide.
INTRODUCTION 1
Flunitrazepam (Rohypnol ; 5 (o-¯ourophenyl)-1,3dihydro-1-methyl-7-nitro-2H-1,4-benzodiazepine-2one), is a benzodiazepine agonist manufactured by Homann-LaRoche Pharmaceuticals. Flunitrazepam (FNZ) is a benzodiazepine derivative, the hypnotic eect of which predominates over the sedative, anxiolytic and anticonvulsant eects characteristic of benzodiazepines (Matilla and Larni, 1980). FNZ is not marketed in the United States, but is available by prescription in Latin America, Europe, Asia and Australia (Calhoun et al., 1996; Woods and Winger, 1997). The relative weight-for-weight potency of FNZ is high, on average about 10 times as potent as diazepam. Metabolism, which is age and dose dependent, of *Corresponding author.
FNZ is nearly complete through several hepatic pathways and is excreted in the form of several metabolites through the kidneys. The most important metabolites are the 7-amino, the 1-desmethyl, and the 3-hydroxy derivatives. FNZ abuse causes dependence in humans (Calhoun et al., 1996; CEWG, 1995). Reports of FNZ abuse in the US have emerged in the past few years, and of particular concern are the low cost of the drug and its growing popularity among young people. Calhoun et al. (1996) conducted a street survey and reported that the majority of FNZ abusers are ingesting the pills with alcoholic beverages. The desirable eect is to achieve a slowed-down relaxed feeling similar to being drunk on alcohol or to enhance the euphoric eect. FNZ is used also to ameliorate symptoms of heroin withdrawal or to reduce agitation and anxiety produced by cocaine. Numerous users said that they became more talka-
0887-2333/99/$ - see front matter # 1999 Elsevier Science Ltd. All rights reserved. Printed in Great Britain PII: S0887-2333(99)00004-1
394
M. S. Assaf and M. S. Abdel-Rahman
tive, more at ease or more comfortable, particularly in social situations in which otherwise they would be uncomfortable. In addition, FNZ causes complete or partial amnesia and impaired judgment for varying periods following ingestion of the pills. Even though deaths caused by other benzodiazepines alone in the absence of other signi®cant toxicological agents or pathology are uncommon (Drummer and Ranson, 1996; Ruch et al., 1994), Heyndrickx (1987) and Drummer et al. (1993) reported death by FNZ that was either the causative factor or a signi®cant contributory factor. FNZ has the highest toxicity index, a higher relative toxicity compared with other benzodiazepines and is more toxic if misused (Drummer and Ranson, 1996). FNZ can also contribute to the toxicity of EtOH. Abusing EtOH is common among drug abusing and non-abusing populations. Some of the eects that result from EtOH exposure arise from direct action of EtOH or its metabolites, while others occur from an interaction with endogenous or exogenous substances that share a common metabolic pathway (Battiston et al., 1995; Lieber, 1990; Masini et al., 1994). The liver is the main site of EtOH degradation and metabolism with numerous and varied eects. Recent studies, which used primary cultured hepatocytes incubated with EtOH,
have suggested that agents generated during the metabolism of ETOH are responsible, in part, for cell injury. Glutathione (GSH) is important in maintaining the structural integrity of cell and organelle membranes (Deleve and Kaplowitz, 1991). Being in all mammalian cells, GSH is involved in detoxifying many xenobiotics. The main source of blood GSH is hepatic GSH transported across the sinusoidal membranes, which responds to the needs of peripheral tissues. GSH is important for organs such as the liver with intensive exposure to exogenous toxins. EtOH intake depletes the mitochondrial pool of reduced glutathione (GSH) by impairing the transport of GSH from cytosol into mitochondria (Colell et al., 1997). Isolated hepatocytes have proved useful for metabolic studies (Berry et al., 1992). Qualitatively, hepatocytes are useful for studying the pathways that mediate the conversion of a parent compound to its metabolites and to investigate the extent of hepatotoxicity. The high yield preparation of isolated hepatocytes has made hepatocytes a frequently used tool for the study of hepatic uptake, excretion, metabolism and the toxicity of drugs and other xenobiotics. This study is designed to investigate the hepatotoxicity of FNZ and FNZ in combi-
Fig. 1. Viability of rat liver cells following exposure to ¯unitrazepam, ethanol and their combinations. Hepatocytes (2 106 cells/ml) were incubated at 378C in a shaking water-bath at 30 oscillations/min with ¯unitrazepam (0.16 mM), ethanol (32.56 mM) or their combination. Aliquots were removed from incubation following 30, 60 and 120 min of exposure. Cell viability (mean2 SE) is expressed as the percentage of the total cells counted (four to six rats/experiment, three replicates/rat) at each time point studied. The viability of hepatocytes at the beginning of the experiment was greater than 90%. FNZ = ¯unitrazepam; EtOH = ethanol. a = Signi®cantly dierent from the control or ethanol (P < 0.05) using Shee's Multiple Range Test and Tukey-Kramer HSD. b = Signi®cantly dierent from FNZ (P < 0.05) using Shee's Multiple Range Test and Tukey-Kramer HSD. c = Signi®cantly dierent from the control (P < 0.05) using Shee's Multiple Range Test and Tukey-Kramer HSD.
Hepatotoxicity of ¯unitrazepam and alcohol in vitro
395
Fig. 2. Eect of ¯unitrazepam and/or ethanol on AST enzyme leakage from rat liver cells. Hepatocytes (2 106 cells/ml) were incubated at 378C in a shaking water-bath at 30 oscillations/min with ¯unitrazepam (0.16 mM), ethanol (32.56 mM) or their combination. Aliquots were removed from incubation following 30, 60 and 120 min of exposure. The data (mean2SE) is expressed as the percentage of the total AST leakage at each time point studied (four to six rats/experiment, three replicates/rat). FNZ = ¯unitrazepam; EtOH = ethanol. a = Signi®cantly dierent from the control, EtOH or FNZ/ EtOH (P < 0.05) using Shee's Multiple Range Test and Tukey-Kramer HSD. b = Signi®cantly dierent from FNZ or FNZ/EtOH (P < 0.05) using Shee's Multiple Range Test and Tukey-Kramer HSD. c = Signi®cantly dierent fron the control, FNZ or EtOH (P < 0.05) using Shee's Multiple Range Test and Tukey-Kramer HSD. d = Signi®cantly dierent fron the control or EtOH (P < 0.05) using Shee's Multiple Range Test and Tukey-Kramer HSD. e = Signi®cantly dierent fron the control or FNZ/EtOH (P < 0.05) using Shee's Multiple Range Test and Tukey-Kramer HSD.
