Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Many physiological and biochemical processes in laboratory animals and humans have been found to vary rhythmically over a 24-h period (i.e., to exhibit a circadian or diurnal rhythm). Rhythms are considered as sequences of biological events, repeating themselves in the same order and at the same intervals. Temporal differences in drug action and toxicity have been recognized for over 50 years in humans and in laboratory animals. A number of cyclic physiological processes can affect the absorption, distribution, metabolism, and elimination of drugs and other chemicals (Labrecque and Belanger, 1991).

The chronotoxicity of solvents and other industrial chemicals has received relatively little attention. Volatile organic compounds (VOCs) are a class of solvents to which many people are exposed occupationally and environmentally. Early studies of chloroform (Lavigne et al., 1983), carbon tetrachloride (CCl₄) (Harris and Anders, 1980; Bruckner et al., 1984), and 1,1-dichloroethylene (Jaeger et al., 1973) revealed circadian rhythms in susceptibility of rats to liver damage by the chemicals. Some of the investigators speculated that periods of enhanced susceptibility might be due to cyclic increases in metabolic activation of the VOCs. Temporal variations in P450-catalyzed metabolism of a variety of substrates by rat liver microsomes have been reported, but correlations of substrate metabolism with P450 levels have generally been poor (Labrecque and Belanger, 1991). Miyazaki et al. (1990) described the time dependence of several P450 isozymes active in catalyzing testosterone hydroxylation in rat liver in a preliminary report. Furukawa et al. (1999) observed high alkoxycoumarin O-dealkylase activities in the liver of rats sacrificed during the animals' dark cycle but relatively stable total P450 levels. It is possible that diurnal rhythms in certain P450 isozymes account for the periodicity in xenobiotic metabolism. Unfortunately, very little information relevant to this postulate is available. The P450 substrates used by Furukawa et al. (1999) had low isoform specificity.

Abstinence from food during the sleep cycle may be linked to circadian rhythmicity in CCl₄ metabolic activation and acute toxicity. Fasting for 16 to 48 h is known to enhance the metabolism and hepatotoxicity of CCl₄ (Harris and Anders, 1980; Nakajima et al., 1982). Fasting has also been found to induce hepatic microsomal cytochrome P450IIE1 (CYP2E1) (Brown et al., 1995). CYP2E1 plays a major role in metabolism of CCl₄ and other low molecular weight VOCs (Guengerich et al., 1991; Zangar et al., 2000). It is possible that levels of CYP2E1 vary rhythmically over a 24-h period and that the isozyme's activity at any given clock time will determine the extent of CCl₄ metabolic activation and hepatotoxicity. Fasting can result in significant lowering of hepatic glutathione (GSH) levels. The GSH redox cycle is thought to play a protective role against CCl₄-induced oxidative stress by metabolism of hydroperoxides and lipid hydroperoxides (Nishida et al., 1996). Fasting is also known to result in increased circulating levels of ketone bodies, including acetone. CYP2E1 plays a major role in acetone catabolism, as demonstrated by a marked elevation in blood acetone in diallyl sulfide-treated rats (Chen et al., 1994). Conversely, acetone is an inducer of CYP2E1 and certain other P450s (Johansson et al., 1988). Thus, it is possible that lack of food intake during animals' sleep cycle may trigger the formation of acetone, which can contribute to CYP2E1 induction sufficiently to increase the production of cytotoxic CCl₄ metabolites.

Objectives of this investigation were to 1) determine whether acute CCl₄ hepatotoxicity and hepatic microsomal CYP2E1 exhibit predictable circadian rhythms in rats maintained on a constant light/dark cycle; 2) test the hypothesis that abstinence from food during sleep results in lipolysis and formation of acetone, which participates in induction of hepatic CYP2E1, resulting in a diurnal rhythm in CCl₄ metabolic activation and acute hepatotoxicity; and 3) assess the potential contribution of circadian rhythms in glutathione and glutathione peroxidase activity to CCl₄ chronotoxicity. Information gained from this study may be helpful in understanding the synchronization of metabolism of CCl₄ and other CYP2E1 substrates.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Chemicals. Analytical-grade CCl₄ was purchased from J. T. Baker Chemical Co. (Philadelphia, PA). ¹⁴CCl₄ with specific activities of 4 and 61 mCi/mmol was supplied by PerkinElmer Life Sciences (Boston, MA) and Amersham Pharmacia Biotech (Arlington Heights, IN), respectively. Carbasorb was purchased from Packard Instrument Co. (Downers Grove, IL). All other chemicals and biochemicals were obtained from Sigma Chemical Co. (St. Louis, MO).

