Influence of Gender on Cocaine Hepatotoxicity in CF-1 Mice

Abstract

Gender is known to play a role in the bioavailability, metabolism, and lethality of many toxic substances. This study was conducted to investigate the influence of gender on cocaine hepatotoxicity (CH) and lipopolysaccharide (LPS) potentiation of CH. Male and female CF-1 mice were orally administered 20 mg/kg body weight cocaine hydrochloride once daily for 7 days. Four hours after the last cocaine administration, the mice were administered 12 × 10⁶ EU LPS (or equal volume of sterile saline) intraperitoneally. Plasma alanine aminotransferase (ALT) and aspartate aminotransferase (AST) were evaluated as indices of liver injury. Blood and liver glutathione (GSH), glutathione reductase (GRx), and catalase (CAT) activities were also determined to investigate the extent of oxidative stress induced by the treatments. Serum ALT and AST concentrations were elevated in all males receiving cocaine alone or cocaine + LPS. Furthermore, blood GSH and CAT were decreased and GRx activity was elevated in these same animals. Histological analysis revealed a high degree of hepatic focal necrosis in the male cocaine group, and severe hemorrhagic necrosis in the male cocaine + LPS group. Unlike males, females showed no damage resulting from cocaine or cocaine + LPS exposure, whereas testosterone-supplemented ovariectomized females displayed histological and biochemical profiles statistically similar to males. The results demonstrate that the extent of CH or LPS-potentiated CH is influenced by gender and sex hormones, particularly testosterone.

Cocaine is one of the most potent naturally occurring stimulants, and its nonmedicinal use has taken place for centuries (Evans 1981; Karch 2001). Today, cocaine abuse remains an extensive problem in the United States, with an estimated 15.4% of 18- to 25-year-old adults using cocaine at least once in their lifetime. Since 1975, males have generally comprised the majority of cocaine users. However, in 2001, there were 0.5 million new female users compared to 0.7 million new male users, indicating that the gender use gap is decreasing (Substance Abuse and Mental Health Services Administration [SAMHSA] 2002).

The metabolism pathways for cocaine, as well as peak plasma concentrations depend on route of administration (Cone 1995; Karch 2001; Schuckit 2000). Oral abuse is a route often used that results in many of the same effects as other exposure routes, and produces similar peak plasma concentrations as intranasal administration (Van Dyke et al. 1978; Wilkinson et al. 1980). Oral abuse, however, dramatically increases the liver’s exposure to the drug by way of the first-pass effect. Therefore, a greater portion of an oral cocaine dose is subject to oxidative metabolism by liver cytochrome P450 enzymes versus hydrolysis by esterases in the blood that occurs for other routes of exposure (Ma, Falk, and Lau 1999). Hydrolysis yields pharmacologically inactive and nonhepatotoxic metabolites (Karch 2001). On the other hand, sequential oxidation of cocaine by P450 enzymes produces hepatotoxic metabolites and reactive oxygen species (ROS) that irreversibly bind to essential hepatocellular proteins, thereby preventing them from carrying out their biological functions (Bornheim 1998; Woods, Plessinger, and Fantel 1998). Antioxidant defense mechanisms such as glutathione (GSH) and catalase (CAT) combat this stress. However, when the rate at which ROS produced by oxidation of cocaine exceeds the capacity of the cells to remove them, oxidative stress occurs.

Moreover, cocaine users are at great risk of exposure to bacterial infection for a number of reasons, including nutritional deficiencies and the vasoconstrictive effect of cocaine upon bowel vasculature. These conditions significantly increase the translocation of gram-negative bacteria from the gastrointestinal (GI) tract to the systemic circulation (Ganey and Roth 2001; Van Leeuwen et al. 1994). In particular, lipopolysaccharide (LPS), a component of the gram-negative bacterial cell wall, is a potent inducer of hepatic inflammation in both human and experimental animals (Cadenas and Cadenas 2001; Champion et al. 1976; Moulin et al. 2001). The presence of harmful cocaine metabolites coincident with LPS-induced hepatic inflammation dramatically increases the hepatotoxicity of cocaine (Labib, Turkall, and Abdel-Rahman 2002; Roth et al. 1997).