nation with EtOH on freshly isolated rat liver cells. Additionally, EtOH contribution to the eects of FNZ on the liver is studied. Measuring trypan blue exclusion and the leakage of the intracellular enzyme ALT and AST will assess concentration and time dependent hepatotoxic eects of FNZ and FNZ combination with EtOH into the extracellular media. MATERIALS AND METHODS
Adult male Sprague±Dawley rats (200±250 g) were purchased from Taconic Farms (Germantown, NY, USA) and used in this study. The animals were housed three per cage under conditions of a controlled temperature of (23±258C) and humidity (50±55%) on a 12-hr light/dark cycle (NIH, 1985). Purina Laboratory Chow and water were provided ad lib. Hepatocytes were harvested according to Seglen (1976) using Collagenase (Type IV, Sigma Chemical Co., St Louis, MO USA) with some modi®cation. Animals were anaesthetized with Ketamine (100 mg/kg, ip) (Ketalar, Park-Davis, Morris Plains, NJ USA). The liver was perfused in situ for 8 min with calcium-free Hanks' bicarbonate buer maintained at 378C. Then, the liver was transferred
to a collagenase buer and was perfused with the collagenase buer for 15 min. After perfusion, the cells were harvested, ®ltered through four layers of cotton gauze and the ®ltrate was centrifuged for 2 min at 35 g at 48C. Isolated cells were resuspended immediately in Krebs±Hanseleit (K-H) buffer at pH 7.4 (2 0.01) containing 1% bovine serum albumin (Sigma). Cell concentration and viability were determined by 0.4% trypan blue exclusion using a hemocytometer and a light microscope (Baur et al., 1985). Hepatocytes incubation and enzyme studies Freshly isolated Hepatocytes (2 106 cells/ml) resuspended in K-H buer were incubated at 378C in a shaking water-bath at 30 oscillations/min. One to three replicates were used for each treatment. The concentration of FNZ used in this investigation (0.16 mM) was chosen on the basis of post mortem blood concentration reported in fatality victims (Drummer et al 1993; Heyndrickx, 1987). In addition, a dose±response study with FNZ was carried out to determine the dose that produced signi®cant cell death compared with control. Hepatocytes were incubated with FNZ (Sigma) at a ®nal concentration of 0.016, 0.08 and 0.16 mM dissolved in saline. A signi®cant cell death was observed at
396
M. S. Assaf and M. S. Abdel-Rahman
0.16 mM FNZ. Ethanol dose (32.56 mM) was chosen based on a previous study in our laboratory (Figliomeni and Abdel-Rahman, 1997). Therefore, we incubated hepatocytes with 0.16 mM FNZ, 32.56 mM EtOH (Pharmco Inc., Brook®eld, CT USA) or their combination in subsequent incubations. In each incubation, one to three replicates of untreated hepatocytes in suspension were used as control. Aliquots of control and treated cells were taken at 30, 60 and 120 min for trypan blue exclusion to determine cell viability. Liver cell viability and the leakage of intracellular enzymes were used to assess biochemical damage that occurred following exposure to xenobiotics. The leakage of the intracellular enzymes alanine aminotransferase (ALT) and aspartate aminotransferase (AST) from rat hepatocytes into the extracellular media were determined according to Kachmer (1976) and Story et al. (1983) after exposure to 0.16 mM FNZ, 32.56 mM EtOH or their combination. ALT activity was determined by the conversion of alanine and ketoglutamic acid to glutamic and pyruvic acids. AST activity was determined also by conversion of aspartic and ketoglutamic acids to glutamic and oxaloacetic acids. Conversion of the reduced form of nicotinamide adenine dinucleotide (NADH) to the oxidized form (NAD+) is coupled to the enzyme processes. The decrease in absorbance produced by the oxidation of NADH is measured at 340 nm.
Enzymatic leakage of AST and ALT was monitored in an aliquot of cell-free medium and compared to the total activity achieved after cell lysis. Cell-free media was obtained by centrifuging 200 ml treated cells and 200 ml saline at 1300 g at 48C for 15 min to obtain the supernatant. Cell lysis was obtained by adding 200 ml 1% Triton X-100 (Sigma) instead of saline. Control replicates were carried out simultaneously using the same conditions. The lysates were kept for enzyme assay and stored at (ÿ18 to ÿ 208C). The activity of ALT and AST was monitored by using assay kits obtained from Sigma (No. 58-50 for ALT and 59-50 for AST) and were compared to the total activity achieved after cell lysis (Triton-X tubes). Enzyme leakage was expressed as a percent of the total enzymatic activity at each time interval.
Microsomal enzyme induction Animal pretreatment with Phenobarbital causes liver microsomal enzyme induction (Mills and Jones, 1974; Roots et al., 1975) which changes the metabolic pro®le. Another group of animals was injected with Phenobarbital (100 mg/kg, ip) for three days. On day 4, animals were sacri®ced and liver perfusion was performed. Hepatocytes were isolated and incubated with FNZ, EtOH or their combination as described above.
Fig. 3. Eect of ¯unitrazepam and/or ethanol on AST leakage following phenobarbital pretreatment on rat liver cells. Hepatocytes (2 106 cells/ml) were incubated at 378C in a shaking water-bath at 30 oscillations/min with ¯unitrazepam (0.16 mM), ethanol (32.56 mM) or their combination. Aliquots were removed from incubation following 30, 60 and 120 min of exposure. The data (mean ( SD) is expressed as the percentage of the total AST leakage at each time point studied (four to six rats/experiment, three replicates/rat). FNZ = ¯unitrazepam; EtOH = ethanol. a = Signi®cantly dierent from the control (P < 0.05) using Shee's Multiple Range Test and Tukey-Kramer HSD.
Hepatotoxicity of ¯unitrazepam and alcohol in vitro
397
Fig. 4. Eect of ¯unitrazepam and/or ethanol on ALT leakage from rat liver cells. Hepatocytes (2 106 cells/ml) were incubated at 378C in a shaking water-bath at 30 oscillations/min with ¯unitrazepam (0.16 mM), ethanol (32.56 mM) or their combination. Aliquots were removed from incubation following 30, 60 and 120 min of exposure. The data (mean2SD) is expressed as the percentage of the total ALT leakage at each time point studied (four to six rats/experiment, three replicates/rat). FNZ = ¯unitrazepam; EtOH = ethanol. a = Signi®cantly dierent from the control and FNZ/EtOH (P < 0.05) using Shee's Multiple Range Test and Tukey-Kramer HSD. b = Signi®cantly dierent from the control, FNZ or EtOH (P < 0.05) using Shee's Multiple Range Test and Tukey-Kramer HSD.
Glutathione assay GSH content in isolated hepatocytes was determined as described by Battiston et al. (1995) and Beutler et al. (1963). Freshly isolated hepatocytes were resuspended in K-H buer to a ®nal concentration of 60±80 106 cells/ml. Hepatocytes were treated with 0.16 mM FNZ and 32.56 mM EtOH as described above. GSH concentration was calculated from a standard curve ranging from 0 to 20 mg% (using diluted hepatocytes as a matrix). Data analysis Analysis of variance (ANOVA) with Shee's multiple range test and Tukey-Kramer HSD test were performed for the results (mean 2 SE). Level of signi®cance in all analyses was P < 0.05. All statistical analyses were performed using Microsoft Excel 5.0 and JMP 2.0 (SAS Institute, Inc. Cary, NC, USA) software packages.
RESULTS
The cytotoxic eects of FNZ and/or EtOH were measured by the uptake of trypan blue into hepatocytes. Viability of isolated hepatocytes was greater than 91% at the time of isolation and before incubation. Control samples maintained viability higher than 77% after 2 hr of incubation at 378C. Neither
0.0016 nor 0.016 mM FNZ produced signi®cant cell death compared with the control. However, after 2 hr of incubation, exposing hepatocytes to 0.16 mM FNZ resulted in a signi®cant reduction (P < 0.05) of cell viability (e35% lower than the control). Figure 1 describes FNZ and EtOH eects on the viability of isolated rat hepatocytes. A signi®cant reduction (P < 0.05) of viability was observed as early as 30 min of hepatocyte exposure to FNZ or FNZ-EtOH in combination. The viability of FNZ or FNZ-EtOH samples was also signi®cantly lower than EtOH-treated samples over the incubation period. EtOH-treated samples produced signi®cant cell death only after 2 hr of exposure. FNZ-EtOH in combination resulted in a signi®cant reduction in cell viability (e45% lower than control) after 120 min of incubation. The hepatotoxicity of FNZ and EtOH was also evaluated by the leakage of ALT and AST from the hepatocytes into the extracellular space. Enzymatic leakages were expressed as a percent of the total enzymatic activity at each time interval. The eect of FNZ, EtOH and FNZ-EtOH in combination on AST enzyme leakage is illustrated in Fig. 2. Exposing hepatocytes to FNZ caused a signi®cant increase (P < 0.05) in AST leakage in the incubation media at 30, 60 and 120 min of incubation compared with control and EtOH. Only after 2 hr of incubation, we observed AST leakage that was
398
M. S. Assaf and M. S. Abdel-Rahman
Fig. 5. Eect of ¯unitrazepam and/or ethanol on ALT leakage following Phenobarbital pretreatment on rat liver cells. Hepatocytes (2 106 cells/ml) were incubated at 378C in a shaking water-bath at 30 oscillations/min with ¯unitrazepam (0.16 mM), ethanol (32.56 mM) or their combination. Aliquots were removed from incubation following 30, 60 and 120 min of exposure. The data (mean ( SD) is expressed as the percentage of the total ALT leakage at each time point studied (four to six rats/experiment, three replicates/rat). FNZ = ¯unitrazepam; EtOH = ethanol. a = Signi®cantly dierent from the control, FNZ or FNZ/EtOH (P < 0.05) using Shee's Multiple Range Test and Tukey-Kramer HSD. b = Signi®cantly dierent from the control, FNZ or EtOH (P < 0.05) using Shee's Multiple Range Test and Tukey-Kramer HSD.