Animals. Male Sprague-Dawley (S-D) rats (250-265 g) were purchased from Harlan Sprague Dawley Inc. (Indianapolis, IN). The animals were housed in their own room in an American Association for the Accreditation of Laboratory Animal Care-approved animal care facility. The experimental protocol was reviewed and approved by an institutional animal care committee. The rats were acclimated to a 12-h light/dark cycle (light 6:00 AM-6:00 PM, i.e., 0600-1800 h) in a temperature (25°C)- and humidity (40%)-controlled room for at least 2 weeks prior to use. Animals were housed in groups of six in stainless steel cages in a ventilated rack. Tap water and food (Purina Rat Chow 5001; Purina, St. Louis, MO) were available ad libitum to some animals. Other groups of rats were fasted for 24 h prior to blood/tissue sampling or CCl₄ dosing in experiments requiring fasting. Food was withheld from all animals for 3 h post dosing, but water was provided at all times.

CCl₄ Dosing and Blood Sampling. The body weight of animals at the time of experimentation ranged from 320 to 360 g. Appropriate amounts of CCl₄ were diluted in corn oil, such that an oral dose of 800 mg of CCl₄/kg b.wt. could be given in a total volume of 1.5 ml/kg. Randomized groups of rats were administered a single dose of 800 mg of CCl₄/kg by gavage, at 2- to 4-h intervals over a 24-h period. A curved, ball-tipped intubation needle was used to give the oral bolus dose. Controls similarly received 1.5 ml/kg of corn oil. The rats were sacrificed 24 h after CCl₄ dosing. Blood was drawn by cardiac puncture, and serum was obtained for assays of enzyme activities.

Serum Enzyme Assays. The activities of three enzymes in serum were monitored as measures of hepatocellular damage by CCl₄. Alanine aminotransferase (ALT), sorbitol dehydrogenase (SDH), and isocitrate dehydrogenase (ICD) activities were measured by standard spectrophotometric techniques. ALT, SDH, and ICD are each expressed as milliunits per milliliter of serum, in which 1 unit of activity is defined as 1 µmol of NAD(H) converted/min/ml of serum.

Acetone Dose-Response Study. The first of a set of experiments was conducted to determine the area under the blood concentration versus time curve (AUC) for each of a series of doses of acetone. An indwelling cannula was surgically implanted into the right carotid artery of rats of 350 to 375 g. The cannula was passed under the skin and exteriorized at the nape of the neck. Food and water were supplied ad libitum during a 24-h recovery period. Each group of the unanesthetized, freely moving animals was given one of the following doses of acetone by corn oil gavage between 9:00 and 10:00 AM: 100, 250, 500, 1000, or 2000 mg/kg. Serial blood samples of 5 to 50 µl were taken at intervals of 2 to 120 min for 8 h post dosing and analyzed for their acetone content as described below.

Acetone Analysis. Blood samples of 5 to 50 µl, taken from an indwelling carotid arterial cannula, were transferred to a 20-ml headspace vial. The vial was immediately capped with a polytetrafluoroethylene-lined septum and tightly crimped. A PerkinElmer model 8500 gas chromatograph (GC) fitted with a flame ionization detector and an HS-101 headspace autosampler was used. Analyses were carried out on a 6' × 1/8" stainless steel column packed with 3% OV-17 that was supplied by Alltech Associates (Deerfield, IL). The temperatures for analysis were as follows: column run isothermal at 75°C, injector at 110°C, and detector at 200°C. Nitrogen was used as the carrier gas at 5 psi. The headspace vials were heated at 110°C in the thermostat-controlled autosampler chamber for 30 min before being vented into the GC. Acetone was well resolved without interfering peaks. Standard solutions of acetone in distilled water were analyzed each day as described above to derive a standard curve.