Previous studies from our laboratory show that estrogen and testosterone have a major role in the bioavailability of drugs and chemicals (McCormick and Abdel-Rahman 1991; Dahab and Abdel-Rahman 1995; Abou-Hadeed et al. 1998). In addition, recent studies performed in our laboratory show a strong gender influence on cocaine hepatotoxicity (CH) (Visalli, Turkall, and Abdel-Rahman 2004). Previous studies in other laboratories confirm this gender difference utilizing other routes of cocaine administration (Boyer, Ross, and Petersen 1988; Thompson, Shuster, and Casey 1984). To the best of our knowledge, this is the first study to utilize the oral route of cocaine administration to explore sex hormone involvement in the mechanism of gender difference seen in CH with and without LPS exposure.

MATERIALS AND METHODS

Animals

Adult male and female CF-1 mice (Charles River Laboratories, Wilmington, MA), weighing 25 to 30 g and 20 to 25 g, respectively, were used in these studies. Cocaine-induced liver necrosis has been well documented in this strain (Mehanny and Abdel-Rahman 1991; Bornheim 1998; Labib, Turkall, and Abdel-Rahman 2002, 2003b). The animals were quarantined for 1 week before the initiation of the experiment. They were housed in plastic cages in environmentally controlled rooms with a 12-h light/dark cycle at a constant temperature (23°C to 25°C) and a relative humidity of 50%. The animals were maintained on Rodent Diet 5001 (Lab Diet, St. Louis, MO) and water ad libitum. The use of animals for this experiment was approved by the Institutional Animal Care and Use Committee (IACUC) of the University of Medicine and Dentistry of New Jersey, Newark, NJ.

Reagents and Chemicals

Cocaine hydrochloride was provided by the National Institute on Drug Abuse (NIDA), Bethesda, MD. LPS derived from Salmonella enteriditis with an activity of 3 × 10⁶ EU/mg and testosterone were purchased from Sigma Chemical (St. Louis, MO). Drugs were dissolved in physiological saline prior to administration. Testosterone was dissolved in peanut oil. Administration volume was constant at 0.1 ml. The reagent kits used for measuring serum markers of liver injury (alanine aminotransferase [ALT] and aspartate aminotransferase [AST]) were purchased from ThermoDMA (Louisville, CO). Testosterone and estrogen test kits were obtained from and Cayman Chemical (Ann Arbor, MI). Unless otherwise stated, all the chemicals were reagent grade and were purchased from Sigma Chemical.

Ovariectomy/Hormone Administration

Female mice were bilaterally ovariectomized and allowed to recover for 10 days. On the 10th day post surgery, animals received daily subcutaneous injections of 500 μg testosterone or sterile peanut oil based on Barnes and Eltherington (1973) and Dickerson et al. (1991). Hormone plasma concentrations were monitored on the 20th day post surgery.

Treatment Protocol

Mice were randomly assigned to one of four treatment groups (four to five animals per group), with a male group, a female group, an ovariectomized group, and an ovariectomized + testosterone group assigned to each treatment:

Treatment 1: Saline (0.1 ml, per os [p.o.])

Treatment 2: Saline (0.1 ml, p.o.) followed by LPS (12 × 10⁶ EU, intraperitoneal [i.p.])

Treatment 3: Cocaine only (20 mg/kg, p.o.)

Treatment 4: Cocaine (20 mg/kg, p.o.) followed by LPS (12 × 10⁶ EU, i.p.) 4 hours after the last cocaine administration