signi®cantly higher (13%) than the control for EtOH-treated samples. FNZ-EtOH in combination resulted in AST leakage that was signi®cantly higher than control, FNZ or EtOH over all incubation periods. These observations suggest that the increase in AST leakage and the reduction in cell viability are related to the hepatotoxic eect of FNZ. The eects of phenobarbital pretreatment on the leakage of AST enzyme are shown in Fig. 3. Hepatocyte exposure to EtOH and FNZ-EtOH in combination caused AST leakage which was statistically dierent (P < 0.05) from control at 30 min of incubation and after 2 hr of incubation for FNZEtOH in combination. At 60 min of incubation, there were no signi®cant dierences between control and treated samples. These ®ndings indicate that in the phenobarbital-pretreated group, the leakage of AST enzyme after exposure to FNZ, EtOH and FNZ-EtOH are less than the leakage in the group without Phenobarbital pretreatment. Figure 4 represents the eects of exposing hepatocytes to FNZ, EtOH and FNZ-EtOH in combination on the leakage of ALT enzyme. Enzymatic leakage from cells exposed to FNZ, EtOH or FNZEtOH was statistically dierent from the control samples over the 2 hr incubation periods. After 2 hr of incubation, FNZ-EtOH treated samples pro-
duced ALT leakage that was signi®cantly (P < 0.05) greater than control and FNZ or ETOH incubations (30%, 23% and 15%, respectively). The observations from Figs 2 and 4 indicate that the enzymatic leakage of ALT or AST was caused by an additive eect of FNZ and EtOH in combination. Figure 5 shows the eect of microsomal enzyme induction on ALT enzyme leakage. The exposure of hepatocytes to the combination of FNZ and EtOH caused ALT leakage that was signi®cantly higher (P < 0.05) than control, FNZ or EtOH at 30, 60 and 120 min. Even though FNZ did not cause signi®cant ALT enzyme leakage, FNZ-EtOH in combination enhanced the cytoplasmic enzyme ALT leakage from the isolated hepatocytes. Inducing metabolism of FNZ or EtOH by microsomal enzyme induction did not result in statistically signi®cant ALT or AST enzyme leakage (Figs 3 and 5). Therefore, the increase in ALT or AST leakage because of FNZ-EtOH incubation is related to an additive eect of FNZ and EtOH. The eect on GSH content after exposing freshly isolated hepatocytes to FNZ, EtOH and FNZEtOH in combination is shown in Fig. 6. Throughout the time course studied, a slight depletion, statistically not dierent from the control, of GSH was observed after incubation with FNZ. On the other hand, EtOH and FNZ-EtOH in com-
Hepatotoxicity of ¯unitrazepam and alcohol in vitro
399
Fig. 6. Eects of ¯unitrazepam and/or ethanol on glutathione content in isolated rat liver cells. Hepatocytes (60±80 106 cells/ml) were incubated at 378C in a shaking water-bath at 30 oscillations/ min with ¯unitrazepam (0.16 mM), ethanol (32.56 mM) or their combination. Aliquots were removed from incubation following 30, 60 and 120 min of exposure. The data is expressed as (mean2SD) of glutathione concentration (nM/mg protein) which determined at each time point studied (four to six rats/experiment, three replicates/rat). FNZ = ¯unitrazepam; EtOH = ethanol. a = Signi®cantly dierent from the control or FNZ (P < 0.05) using Shee's Multiple Range Test and Tukey-Kramer HSD.
bination resulted in signi®cant GSH depletion compared with control or FNZ treated samples. The percent of GSH depletion at 30, 60 and 120 min for EtOH and FNZ-EtOH was (59.3 and 61.1), (56.5 and 60.9) and (48.8 and 51.4), respectively. The results revealed that the decline of GSH reached a steady state after 30 min of treatment. DISCUSSION
Because of damage to the hepatocellular membrane, the cell releases cytosolic enzymes into the incubation medium and loses the ability to exclude trypan blue (Baur et al., 1985). As cellular injury is accompanied by an increase in plasma membrane permeability (Smith and Orrenius, 1984), cell damage showed a good correlation with the leakage of enzymes, such as ALT from cytoplasm and AST from mitochondria (Rej, 1978). In addition, enzyme elevation in the serum of treated animals correlated with the degree of liver damage that was observed histologically (Thompson et al., 1979). The release of aminotransferases from isolated hepatocytes suspensions treated with hepatotoxins correlated also with the hepatotoxins induced toxic response in whole animal studies (Story et al., 1983). An experimental model that does not include prior EtOH exposure was utilized. EtOH can modify the eect produced by other agents because of induction or reduction in the rate by which the
parent agent is converted to its reactive metabolites (Lieber, 1990). The acute administration of EtOH can result in inhibition, while chronic exposure can result in the induction of the cytochrome P450 monooxygenase system (Sato et al., 1981a,b). For example, following chronic ETOH exposure, the hepatotoxicity produced by acetaminophen is increased (Sato et al., 1981a). However, acute EtOH administration attenuates the elevation in serum AST concentration produced by acetaminophen (Sato et al., 1981b). The method for determining cell viability is based on the principle that viable cells do not take up certain dyes, whereas dead (non-viable) cells take up the dye, particularly the nucleus staining dyes. Hepatocytes exposed to FNZ, or FNZ-EtOH in combination produced hepatocellular damage as measured by the increase of trypan blue intake (Fig. 1). The reduction in cell viability is an indication of compromising the integrity and cellular injury of hepatocytes. Exposing hepatocytes to EtOH produced hepatocellular damage only after 2 hr of exposure as indicated by an increase in the uptake of trypan blue at this time point. This observation of EtOH response is consistent with previous work in our laboratory (Figliomeni and AbdelRahman, 1997). Elevations in serum aminotransferases concentrations and mitochondrial disruption are manifestation of liver injury (Boyer and Petersen, 1991).
400
M. S. Assaf and M. S. Abdel-Rahman
The increases of AST and ALT enzymes leakage correlated with the increases in trypan blue uptake. AST enzyme was a more sensitive indicator than ALT enzyme for FNZ hepatotoxicity. This may be an eect related to the dierence in the subcellular localization of ALT compared with AST. It is important to note that the toxic eect of FNZ on AST enzyme is more than EtOH, while the eect of FNZ or EtOH alone on ALT leakage is approximately the same. The eects of EtOH on ALT and AST enzymes leakage are consistent with data obtained by other investigators (Devi et al., 1993; Figliomeni and Abdel-Rahman, 1997). Our ®ndings suggest that EtOH have an additive eect on ¯unitrazepam's hepatotoxicity. In order to examine the eect of microsomal enzyme induction on FNZ or EtOH hepatotoxicity, animals were pretreated with Phenobarbital injections (Figs 3 and 5). Treatment of animals with Phenobarbital enhanced N-demethylation and 3-hydroxylation enzymes (Levine, 1983). Thus, Phenobarbital most likely induces FNZ metabolism generating more of the 3-hydroxy and desmethyl metabolites of FNZ. After 2 hr of incubation, it is worth noting that pretreatment with phenobarbital decreased AST and ALT percent leakage in the combination treatment compared with the one without Phenobarbital pretreatment (15.5% v. 59% and 52.8% v. 68%, respectively). The leakage of ALT enzyme is higher than AST enzyme from the phenobarbital-pretreated group. This observation suggests that while AST is a more sensitive indicator than ALT for FNZ hepatotoxicity on the group without microsomal enzyme induction, ALT enzyme is an appropriate indicator for hepatotoxicity determination after the induction of FNZ metabolism. In conclusion, ALT and AST leakage in this study revealed that the hepatotoxic eect of FNZ is more signi®cant with the parent compound than the metabolites. Liver cells are probably capable of sustaining severe depletion of the cytosolic pool of glutathione without signi®cant cellular damage (Wendel and Feuerstein, 1981). Consistent with previous published work, EtOH caused signi®cant depletion of GSH (Colell et al., 1997; Devi et al., 1993). FNZEtOH in combination led to signi®cant glutathione depletion; however, the extent of depletion is not statistically dierent from the depletion caused by EtOH alone. Therefore, under these investigation conditions, EtOH decreases liver glutathione content compared with FNZ and the eect in the combination treatment is very likely due to EtOH. Therefore, FNZ has no eect on GSH content in freshly isolated hepatocytes. This investigation is an in vitro model for FNZ hepatotoxicity. Further investigations are currently in progress at our laboratory in order to determine FNZ hepatotoxicity in vivo and to look at other parameters of cellular injury.