Liver Assays. The liver nonprotein sulfhydryl (NPSH) content of freshly prepared liver homogenates from undosed rats was measured colorimetrically. Other portions of liver were homogenized in cold 0.02 M Tris-KCl buffer at pH 7.4. Microsomes were prepared by differential centrifugation and stored at 80°C. Glutathione peroxidase (GSHPx) activity in the 105,000g supernatant was assayed according to Lawrence and Burke (1976). The protein concentration and total cytochrome P450 content of hepatic microsomes were determined by conventional spectrophotometric procedures. CYP2E1 activity was estimated by measuring p-nitrophenol (PNP)-hydroxylation. The hydroxylation of PNP to 4-nitrocatechol (4-NC) was determined by the procedure of Koop (1990). A substrate concentration of 100 µM PNP was used in this assay. CYP2E1 activity was expressed as nmol of 4-NC/min/mg of protein.

Covalent Binding Study. Groups of undosed rats served as liver donors for a study of covalent binding of ¹⁴CCl₄ as a function of clock time. Liver samples were taken every 2 to 4 h over a 24-h period. Microsomes were prepared as described previously and diluted with 0.05 M Tris-0.15 M KCl/EDTA buffer (pH 7.4) to a microsomal protein concentration of 2 mg/ml. This preparation was incubated with NADPH (1 mM), MgCl₂ (5 mM), nicotinamide (2.5 mM), and 400 nmol/ml ¹⁴CCl₄ (0.25 µCi/ml). The flasks were capped with latex rubber septa immediately after addition of the ¹⁴CCl₄ and incubated for 20 min at 37°C. The reaction was stopped with 2 ml of ice-cold ethanol, and the samples were centrifuged in the cold at 1800 g for 15 min. The pellet was washed twice with cold ethanol, dissolved in 0.5 N NaOH, neutralized with 1 N HCl, and counted in a Beckman LS-7500 scintillation counter (Beckman Coulter, Inc., Fullerton, CA) after addition of an appropriate scintillation cocktail. Blanks were prepared using boiled microsomes. The results are expressed as pmol of ¹⁴C bound/mg of microsomal protein.

Safety Precautions. Studies were conducted in a biohazard facility to avoid CCl₄ contamination of laboratory personnel and surrounding areas. Preparation of test solutions, dosing, sacrifices, and blood and tissue sampling were performed under a 100% exhaust hood. The rats were maintained in vented cage racks to avoid exposure of each other and personnel. Animal remains and other CCl₄-contaminated materials were sealed in double plastic bags, placed into biohazard barrels, and disposed of by a licensed disposal firm. ¹⁴CCl₄ was used, monitored for, and disposed of in accordance with university policy.

Statistical Analyses. Data were analyzed by one-way analysis of variance. The significance of difference between fed and fasted groups was determined by use of Tukey's least significant differences test. Area under the blood acetone concentration versus time curve from 0 to 8 h (AUC⁸₀) values were calculated according to the linear trapezoidal rule. The significance of apparent acetone dose-dependent differences in blood acetone AUC⁸₀, hepatic CYP2E1 activity, and serum SDH activity were assessed by the Student-Newman-Keuls test using Modstat software by Statistics.Com (Bellingham, WA). The minimum level of statistical significance for these tests was p  0.05. Correlation coefficients between AUC⁸₀ and hepatic CYP2E1 activity and between AUC⁸₀ and serum SDH activity were determined using Prism Version 3.02 software by GraphPad Software (San Diego, CA).

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Toxicity. Rats dosed with 800 mg/kg CCl₄ exhibited a pronounced circadian rhythm in susceptibility to the chemical's hepatotoxic action. The three enzymes used as indices of hepatocellular damage rose and fell in parallel over the 24-h monitoring period (Fig. 1). CCl₄ was most hepatotoxic to fed rats when given at 6:00 PM, the beginning of the animals' active/dark cycle. The chemical was least hepatotoxic during the middle (i.e., 10:00 AM-2:00 PM) of their inactive/light cycle (Fig. 1A). Serum enzyme levels in these subjects rose very rapidly from a minimum at 10:00 AM to 2:00 PM to a maximum at 6:00 PM. These increases were about 3- to 5-fold. SDH, ALT, and ICD levels did not vary significantly during 24 h in undosed rats (data not shown). It is evident when comparing Fig. 1A with 1B that fasted rats were much more sensitive to CCl₄ than fed rats. The diurnal rhythm of cytotoxicity in the fasted animals had a substantially greater amplitude. Maximal hepatocellular injury, as reflected by the serum enzyme levels, however, occurred some 2 h sooner than in the group fed ad libitum. The serum enzyme levels declined more rapidly thereafter in these animals than in their fed counterparts.