Mice were food fasted 12 h before all treatments. Cocaine hydrochloride (20 mg/kg) and LPS (12 × 10⁶ EU) were prepared fresh daily in saline. Mice in treatment group 1 were administered 0.9% saline daily for 7 consecutive days by oral gavage. The mice in treatment group 2 were administered LPS 4 h after the last saline treatment. Mice from treatments 3 and 4 were administered cocaine hydrochloride by gavage once daily for 7 consecutive days. In treatment 4, 4 h after the last cocaine treatment, the mice were intraperitoneally administered 12 × 10⁶ EU LPS. Mice from groups 3 and 4 were treated by gavage once daily for 7 consecutive days with 20 mg cocaine hydrochloride/kg body weight. In groups 2 and 4, 4 h after the last cocaine or saline treatment, the mice were intraperitoneally administered 12 × 10⁶ EU LPS/kg. The animals were then anesthetized and sacrificed by decapitation 18 h after LPS or saline administration. Selection of cocaine and LPS dosages were based on previously published data from our laboratory (Labib, Turkall, and Abdel-Rahman 2001; Visalli, Turkall, and Abdel-Rahman 2004).

Determination of Hepatotoxicity

Liver cell integrity was monitored by measuring the release of ALT and AST into plasma. Plasma samples for the transaminase assays were obtained upon sacrifice. Whole blood was collected in heparinized tubes and centrifuged for 10 min at 1000 × g. Results were expressed as international units per liter (IU/L).

Histological Evaluation

The livers were immediately excised upon sacrifice, washed in ice-cold saline, and fixed in neutral-buffered 10% formalin for at least 24 h. The liver was then trimmed, processed, embedded in paraffin, sectioned at 4 to 6 μm, and stained with hematoxylin and eosin (H&E) for light microscopic evaluation.

Measurement of CAT, GSH, and GSH-Related Enzymes

Indices of oxidative stress (GSH, CAT, glutathione reductase [GRx]) were measured. Whole blood and liver samples were obtained from mice upon sacrifice. Liver samples were removed and immediately washed in ice-cold saline and homogenized in sodium phosphate buffer (pH = 7.4) containing 1% Triton X-100. Homogenates were centrifuged at 10, 000 × g for 10 minutes at 4°C. Blood hemolysate was prepared by washing the blood in ice-cold physiologic saline (pH = 7.4) and lysing by resuspension of cells in distilled water.

GSH was determined in whole blood and in liver homogenate according to the method of Beutler, Duran, and Kelly (1963). The activities of related antioxidant enzymes, GRx and CAT, were measured in liver supernatant and blood hemolysate. CAT activity was measured according to the method of Aebi (1981), by monitoring the decomposition of H₂O₂ spectrophotometrically at 240 nm. Enzyme activity was expressed per milligram of hemoglobin and milligram of protein for blood and liver samples, respectively. GRx activity in liver homogenate was assayed according to the method of Couri and Abdel-Rahman (1980). This assay is based on the reduction of oxidized GSH, coupled with oxidation of NADPH. The unit activity of GRx was defined as an increase in log of GSH absorbance at 340 nm versus time. Protein was determined with serum albumin as a standard (Bradford 1976) and total hemoglobin was measured in whole blood using a method by Tietz (1987).

Statistical Analysis

Data were represented as the mean ± SEM. Multiple comparisons were performed by analysis of variance (ANOVA) followed by the Tukey-Kramer honestly significant difference (HSD) test. Statistical analysis between two groups was performed by Student’s independent t test. In all analyses, the level of significance was set to p < .05. Statistical analyses were performed with the JMP 4.0.4 statistical software package (SAS Institute, Cary, NC).

RESULTS

Hepatotoxicity Assessment

Male and female animals showed no elevation in serum transaminases after treatment with saline or LPS alone (Figure 1A, B ), and the livers from these mice had no histological evidence of injury (Figure 2A, B ). The architecture of the liver was normal for both saline and LPS treatments, with a slight increase in phagocytic cell infiltration in animals treated with LPS (Figure 2C, D ). Male animals treated with cocaine alone consistently exhibited localized areas of centrilobular, zone 3 coagulative necrosis (Figure 3A ), with a 15-fold and 2.5-fold increase in ALT and AST versus surgically unaltered females, respectively (Figure 1A, B ). Males receiving cocaine + LPS displayed severe hemorrhagic, submassive necrosis with complete loss of cell boundaries (Figure 4A ). Necrotic areas spanned central vein to central vein in many cases. ALT and AST activity increased three-fold and nearly two-fold versus surgically unaltered females, respectively (Figure 1A, B ). A statistically similar biochemical and histological profile to males was seen in testosterone-supplemented ovariectomized females (Figures 1A, B ; 3D ; 4D ). Hepatocytes from surgically unaltered females and ovariectomized females receiving cocaine or cocaine + LPS displayed minor variation in nuclear size (Figures 3B, C ; 4B, C ), but no structural liver damage. An increase of less than twofold in ALT and AST was apparent in ovariectomized females (Figure 1A, B ) versus surgically unaltered females. A small amount of phagocytic cell infiltration was present in unaltered and ovariectomized females treated with cocaine + LPS, consistent with that produced by exposure to LPS alone (Figure 4B, C, D ).