REFERENCES
Battiston L., Moretti M., Tulissi P., Michili L., Marchi P., Mazzoran L., Lunazzi G. and Pozzato G. (1995) Hepatic glutathione determination after ethanol administration in rat: Evidence of the ®rst pass metabolism of ethanol. Life Sciences 56, 241±248. Baur H., Kasperek S. and Pfa E. (1985) Criteria of viability of isolated liver cells. Physiological Chemistry 356, 827±838. Berry M. N., Halls H. J. and Grivell M. B. (1992) Techniques for pharmacological and toxicological studies with isolated hepatocytes suspension. Life Sciences 51, 1±16. Beutler E., Duron O. and Kelly B. M. (1963) Improved method for the determination of blood glutathione. Journal of Laboratory Clinical Medicine 61, 882±888. Boyer C. S. and Petersen D. R. (1991) Hepatic biochemical changes as result of acute cocaine administration in the mouse. Hepatology 14, 1209±1216. Calhoun C. S., Wesson D. R., Galloway G. P. and Smith D. E. (1996) Abuse of ¯unitrazepam (Rohypnol) and other benzodiazepine in Austin and south Texas. Journal of Psychoactive Drugs 28, 183±189. Colell A., Garcia-Ruiz C., Morales A., Ballesta A., Ookhtens M., Rodes J., Kaplowitz N. and FernandezCheca J. C. (1997) Transport of reduced glutathione in hepatic mitochondria and mitoplasts from ethanol-treated rats: Eects of membrane physical properties and Sadenosyl-l-methionine. Hepatology 26, 699±708. CEWG (1995) Epidemiologic trends in drug abuse-advance report: Fact sheet on Rohypnol. Community Epidemiology Work Group National Institute on Drug Abuse, http:// www.health.org/pubs/factsht/rohypnol.htm. Deleve L. D. and Kaplowitz N. (1991) Glutathione metabolism and its role in hepatotoxicity. Pharmacology and Therapeutics 52, 287±305. Devi B. G., Henderson G. I., Frosto T. A. and Schenker S. (1993) Eect of ethanol on rat fetal hepatocytes: Studies on cell replication, lipid peroxidation and glutathione. Hepatology 18, 648±659. Drummer O. H. and Ranson D. L. (1996) Sudden death and benzodiazepines. American Journal of Forensic Medicine and Pathology 17, 336±342. Drummer O. H., Syrjanen M. L. and Cordner S. M. (1993) Deaths involving the benzodiazepine ¯unitrazepam. American Journal of Forensic Medicine and Pathology 14, 238±243. Figliomeni M. L. and Abdel-Rahman M. S. (1997) The role of ethanol exposure on cocaine metabolism in rat hepatocytes. Journal of Applied Toxicology 17, 105±112. Heyndrickx B. (1987) Fatal intoxication due to ¯unitrazepam. Journal of Analytical Toxicology 11, 278. Kachmer J. R. (1976) Enzymes. In Fundamentals of Clinical Chemistry, ed. N. W. Teitz, pp. 674. W. B. Saunders, Philadelphia. Levine W. G. (1983) Glutathione and hepatic mixed-function oxidase activity. Drug Metabolism Reviews 14, 909± 930. Lieber C. S. (1990) Mechanism of ethanol induced hepatic injury. Pharmacology and Therapeutics 46, 1±41. Masini A., Ceccarelli D., Gallesi D., Giovannini F. and Trenti T. (1994) Lipid hydroperoxide induced mitochondrial dysfunction following acute ethanol intoxication in rats. The critical role for mitochondrial reduced glutathione. Biochemical Pharmacology 47, 217±224. Mattila M. A. K. and Larni H. M. (1980) Flunitrazepam: A review of its pharmacological properties and therapeutic use. Drugs 20, 353±374. Mills E. S. and Jones A. L. (1974) Ultrastructure concepts of drug metabolism. II. The hepatocytes: phenobarbital and microsomal enzyme induction. American Journal of Drug and Alcohol Abuse 1, 271±298.
Hepatotoxicity of ¯unitrazepam and alcohol in vitro NIH (1985) Guide for care and use of laboratory animals. 11±28. NIH Contact No. NOI-RR-2-2135. National Institute of Health, Bethesda, MD. Rej R. (1978) Aspartate aminotransferase activity and isozyme proportion in human liver tissues. Clinical Chemistry 24, 1971±1979. Roots I., Saalfrank K. and Hildebrandt A. G. (1975) Comparison of methods to study enzyme induction in man. Advanced Experimental Medicine and Biology 58, 485±502. Ruch R., Jaber B., Hilton S. and Goertz E. (1994) Oral benzodiazepines safety in overdose: 20 years of experience. Neuropsychopharmacology 10, 227S. Sato C., Matsuda Y. and Lieber C. S. (1981a) Increased hepatotoxicity of acetaminophen after chronic ethanol consumption in the rat. Gastroenterology 80, 140±148. Sato C., Nakano M. and Lieber C. S. (1981b) Prevention of acetaminophen-induced hepatotoxicity by acute ethanol administration in the rat: Comparison with carbon tetrachloride-induced hepatotoxicity. Journal of
401
Pharmacology and Experimental Therapeutics 218, 805± 809. Seglen P. O. (1976) Preparation of rat liver cells. Methods in Cell Biology 13, 29±83. Smith M. T. and Orrenius S. (1984) Drug Metabolism and Drug Toxicity, ed. J. R. Mitchell and M. G. Horning, pp. 71. Raven Press, New York. Story D. L., Gee S. J. and Tyson C. A. (1983) Response of isolated hepatocytes to organic and inorganic cytotoxins. Journal of Toxicology and Environmental Health 11, 443±450. Thompson M. L., Shuster L. and Shaw K. (1979) Cocaine induced hepatic necrosis in mice-the role of cocaine metabolism. Biochemical Journal 225, 565±572. Wendel A. and Feuerstein S. (1981) Drug induced lipid peroxidation in mice: Modulation by monoxygenase activity, glutathione and selenium status. Biochemical Pharmacology 30, 2513±2520. Woods J. H. and Winger G. (1997) Abuse liability of ¯unitrazepam. Journal of Clinical Psychopharmacology 17 (Suppl. 2), 1S-57S.