View larger version (19K): [in this window] [in a new window]   Fig. 1.   Circadian rhythmicity of susceptibility of rats to CCl4 hepatotoxicity. Groups of rats were dosed at 2- to 4-h intervals over a 24-h period with 800 mg of CCl4/kg b.wt. SDH, ALT, and ICD activities in serum were measured 24 h after dosing. Each point represents the mean ± S.E. for six animals. For some groups, S.E. bars are too small to be visible. The rats were fed ab libitum (A) or fasted for 24 h (B) before CCl4. ICD activity is not included (in A) for the sake of clarity. SDH levels in rats pretreated with 20 mg of DAS/kg p.o. 12 h before CCl4 are shown instead. As illustrated by the bar parallel to the x-axis, all animals were maintained on a 12-h light/dark cycle with light from 6:00 AM to 6:00 PM (0600-1800 h) and dark from 6:00 PM to 6:00 AM (1800-0600 h).

Metabolism. A circadian rhythm in hepatic microsomal total P450 was not apparent in fed or in fasted rats. The P450 levels were comparable in both groups over the 24-h monitoring period (data not shown). Hepatic microsomal CYP2E1 activity, in contrast to total P450 levels, exhibited a definite circadian rhythm (Fig. 2). CYP2E1 activity in fed rats progressively diminished during the animals' inactive cycle, reaching its lowest level at 2:00 PM. There was an increase in CYP2E1 between 2:00 and 6:00 PM of ~35% in the fed animals. This interval was followed by stable CYP2E1 activity for the remainder of the active cycle. In contrast, CYP2E1 diminished rapidly during the active cycle, but rose steadily throughout the inactive cycle in the fasted rats, attaining its maximal level at 6:00 PM, the beginning of the active cycle. There was an increase of ~110% during this period (i.e., 6:00 AM-6:00 PM). The isozyme's activity was significantly higher in the fasted than in the fed rats at all times except 6:00 AM.

View larger version (14K): [in this window] [in a new window]   Fig. 2.   Diurnal variation in CYP2E1 activity in hepatic microsomes of fasted and nonfasted rats. Six rats in each group were sacrificed every 4 h over a 24-h period. CYP2E1 activity was assessed by measuring hydroxylation of p-nitrophenol to 4-NC. Points represent mean ± S.E. All values for the fasted groups are significantly higher than corresponding values for the nonfasted groups at p  0.05, except at 6:00 AM.

View larger version (16K): [in this window] [in a new window]   Fig. 3.   Diurnal variation in extent of covalent binding of 14CCl4 metabolites in vitro to hepatic microsomal proteins of fasted and nonfasted rats. 14CCl4 was incubated with microsomes harvested every 2 to 4 h over a 24-h period from untreated rats. Each point represents the mean ± S.E. for six animals. All values in the fasted group are significantly different from corresponding values in the nonfasted group at p  0.05.

Acetone, CYP2E1, and Potentiation of CCl₄ Toxicity. Levels of acetone in the blood exhibited a unique circadian rhythm (Fig. 4). There was a pronounced peak in the acetone concentration at the beginning of the active cycle in both fed and fasted rats. The blood levels rose abruptly between 2:00 and 6:00 PM and then dropped rapidly during the next 4 h. Blood acetone concentrations were significantly higher throughout the 24-h monitoring period in the fasted animals, particularly at 6:00 PM.

View larger version (14K): [in this window] [in a new window]   Fig. 4.   Circadian rhythm of blood acetone levels in fasted and nonfasted rats. Six rats from each group were sacrificed every 4 h over a 24-h period. Blood acetone concentrations were measured by headspace gas chromatography. Points represent mean ± S.E. S.E. bars are too small to be visible for some groups. All values for the fasted groups are significantly higher than corresponding values for the nonfasted groups at p  0.05.

View larger version (11K): [in this window] [in a new window]   Fig. 5.   A, correlation between area under the blood acetone concentration versus time curves from 0 to 8 h post-acetone dosing (AUC80) and hepatic microsomal CYP2E1 activity. Best linear regression (solid line): y = 0.0074X ± 1.27, r2 = 0.96, p = 0.004, n = 6. B, correlation between AUC80 and serum SDH activity. Best linear regression (solid line): y = 2.696X ± (47.63), r2 = 0.95, p = 0.004, n = 6.