GSH and Antioxidant Enzymes

Animals administered saline or LPS only showed no change in blood and liver GSH or GRx (Figures 5, 6). Cocaine and cocaine + LPS treatments caused blood GSH concentration to decrease significantly by approximately 30% in male animals versus saline treatment (Figure 5A ). Further, hepatic GSH concentration doubled for males receiving cocaine only compared to saline treatment (Figure 5B ). A similar pattern was observed in hepatic GRx activity, which nearly doubled in males receiving cocaine only compared to saline (Figure 6). Statistically similar results in these same biochemical indices were obtained in testosterone-supplemented ovariectomized females. However, surgically unaltered females did not exhibit significant changes in either of the above indices after cocaine or cocaine + LPS treatment (Figures 5, 6). Ovariectomized females showed a small but significant decrease in blood GSH (Figure 5A ) and a corresponding increase in GRx (Figure 6) after cocaine and cocaine + LPS treatment versus surgically unaltered females.

In addition, cocaine and cocaine + LPS treatment reduced blood and liver CAT concentrations by 30% to 35% and 55% to 60%, respectively, in male and testosterone-supplemented ovariectomized females, whereas LPS alone caused a significant but only slight decrease in blood and liver CAT in all animals (Figure 7A, B ). Cocaine + LPS treatment in surgically unaltered females caused a small decrease in blood and hepatic CAT consistent with that produced by LPS exposure alone (Figure 7A, B ). Blood and liver CAT activities of females receiving cocaine alone were not significantly different from saline treatment (Figure 7A, B ), whereas ovariectomized females showed a small but significant decrease in blood and hepatic CAT after exposure to cocaine alone.

DISCUSSION

Previously, our laboratory reported that CH with and without LPS exposure is influenced by gender in CF-1 mice (Visalli, Turkall, and Abdel-Rahman 2004). The results of the current study exemplify the role of sex hormones in the underlying mechanism of this gender difference. In particular, surgically unaltered female mice and ovariectomized females lack the severe histological changes observed in the livers of male mice after seven day cocaine or cocaine + LPS treatment. Only a minor variation in the hepatocyte nuclear size indicative of oxidative stress was observed in both female treatment groups. Further, ovariectomized females showed significant increases in serum ALT and AST activities compared to unaltered females. However, these values were much lower than that of males. This suggests that estrogen plays a minor role in the observed gender difference.

On the other hand, testosterone supplemented ovariectomized females treated with cocaine or cocaine + LPS exhibited a his-tological and biochemical pattern of CH similar to male mice. Specifically, cocaine treatment produced a statistically equivalent elevation of serum ALT and AST as well as an identical pattern of focal, zone 3 necrosis in both males and testosterone supplemented ovariectomized females. Further, in these same groups cocaine + LPS treatment further exacerbated CH as evidenced by the production of severe hemorrhagic liver necrosis. The slight increase in serum ALT and AST seen with cocaine + LPS treatment is consistent with severely reduced amounts of viable tissue no longer able to contribute these enzymes to the blood (Speicher 1993; Wallach 1992). The similarity of treatment results between testosterone-supplemented ovariectomized females and males suggests that testosterone plays a major role in the observed gender differences.