Hepatotoxicity of Flunitrazepam and Alcohol In Vitro M. S. ASSAF and M. S. ABDEL-RAHMAN* University of Medicine & Dentistry of New Jersey, New Jersey Medical School, Department of Pharmacology and Physiology, Newark, NJ 07103-2714, USA (Accepted 2 December 1998) AbstractÐFlunitrazepam (FNZ) is a benzodiazepine derivative more potent than diazepam. FNZ abuse in the US has emerged in the last few years and has a growing popularity among young people and drug abusing populations. Ethanol (EtOH) consumption with FNZ enhances euphoria and onset of action. It is postulated that FNZ and EtOH cause liver cell injury. In this study, hepatocytes are employed to study the hepatotoxicity of FNZ, EtOH and their combination (FNZ-EtOH). Hepatocytes (2 106 cells/ml) isolated from male Sprague±Dawley rats were exposed to saline, FNZ, EtOH or FNZ-EtOH in combination. The uptake of 0.4% trypan blue and the leakage of alanine aminotransferase (ALT) and aspartate aminotransferase (AST) enzymes into the incubation media were used to assess cell membrane damage of hepatocytes. Where metabolism of FNZ is nearly complete through several hepatic pathways, animal pretreatment with Phenobarbital was used to study the eect of microsomal enzyme induction on cellular injury. FNZ (0.16 mM), EtOH (32.56 mM) or their combination caused a signi®cant (P < 0.05) decrease in cell viability. Compared with control, FNZ and FNZ-EtOH in combination caused signi®cant AST leakage over the 2-hour incubation period. EtOH alone caused signi®cant AST leakage after 2 hours of incubation. The leakage of ALT enzyme was signi®cant for FNZ, EtOH and FNZ-EtOH over the 2-hour incubation period. While FNZ alone did not produce any signi®cant enzymatic leakage in the Phenobarbital pretreated groups, the leakage of ALT and AST were signi®cant for FNZ-EtOH in combination as early as 30 minutes of incubation. A signi®cant depletion (P < 0.05) of glutathione (GSH) was observed for EtOH and FNZ-EtOH in combination treated samples. This investigation suggests that FNZ and EtOH cause hepatotoxicity, and their combinations have an additive eect in increasing liver toxicity. Induction of microsomal enzymes revealed that FNZ is more hepatotoxic than the metabolites. And FNZ alone has no eect on GSH content. # 1999 Elsevier Science Ltd. All rights reserved Abbreviations: ALT = alanine aminotransferase; AST = aspartate aminotransferase; EtOH = ethanol; FNZ = ¯unitrazepam; GSH = glutathione; NADH = nicotinamide adenine dinucleotide.
INTRODUCTION 1
Flunitrazepam (Rohypnol ; 5 (o-¯ourophenyl)-1,3dihydro-1-methyl-7-nitro-2H-1,4-benzodiazepine-2one), is a benzodiazepine agonist manufactured by Homann-LaRoche Pharmaceuticals. Flunitrazepam (FNZ) is a benzodiazepine derivative, the hypnotic eect of which predominates over the sedative, anxiolytic and anticonvulsant eects characteristic of benzodiazepines (Matilla and Larni, 1980). FNZ is not marketed in the United States, but is available by prescription in Latin America, Europe, Asia and Australia (Calhoun et al., 1996; Woods and Winger, 1997). The relative weight-for-weight potency of FNZ is high, on average about 10 times as potent as diazepam. Metabolism, which is age and dose dependent, of *Corresponding author.
FNZ is nearly complete through several hepatic pathways and is excreted in the form of several metabolites through the kidneys. The most important metabolites are the 7-amino, the 1-desmethyl, and the 3-hydroxy derivatives. FNZ abuse causes dependence in humans (Calhoun et al., 1996; CEWG, 1995). Reports of FNZ abuse in the US have emerged in the past few years, and of particular concern are the low cost of the drug and its growing popularity among young people. Calhoun et al. (1996) conducted a street survey and reported that the majority of FNZ abusers are ingesting the pills with alcoholic beverages. The desirable eect is to achieve a slowed-down relaxed feeling similar to being drunk on alcohol or to enhance the euphoric eect. FNZ is used also to ameliorate symptoms of heroin withdrawal or to reduce agitation and anxiety produced by cocaine. Numerous users said that they became more talka-
0887-2333/99/$ - see front matter # 1999 Elsevier Science Ltd. All rights reserved. Printed in Great Britain PII: S0887-2333(99)00004-1
394
M. S. Assaf and M. S. Abdel-Rahman
tive, more at ease or more comfortable, particularly in social situations in which otherwise they would be uncomfortable. In addition, FNZ causes complete or partial amnesia and impaired judgment for varying periods following ingestion of the pills. Even though deaths caused by other benzodiazepines alone in the absence of other signi®cant toxicological agents or pathology are uncommon (Drummer and Ranson, 1996; Ruch et al., 1994), Heyndrickx (1987) and Drummer et al. (1993) reported death by FNZ that was either the causative factor or a signi®cant contributory factor. FNZ has the highest toxicity index, a higher relative toxicity compared with other benzodiazepines and is more toxic if misused (Drummer and Ranson, 1996). FNZ can also contribute to the toxicity of EtOH. Abusing EtOH is common among drug abusing and non-abusing populations. Some of the eects that result from EtOH exposure arise from direct action of EtOH or its metabolites, while others occur from an interaction with endogenous or exogenous substances that share a common metabolic pathway (Battiston et al., 1995; Lieber, 1990; Masini et al., 1994). The liver is the main site of EtOH degradation and metabolism with numerous and varied eects. Recent studies, which used primary cultured hepatocytes incubated with EtOH,
have suggested that agents generated during the metabolism of ETOH are responsible, in part, for cell injury. Glutathione (GSH) is important in maintaining the structural integrity of cell and organelle membranes (Deleve and Kaplowitz, 1991). Being in all mammalian cells, GSH is involved in detoxifying many xenobiotics. The main source of blood GSH is hepatic GSH transported across the sinusoidal membranes, which responds to the needs of peripheral tissues. GSH is important for organs such as the liver with intensive exposure to exogenous toxins. EtOH intake depletes the mitochondrial pool of reduced glutathione (GSH) by impairing the transport of GSH from cytosol into mitochondria (Colell et al., 1997). Isolated hepatocytes have proved useful for metabolic studies (Berry et al., 1992). Qualitatively, hepatocytes are useful for studying the pathways that mediate the conversion of a parent compound to its metabolites and to investigate the extent of hepatotoxicity. The high yield preparation of isolated hepatocytes has made hepatocytes a frequently used tool for the study of hepatic uptake, excretion, metabolism and the toxicity of drugs and other xenobiotics. This study is designed to investigate the hepatotoxicity of FNZ and FNZ in combi-
Fig. 1. Viability of rat liver cells following exposure to ¯unitrazepam, ethanol and their combinations. Hepatocytes (2 106 cells/ml) were incubated at 378C in a shaking water-bath at 30 oscillations/min with ¯unitrazepam (0.16 mM), ethanol (32.56 mM) or their combination. Aliquots were removed from incubation following 30, 60 and 120 min of exposure. Cell viability (mean2 SE) is expressed as the percentage of the total cells counted (four to six rats/experiment, three replicates/rat) at each time point studied. The viability of hepatocytes at the beginning of the experiment was greater than 90%. FNZ = ¯unitrazepam; EtOH = ethanol. a = Signi®cantly dierent from the control or ethanol (P < 0.05) using Shee's Multiple Range Test and Tukey-Kramer HSD. b = Signi®cantly dierent from FNZ (P < 0.05) using Shee's Multiple Range Test and Tukey-Kramer HSD. c = Signi®cantly dierent from the control (P < 0.05) using Shee's Multiple Range Test and Tukey-Kramer HSD.
Hepatotoxicity of ¯unitrazepam and alcohol in vitro
395
Fig. 2. Eect of ¯unitrazepam and/or ethanol on AST enzyme leakage from rat liver cells. Hepatocytes (2 106 cells/ml) were incubated at 378C in a shaking water-bath at 30 oscillations/min with ¯unitrazepam (0.16 mM), ethanol (32.56 mM) or their combination. Aliquots were removed from incubation following 30, 60 and 120 min of exposure. The data (mean2SE) is expressed as the percentage of the total AST leakage at each time point studied (four to six rats/experiment, three replicates/rat). FNZ = ¯unitrazepam; EtOH = ethanol. a = Signi®cantly dierent from the control, EtOH or FNZ/ EtOH (P < 0.05) using Shee's Multiple Range Test and Tukey-Kramer HSD. b = Signi®cantly dierent from FNZ or FNZ/EtOH (P < 0.05) using Shee's Multiple Range Test and Tukey-Kramer HSD. c = Signi®cantly dierent fron the control, FNZ or EtOH (P < 0.05) using Shee's Multiple Range Test and Tukey-Kramer HSD. d = Signi®cantly dierent fron the control or EtOH (P < 0.05) using Shee's Multiple Range Test and Tukey-Kramer HSD. e = Signi®cantly dierent fron the control or FNZ/EtOH (P < 0.05) using Shee's Multiple Range Test and Tukey-Kramer HSD.