Nonprotein Sulfhydryls. A prominent diurnal rhythm was seen in the NPSH content of the liver of rats fed ad libitum (Fig. 6). Fasted rats displayed a similar rhythm of lower amplitude. NPSH levels were significantly lower in the fasted animals at each measurement period. Hepatic NPSH in the fed rats diminished throughout the 12-h light period and steadily increased during the last 10 h of the dark period. There was a lag time of 2 to 4 h in the diurnal rhythm of the fasted rats. Hepatic GSHPx activity did not exhibit a circadian rhythm nor was there a significant difference between fed and fasted animals (data not shown).

View larger version (14K): [in this window] [in a new window]   Fig. 6.   Diurnal variation in hepatic NPSH levels in fasted and nonfasted rats. Six rats from each group were sacrificed every 2 to 4 h over a 24-h period for measurement of liver NPSH content. Points represent mean ± S.E. For some groups, S.E. bars are too small to be visible. All values for the fasted groups are significantly different from the corresponding values for the nonfasted groups at p  0.05.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

Findings in the current investigation are consistent with the hypothesis that abstinence from food during the inactive cycle results in lipolysis and formation of acetone, which participates in CCl₄ metabolic activation and hepatotoxicity. The susceptibility of male S-D rats to CCl₄ hepatotoxicity varies significantly during a 24-h period. The extent of liver damage upon CCl₄ exposure during the initial part of the animals' active cycle is substantially greater than that caused by the same dose of chemical given during the inactive cycle. The abrupt increase from minimal to maximal susceptibility consistently occurs just prior to and during the first hours of feeding and other nocturnal activities.

Circadian rhythmicity in metabolic activation of CCl₄ is apparently important in the chemical's chronotoxicity in male S-D rats. CCl₄ is metabolized by P450-dependent reductive dehalogenation to a trichloromethyl radical. This moiety can covalently bind to lipids and proteins or react further to form organic free radicals, peroxides, and other reactive cytotoxic metabolites. We found covalent binding of ¹⁴CCl₄ metabolites to hepatic microsomal proteins to be relatively high in rats during their active cycle. Total P450 levels were relatively constant over a 24-h period in the current study. In contrast, CYP2E1 activity exhibited a diurnal rhythm, increasing significantly (by 6:00 PM), the time of maximum ¹⁴CCl₄ covalent binding and toxicity. CYP2E1 is the principal P450 isozyme that catalyzes the metabolism of CCl₄ and other VOCs in rodents and humans (Guengerich et al., 1991; Wong et al., 1998; Zangar et al., 2000). In the current study, pretreatment of rats with DAS, a CYP2E1 inhibitor, negated the rhythmicity of CYP2E1 activity and CCl₄ cytotoxicity, demonstrating the central role of CYP2E1 in this phenomenon.

Circadian variations in CCl₄ metabolism and toxicity appear to be related to restricted food intake during inactive/sleep periods, because fasting is known to potentiate CCl₄ hepatotoxicity. Nakajima et al. (1982), for example, observed a 3.3-fold increase in CCl₄ metabolism and a marked elevation of serum enzymes in male Wistar rats fasted for 24 h before CCl₄ exposure. Brown et al. (1995) reported time-dependent CYP2E1 increases in male F-344 rats fasted for 8, 16, and 24 h. We observed an increase in CYP2E1 activity of ~35% between 2:00 and 6:00 PM and ~50% between 2:00 PM and 6:00 AM in rats fed ad libitum. Fasting, stress, and exercise each stimulate lipolysis to satisfy the body's need for energy. Excess fatty acids can be converted to -hydroxybutyrate (HB) and acetoacetate (AA), the latter undergoing decarboxylation to acetone. Acetone is a potent inducer of CYP2E1 and CCl₄ metabolism (Johansson et al., 1988). Acetone, a substrate for CYP2E1, is believed to induce CYP2E1 primarily by binding to it and thereby protecting it from cAMP-dependent phosphorylation (Ronis et al., 1991). Acetone formation during maximal lipolysis each day likely contributes to diurnal induction of CYP2E1. Mlekusch (1982) and Ahlersova et al. (1985) observed sharp peaks in plasma AA and HB like that we saw for acetone at the onset of the active cycle of rats. We observed that CYP2E1 activity and covalent binding of ¹⁴CCl₄ were also maximal or near-maximal during this time. Clark and Powis (1974) reported a peak in aniline hydroxylase activity within 30 min of i.p. injection of female rats with acetone. Such a rapid increase is compatible with the mechanism of ligand stabilization of CYP2E1.