The role of testosterone in gender differences seen in CH and CH potentiated by LPS is supported by gender-dependent differences in blood esterase and hepatic cytochrome P450 activity seen in both humans and experimental animals (Davis 1998; Illsley and Lamartiniere 1981; Smolen and Smolen 1990). In particular, testosterone has been shown to inhibit blood esterase activity (Illsley and Lamartiniere 1981), possibly leaving a greater portion of the cocaine dose available for P450 oxidation in male mice. Further, male mice have been shown to develop CH only after 30 days of age (corresponding with the onset of sexual maturity), suggesting that the enzyme(s) responsible for toxic metabolite production are absent or at very low concentrations in female and immature male mice (Smolen and Smolen 1990).

Specifically, significant gender-related differences in norcocaine tissue/plasma concentration ratios following cocaine administration have been observed in male and female mice and rats. Norcocaine is a P450 oxidative metabolite of cocaine that is thought, in part, to produce hepatotoxicity by binding to critical molecules. Hepatic tissue/plasma ratios of norcocaine have been reported to be three-fold greater in males than females (Whittington et al. 1999). This difference in cocaine metabolite binding in hepatic tissue between males and females is consistent with the gender-dependent difference in CH and LPS potentiation seen in this study.

Recent studies in our laboratory show that hepatocellular damage resulting from LPS and cocaine is mitigated by a depletion of antioxidant systems due to the presence of excessive ROS (Labib, Turkall, and Abdel-Rahman 2003a, 2003b). In this study, blood GSH concentration was decreased only in male and testosterone-supplemented ovariectomized females after cocaine or cocaine + LPS administration. This decrease contrasts with increases in hepatic GSH concentration in these same animals, suggesting enhancement of pathways to replenish depleted GSH. GSH depletion is likely the result of oxidative stress introduced by prolonged cocaine exposure with production of oxidative metabolites as well as ROS produced by cocaine metabolism and LPS (Labib, Turkall, and Abdel-Rahman 2002). Elevated GRx activity in males and testosterone-supplemented ovariectomized females supports this concept, but does not rule out increased synthesis. The relatively smaller increases in hepatic GSH concentration and GRx activity in males and testosterone-supplemented, ovariectomized females treated with cocaine + LPS versus cocaine alone is consistent with a smaller number of viable liver cells remaining following cocaine + LPS treatment. Blood GSH concentrations in ovariectomized females were slightly decreased compared to surgically unaltered females in the same treatment group, whereas hepatic GSH was not significantly different from saline treatment. This small depletion of blood GSH in the absence of corresponding liver damage is consistent with minor oxidative stress that does not necessitate an up-regulation of pathways to replenish GSH. Surgically unaltered females treated with cocaine or cocaine + LPS showed no change in blood GSH and hepatic GSH or GRx versus females receiving saline treatment.

CAT activity was decreased in the blood of male mice receiving saline treatment and cocaine + LPS. This is in agreement with studies performed in male experimental animals where down-regulation of CAT genes occurs due to cocaine or LPS exposure (Diez-Fernandez et al. 1999: Labib, Turkall, and Abdel-Rahman 2001; Sakaguchi et al. 1981, 1996; Zaragoza et al. 2000). A similar decrease in CAT activity was observed in testosterone-supplemented ovariectomized females receiving the same treatments. On the other hand, unaltered females did not show a decrease in CAT after cocaine treatment; lack of cocaine-induced hepatotoxicity in females may be partly explained by this. Unaltered females, however, did show a significant decrease in blood CAT after LPS and cocaine + LPS treatments. The decrease in CAT activity after cocaine + LPS treatment was of lower magnitude than that seen in male mice and is likely the effect of LPS, because cocaine alone had no effect on CAT activity. After LPS and cocaine + LPS exposure, ovariectomized females showed a decrease in blood CAT corresponding to other groups, but also a slight decrease in blood CAT after exposure to cocaine alone. This further supports response to oxidative stress. Mice exhibited a similar pattern of decreased CAT activity in blood as was exhibited in liver.

The results of this study demonstrate that sex hormones, mainly testosterone, are an underlying factor in the mechanism of gender difference in response to cocaine or cocaine + LPS treatment in CF-1 mice.