nation with EtOH on freshly isolated rat liver cells. Additionally, EtOH contribution to the eects of FNZ on the liver is studied. Measuring trypan blue exclusion and the leakage of the intracellular enzyme ALT and AST will assess concentration and time dependent hepatotoxic eects of FNZ and FNZ combination with EtOH into the extracellular media. MATERIALS AND METHODS
Adult male Sprague±Dawley rats (200±250 g) were purchased from Taconic Farms (Germantown, NY, USA) and used in this study. The animals were housed three per cage under conditions of a controlled temperature of (23±258C) and humidity (50±55%) on a 12-hr light/dark cycle (NIH, 1985). Purina Laboratory Chow and water were provided ad lib. Hepatocytes were harvested according to Seglen (1976) using Collagenase (Type IV, Sigma Chemical Co., St Louis, MO USA) with some modi®cation. Animals were anaesthetized with Ketamine (100 mg/kg, ip) (Ketalar, Park-Davis, Morris Plains, NJ USA). The liver was perfused in situ for 8 min with calcium-free Hanks' bicarbonate buer maintained at 378C. Then, the liver was transferred
to a collagenase buer and was perfused with the collagenase buer for 15 min. After perfusion, the cells were harvested, ®ltered through four layers of cotton gauze and the ®ltrate was centrifuged for 2 min at 35 g at 48C. Isolated cells were resuspended immediately in Krebs±Hanseleit (K-H) buffer at pH 7.4 (2 0.01) containing 1% bovine serum albumin (Sigma). Cell concentration and viability were determined by 0.4% trypan blue exclusion using a hemocytometer and a light microscope (Baur et al., 1985). Hepatocytes incubation and enzyme studies Freshly isolated Hepatocytes (2 106 cells/ml) resuspended in K-H buer were incubated at 378C in a shaking water-bath at 30 oscillations/min. One to three replicates were used for each treatment. The concentration of FNZ used in this investigation (0.16 mM) was chosen on the basis of post mortem blood concentration reported in fatality victims (Drummer et al 1993; Heyndrickx, 1987). In addition, a dose±response study with FNZ was carried out to determine the dose that produced signi®cant cell death compared with control. Hepatocytes were incubated with FNZ (Sigma) at a ®nal concentration of 0.016, 0.08 and 0.16 mM dissolved in saline. A signi®cant cell death was observed at
396
M. S. Assaf and M. S. Abdel-Rahman
0.16 mM FNZ. Ethanol dose (32.56 mM) was chosen based on a previous study in our laboratory (Figliomeni and Abdel-Rahman, 1997). Therefore, we incubated hepatocytes with 0.16 mM FNZ, 32.56 mM EtOH (Pharmco Inc., Brook®eld, CT USA) or their combination in subsequent incubations. In each incubation, one to three replicates of untreated hepatocytes in suspension were used as control. Aliquots of control and treated cells were taken at 30, 60 and 120 min for trypan blue exclusion to determine cell viability. Liver cell viability and the leakage of intracellular enzymes were used to assess biochemical damage that occurred following exposure to xenobiotics. The leakage of the intracellular enzymes alanine aminotransferase (ALT) and aspartate aminotransferase (AST) from rat hepatocytes into the extracellular media were determined according to Kachmer (1976) and Story et al. (1983) after exposure to 0.16 mM FNZ, 32.56 mM EtOH or their combination. ALT activity was determined by the conversion of alanine and ketoglutamic acid to glutamic and pyruvic acids. AST activity was determined also by conversion of aspartic and ketoglutamic acids to glutamic and oxaloacetic acids. Conversion of the reduced form of nicotinamide adenine dinucleotide (NADH) to the oxidized form (NAD+) is coupled to the enzyme processes. The decrease in absorbance produced by the oxidation of NADH is measured at 340 nm.
Enzymatic leakage of AST and ALT was monitored in an aliquot of cell-free medium and compared to the total activity achieved after cell lysis. Cell-free media was obtained by centrifuging 200 ml treated cells and 200 ml saline at 1300 g at 48C for 15 min to obtain the supernatant. Cell lysis was obtained by adding 200 ml 1% Triton X-100 (Sigma) instead of saline. Control replicates were carried out simultaneously using the same conditions. The lysates were kept for enzyme assay and stored at (ÿ18 to ÿ 208C). The activity of ALT and AST was monitored by using assay kits obtained from Sigma (No. 58-50 for ALT and 59-50 for AST) and were compared to the total activity achieved after cell lysis (Triton-X tubes). Enzyme leakage was expressed as a percent of the total enzymatic activity at each time interval.
Microsomal enzyme induction Animal pretreatment with Phenobarbital causes liver microsomal enzyme induction (Mills and Jones, 1974; Roots et al., 1975) which changes the metabolic pro®le. Another group of animals was injected with Phenobarbital (100 mg/kg, ip) for three days. On day 4, animals were sacri®ced and liver perfusion was performed. Hepatocytes were isolated and incubated with FNZ, EtOH or their combination as described above.
Fig. 3. Eect of ¯unitrazepam and/or ethanol on AST leakage following phenobarbital pretreatment on rat liver cells. Hepatocytes (2 106 cells/ml) were incubated at 378C in a shaking water-bath at 30 oscillations/min with ¯unitrazepam (0.16 mM), ethanol (32.56 mM) or their combination. Aliquots were removed from incubation following 30, 60 and 120 min of exposure. The data (mean ( SD) is expressed as the percentage of the total AST leakage at each time point studied (four to six rats/experiment, three replicates/rat). FNZ = ¯unitrazepam; EtOH = ethanol. a = Signi®cantly dierent from the control (P < 0.05) using Shee's Multiple Range Test and Tukey-Kramer HSD.
Hepatotoxicity of ¯unitrazepam and alcohol in vitro
397
Fig. 4. Eect of ¯unitrazepam and/or ethanol on ALT leakage from rat liver cells. Hepatocytes (2 106 cells/ml) were incubated at 378C in a shaking water-bath at 30 oscillations/min with ¯unitrazepam (0.16 mM), ethanol (32.56 mM) or their combination. Aliquots were removed from incubation following 30, 60 and 120 min of exposure. The data (mean2SD) is expressed as the percentage of the total ALT leakage at each time point studied (four to six rats/experiment, three replicates/rat). FNZ = ¯unitrazepam; EtOH = ethanol. a = Signi®cantly dierent from the control and FNZ/EtOH (P < 0.05) using Shee's Multiple Range Test and Tukey-Kramer HSD. b = Signi®cantly dierent from the control, FNZ or EtOH (P < 0.05) using Shee's Multiple Range Test and Tukey-Kramer HSD.
Glutathione assay GSH content in isolated hepatocytes was determined as described by Battiston et al. (1995) and Beutler et al. (1963). Freshly isolated hepatocytes were resuspended in K-H buer to a ®nal concentration of 60±80 106 cells/ml. Hepatocytes were treated with 0.16 mM FNZ and 32.56 mM EtOH as described above. GSH concentration was calculated from a standard curve ranging from 0 to 20 mg% (using diluted hepatocytes as a matrix). Data analysis Analysis of variance (ANOVA) with Shee's multiple range test and Tukey-Kramer HSD test were performed for the results (mean 2 SE). Level of signi®cance in all analyses was P < 0.05. All statistical analyses were performed using Microsoft Excel 5.0 and JMP 2.0 (SAS Institute, Inc. Cary, NC, USA) software packages.
RESULTS
The cytotoxic eects of FNZ and/or EtOH were measured by the uptake of trypan blue into hepatocytes. Viability of isolated hepatocytes was greater than 91% at the time of isolation and before incubation. Control samples maintained viability higher than 77% after 2 hr of incubation at 378C. Neither
0.0016 nor 0.016 mM FNZ produced signi®cant cell death compared with the control. However, after 2 hr of incubation, exposing hepatocytes to 0.16 mM FNZ resulted in a signi®cant reduction (P < 0.05) of cell viability (e35% lower than the control). Figure 1 describes FNZ and EtOH eects on the viability of isolated rat hepatocytes. A signi®cant reduction (P < 0.05) of viability was observed as early as 30 min of hepatocyte exposure to FNZ or FNZ-EtOH in combination. The viability of FNZ or FNZ-EtOH samples was also signi®cantly lower than EtOH-treated samples over the incubation period. EtOH-treated samples produced signi®cant cell death only after 2 hr of exposure. FNZ-EtOH in combination resulted in a signi®cant reduction in cell viability (e45% lower than control) after 120 min of incubation. The hepatotoxicity of FNZ and EtOH was also evaluated by the leakage of ALT and AST from the hepatocytes into the extracellular space. Enzymatic leakages were expressed as a percent of the total enzymatic activity at each time interval. The eect of FNZ, EtOH and FNZ-EtOH in combination on AST enzyme leakage is illustrated in Fig. 2. Exposing hepatocytes to FNZ caused a signi®cant increase (P < 0.05) in AST leakage in the incubation media at 30, 60 and 120 min of incubation compared with control and EtOH. Only after 2 hr of incubation, we observed AST leakage that was
398
M. S. Assaf and M. S. Abdel-Rahman
Fig. 5. Eect of ¯unitrazepam and/or ethanol on ALT leakage following Phenobarbital pretreatment on rat liver cells. Hepatocytes (2 106 cells/ml) were incubated at 378C in a shaking water-bath at 30 oscillations/min with ¯unitrazepam (0.16 mM), ethanol (32.56 mM) or their combination. Aliquots were removed from incubation following 30, 60 and 120 min of exposure. The data (mean ( SD) is expressed as the percentage of the total ALT leakage at each time point studied (four to six rats/experiment, three replicates/rat). FNZ = ¯unitrazepam; EtOH = ethanol. a = Signi®cantly dierent from the control, FNZ or FNZ/EtOH (P < 0.05) using Shee's Multiple Range Test and Tukey-Kramer HSD. b = Signi®cantly dierent from the control, FNZ or EtOH (P < 0.05) using Shee's Multiple Range Test and Tukey-Kramer HSD.
signi®cantly higher (13%) than the control for EtOH-treated samples. FNZ-EtOH in combination resulted in AST leakage that was signi®cantly higher than control, FNZ or EtOH over all incubation periods. These observations suggest that the increase in AST leakage and the reduction in cell viability are related to the hepatotoxic eect of FNZ. The eects of phenobarbital pretreatment on the leakage of AST enzyme are shown in Fig. 3. Hepatocyte exposure to EtOH and FNZ-EtOH in combination caused AST leakage which was statistically dierent (P < 0.05) from control at 30 min of incubation and after 2 hr of incubation for FNZEtOH in combination. At 60 min of incubation, there were no signi®cant dierences between control and treated samples. These ®ndings indicate that in the phenobarbital-pretreated group, the leakage of AST enzyme after exposure to FNZ, EtOH and FNZ-EtOH are less than the leakage in the group without Phenobarbital pretreatment. Figure 4 represents the eects of exposing hepatocytes to FNZ, EtOH and FNZ-EtOH in combination on the leakage of ALT enzyme. Enzymatic leakage from cells exposed to FNZ, EtOH or FNZEtOH was statistically dierent from the control samples over the 2 hr incubation periods. After 2 hr of incubation, FNZ-EtOH treated samples pro-
duced ALT leakage that was signi®cantly (P < 0.05) greater than control and FNZ or ETOH incubations (30%, 23% and 15%, respectively). The observations from Figs 2 and 4 indicate that the enzymatic leakage of ALT or AST was caused by an additive eect of FNZ and EtOH in combination. Figure 5 shows the eect of microsomal enzyme induction on ALT enzyme leakage. The exposure of hepatocytes to the combination of FNZ and EtOH caused ALT leakage that was signi®cantly higher (P < 0.05) than control, FNZ or EtOH at 30, 60 and 120 min. Even though FNZ did not cause signi®cant ALT enzyme leakage, FNZ-EtOH in combination enhanced the cytoplasmic enzyme ALT leakage from the isolated hepatocytes. Inducing metabolism of FNZ or EtOH by microsomal enzyme induction did not result in statistically signi®cant ALT or AST enzyme leakage (Figs 3 and 5). Therefore, the increase in ALT or AST leakage because of FNZ-EtOH incubation is related to an additive eect of FNZ and EtOH. The eect on GSH content after exposing freshly isolated hepatocytes to FNZ, EtOH and FNZEtOH in combination is shown in Fig. 6. Throughout the time course studied, a slight depletion, statistically not dierent from the control, of GSH was observed after incubation with FNZ. On the other hand, EtOH and FNZ-EtOH in com-
Hepatotoxicity of ¯unitrazepam and alcohol in vitro
399
Fig. 6. Eects of ¯unitrazepam and/or ethanol on glutathione content in isolated rat liver cells. Hepatocytes (60±80 106 cells/ml) were incubated at 378C in a shaking water-bath at 30 oscillations/ min with ¯unitrazepam (0.16 mM), ethanol (32.56 mM) or their combination. Aliquots were removed from incubation following 30, 60 and 120 min of exposure. The data is expressed as (mean2SD) of glutathione concentration (nM/mg protein) which determined at each time point studied (four to six rats/experiment, three replicates/rat). FNZ = ¯unitrazepam; EtOH = ethanol. a = Signi®cantly dierent from the control or FNZ (P < 0.05) using Shee's Multiple Range Test and Tukey-Kramer HSD.
bination resulted in signi®cant GSH depletion compared with control or FNZ treated samples. The percent of GSH depletion at 30, 60 and 120 min for EtOH and FNZ-EtOH was (59.3 and 61.1), (56.5 and 60.9) and (48.8 and 51.4), respectively. The results revealed that the decline of GSH reached a steady state after 30 min of treatment. DISCUSSION
Because of damage to the hepatocellular membrane, the cell releases cytosolic enzymes into the incubation medium and loses the ability to exclude trypan blue (Baur et al., 1985). As cellular injury is accompanied by an increase in plasma membrane permeability (Smith and Orrenius, 1984), cell damage showed a good correlation with the leakage of enzymes, such as ALT from cytoplasm and AST from mitochondria (Rej, 1978). In addition, enzyme elevation in the serum of treated animals correlated with the degree of liver damage that was observed histologically (Thompson et al., 1979). The release of aminotransferases from isolated hepatocytes suspensions treated with hepatotoxins correlated also with the hepatotoxins induced toxic response in whole animal studies (Story et al., 1983). An experimental model that does not include prior EtOH exposure was utilized. EtOH can modify the eect produced by other agents because of induction or reduction in the rate by which the
parent agent is converted to its reactive metabolites (Lieber, 1990). The acute administration of EtOH can result in inhibition, while chronic exposure can result in the induction of the cytochrome P450 monooxygenase system (Sato et al., 1981a,b). For example, following chronic ETOH exposure, the hepatotoxicity produced by acetaminophen is increased (Sato et al., 1981a). However, acute EtOH administration attenuates the elevation in serum AST concentration produced by acetaminophen (Sato et al., 1981b). The method for determining cell viability is based on the principle that viable cells do not take up certain dyes, whereas dead (non-viable) cells take up the dye, particularly the nucleus staining dyes. Hepatocytes exposed to FNZ, or FNZ-EtOH in combination produced hepatocellular damage as measured by the increase of trypan blue intake (Fig. 1). The reduction in cell viability is an indication of compromising the integrity and cellular injury of hepatocytes. Exposing hepatocytes to EtOH produced hepatocellular damage only after 2 hr of exposure as indicated by an increase in the uptake of trypan blue at this time point. This observation of EtOH response is consistent with previous work in our laboratory (Figliomeni and AbdelRahman, 1997). Elevations in serum aminotransferases concentrations and mitochondrial disruption are manifestation of liver injury (Boyer and Petersen, 1991).
400
M. S. Assaf and M. S. Abdel-Rahman
The increases of AST and ALT enzymes leakage correlated with the increases in trypan blue uptake. AST enzyme was a more sensitive indicator than ALT enzyme for FNZ hepatotoxicity. This may be an eect related to the dierence in the subcellular localization of ALT compared with AST. It is important to note that the toxic eect of FNZ on AST enzyme is more than EtOH, while the eect of FNZ or EtOH alone on ALT leakage is approximately the same. The eects of EtOH on ALT and AST enzymes leakage are consistent with data obtained by other investigators (Devi et al., 1993; Figliomeni and Abdel-Rahman, 1997). Our ®ndings suggest that EtOH have an additive eect on ¯unitrazepam's hepatotoxicity. In order to examine the eect of microsomal enzyme induction on FNZ or EtOH hepatotoxicity, animals were pretreated with Phenobarbital injections (Figs 3 and 5). Treatment of animals with Phenobarbital enhanced N-demethylation and 3-hydroxylation enzymes (Levine, 1983). Thus, Phenobarbital most likely induces FNZ metabolism generating more of the 3-hydroxy and desmethyl metabolites of FNZ. After 2 hr of incubation, it is worth noting that pretreatment with phenobarbital decreased AST and ALT percent leakage in the combination treatment compared with the one without Phenobarbital pretreatment (15.5% v. 59% and 52.8% v. 68%, respectively). The leakage of ALT enzyme is higher than AST enzyme from the phenobarbital-pretreated group. This observation suggests that while AST is a more sensitive indicator than ALT for FNZ hepatotoxicity on the group without microsomal enzyme induction, ALT enzyme is an appropriate indicator for hepatotoxicity determination after the induction of FNZ metabolism. In conclusion, ALT and AST leakage in this study revealed that the hepatotoxic eect of FNZ is more signi®cant with the parent compound than the metabolites. Liver cells are probably capable of sustaining severe depletion of the cytosolic pool of glutathione without signi®cant cellular damage (Wendel and Feuerstein, 1981). Consistent with previous published work, EtOH caused signi®cant depletion of GSH (Colell et al., 1997; Devi et al., 1993). FNZEtOH in combination led to signi®cant glutathione depletion; however, the extent of depletion is not statistically dierent from the depletion caused by EtOH alone. Therefore, under these investigation conditions, EtOH decreases liver glutathione content compared with FNZ and the eect in the combination treatment is very likely due to EtOH. Therefore, FNZ has no eect on GSH content in freshly isolated hepatocytes. This investigation is an in vitro model for FNZ hepatotoxicity. Further investigations are currently in progress at our laboratory in order to determine FNZ hepatotoxicity in vivo and to look at other parameters of cellular injury.
REFERENCES
Battiston L., Moretti M., Tulissi P., Michili L., Marchi P., Mazzoran L., Lunazzi G. and Pozzato G. (1995) Hepatic glutathione determination after ethanol administration in rat: Evidence of the ®rst pass metabolism of ethanol. Life Sciences 56, 241±248. Baur H., Kasperek S. and Pfa E. (1985) Criteria of viability of isolated liver cells. Physiological Chemistry 356, 827±838. Berry M. N., Halls H. J. and Grivell M. B. (1992) Techniques for pharmacological and toxicological studies with isolated hepatocytes suspension. Life Sciences 51, 1±16. Beutler E., Duron O. and Kelly B. M. (1963) Improved method for the determination of blood glutathione. Journal of Laboratory Clinical Medicine 61, 882±888. Boyer C. S. and Petersen D. R. (1991) Hepatic biochemical changes as result of acute cocaine administration in the mouse. Hepatology 14, 1209±1216. Calhoun C. S., Wesson D. R., Galloway G. P. and Smith D. E. (1996) Abuse of ¯unitrazepam (Rohypnol) and other benzodiazepine in Austin and south Texas. Journal of Psychoactive Drugs 28, 183±189. Colell A., Garcia-Ruiz C., Morales A., Ballesta A., Ookhtens M., Rodes J., Kaplowitz N. and FernandezCheca J. C. (1997) Transport of reduced glutathione in hepatic mitochondria and mitoplasts from ethanol-treated rats: Eects of membrane physical properties and Sadenosyl-l-methionine. Hepatology 26, 699±708. CEWG (1995) Epidemiologic trends in drug abuse-advance report: Fact sheet on Rohypnol. Community Epidemiology Work Group National Institute on Drug Abuse, http:// www.health.org/pubs/factsht/rohypnol.htm. Deleve L. D. and Kaplowitz N. (1991) Glutathione metabolism and its role in hepatotoxicity. Pharmacology and Therapeutics 52, 287±305. Devi B. G., Henderson G. I., Frosto T. A. and Schenker S. (1993) Eect of ethanol on rat fetal hepatocytes: Studies on cell replication, lipid peroxidation and glutathione. Hepatology 18, 648±659. Drummer O. H. and Ranson D. L. (1996) Sudden death and benzodiazepines. American Journal of Forensic Medicine and Pathology 17, 336±342. Drummer O. H., Syrjanen M. L. and Cordner S. M. (1993) Deaths involving the benzodiazepine ¯unitrazepam. American Journal of Forensic Medicine and Pathology 14, 238±243. Figliomeni M. L. and Abdel-Rahman M. S. (1997) The role of ethanol exposure on cocaine metabolism in rat hepatocytes. Journal of Applied Toxicology 17, 105±112. Heyndrickx B. (1987) Fatal intoxication due to ¯unitrazepam. Journal of Analytical Toxicology 11, 278. Kachmer J. R. (1976) Enzymes. In Fundamentals of Clinical Chemistry, ed. N. W. Teitz, pp. 674. W. B. Saunders, Philadelphia. Levine W. G. (1983) Glutathione and hepatic mixed-function oxidase activity. Drug Metabolism Reviews 14, 909± 930. Lieber C. S. (1990) Mechanism of ethanol induced hepatic injury. Pharmacology and Therapeutics 46, 1±41. Masini A., Ceccarelli D., Gallesi D., Giovannini F. and Trenti T. (1994) Lipid hydroperoxide induced mitochondrial dysfunction following acute ethanol intoxication in rats. The critical role for mitochondrial reduced glutathione. Biochemical Pharmacology 47, 217±224. Mattila M. A. K. and Larni H. M. (1980) Flunitrazepam: A review of its pharmacological properties and therapeutic use. Drugs 20, 353±374. Mills E. S. and Jones A. L. (1974) Ultrastructure concepts of drug metabolism. II. The hepatocytes: phenobarbital and microsomal enzyme induction. American Journal of Drug and Alcohol Abuse 1, 271±298.
Hepatotoxicity of ¯unitrazepam and alcohol in vitro NIH (1985) Guide for care and use of laboratory animals. 11±28. NIH Contact No. NOI-RR-2-2135. National Institute of Health, Bethesda, MD. Rej R. (1978) Aspartate aminotransferase activity and isozyme proportion in human liver tissues. Clinical Chemistry 24, 1971±1979. Roots I., Saalfrank K. and Hildebrandt A. G. (1975) Comparison of methods to study enzyme induction in man. Advanced Experimental Medicine and Biology 58, 485±502. Ruch R., Jaber B., Hilton S. and Goertz E. (1994) Oral benzodiazepines safety in overdose: 20 years of experience. Neuropsychopharmacology 10, 227S. Sato C., Matsuda Y. and Lieber C. S. (1981a) Increased hepatotoxicity of acetaminophen after chronic ethanol consumption in the rat. Gastroenterology 80, 140±148. Sato C., Nakano M. and Lieber C. S. (1981b) Prevention of acetaminophen-induced hepatotoxicity by acute ethanol administration in the rat: Comparison with carbon tetrachloride-induced hepatotoxicity. Journal of
401
Pharmacology and Experimental Therapeutics 218, 805± 809. Seglen P. O. (1976) Preparation of rat liver cells. Methods in Cell Biology 13, 29±83. Smith M. T. and Orrenius S. (1984) Drug Metabolism and Drug Toxicity, ed. J. R. Mitchell and M. G. Horning, pp. 71. Raven Press, New York. Story D. L., Gee S. J. and Tyson C. A. (1983) Response of isolated hepatocytes to organic and inorganic cytotoxins. Journal of Toxicology and Environmental Health 11, 443±450. Thompson M. L., Shuster L. and Shaw K. (1979) Cocaine induced hepatic necrosis in mice-the role of cocaine metabolism. Biochemical Journal 225, 565±572. Wendel A. and Feuerstein S. (1981) Drug induced lipid peroxidation in mice: Modulation by monoxygenase activity, glutathione and selenium status. Biochemical Pharmacology 30, 2513±2520. Woods J. H. and Winger G. (1997) Abuse liability of ¯unitrazepam. Journal of Clinical Psychopharmacology 17 (Suppl. 2), 1S-57S.