It was of interest in the current study that blood acetone levels diminished so rapidly after reaching their zenith, but CYP2E1 activity remained high. The acetone bound to CYP2E1, however, is not measured by our analytical GC headspace technique. As noted previously, the binding prolongs the residence time of the CYP2E1. After metabolism of its substrate (i.e., acetone), CYP2E1 is degraded quickly. CYP2E1 has a half-life of 4 to 6 h or less in rat liver (Roberts et al., 1994). Ingested CCl₄ reaches the liver, undergoes metabolic activation, and damages the liver in a matter of minutes (Rao and Recknagel, 1968; Sanzgiri et al., 1997). Thus, there appears to be time during a 24-h period for the necessary sequence of events to occur and to be reversed for subsequent diurnal cycles.

Acetone plays an important role in potentiation of CCl₄ hepatotoxicity in fasted subjects. Sato and Nakajima (1985) found that a 1-day fast enhanced the metabolism of CCl₄ and seven other VOCs in male Wistar rats. If acetone is a mediator of fasting-induced CYP2E1 increase and ensuing CCl₄ metabolic activation/toxicity, the magnitude of each process should vary directly with the duration of fasting. Miller and Yang (1984), indeed, reported reasonably good correlation between the duration of fasting and blood acetone and between acetone and hepatic N-nitrosodimethylamine demethylase activity in male S-D rats. Charbonneau et al. (1986) correlated acetone dose/blood level and CCl₄ hepatotoxicity in male S-D rats but did not examine relationships among acetone, CYP2E1, and CCl₄ toxicity. Our findings of substantially higher blood acetone, hepatic CYP2E1, ¹⁴CCl₄ covalent binding, and serum enzymes in fasted rats rather than in nonfasted rats were anticipated. It was surprising, however, that circadian rhythms in the two sets of animals generally paralleled one another. Although we did not monitor beyond one 24-h fasting period, other researchers have found diurnal rhythms in lipid metabolites to persist during an additional 24 to 48 h of fasting (Ahlersova et al., 1985; Escobar et al., 1998).

Acetone may not be formed in sufficient amounts during the rat's 12-h inactive period to account solely for diurnal CYP2E1 induction. Concentrations of AA and HB are typically higher than acetone levels, so these two ketone bodies could augment CYP2E1 induction by acetone. Results of limited in vivo (Barnett et al., 1992) and in vitro (Zangar and Novak, 1998) studies indicate that neither HB nor AA induces CYP2E1. Johansson et al. (1988) demonstrated that starvation for 3 days exerted a synergistic effect with acetone on induction of CYP2E1 and CCl₄ metabolism. These researchers believed that this synergistic effect of fasting could be the result of an increased rate of CYP2E1 gene transcription or mRNA stabilization, possibly due to action of stress hormones. Plasma levels of a number of hormones, including insulin and corticosterone (Mlekusch, 1982), exhibit circadian rhythms with peaks that coincide in time with the maximum CYP2E1 activity and CCl₄ toxicity we observed. Corticosterone and insulin are active in lipolysis and ketogenesis, although they and other hormones may also be involved at the level of the gene in regulation of CYP2E1. Growth hormone plays a major modulatory role by suppressing CYP2E1 gene transcription in S-D rat liver (Chen et al., 1999). It is not clear whether growth hormone participates in the CYP2E1 diurnal cycle that we observed, because growth hormone secretion in male rats is pulsatile (i.e., every 3-4 h), and apoprotein synthesis requires a period of hours.

Certain factors in addition to CYP2E1 activity may contribute to the diurnal rhythm in CCl₄ hepatotoxicity. Liver blood flow is somewhat higher during periods of physical activity. The net effect of a higher blood flow on CCl₄ metabolism should be modest, however, due to the relatively high dose we administered and the slow metabolism of CCl₄ (Gargas et al., 1986). In the current study, there was a pronounced circadian rhythm in liver NPSH (i.e., primarily GSH), with levels lowest at the time of maximal susceptibility to CCl₄ cytotoxicity. A similar phenomenon has been reported for chloroform (Lavigne et al., 1983), acetaminophen (Schnell et al., 1983), and 1,1-dichloroethylene (Jaeger et al., 1973). Although GSH conjugation plays a major role in detoxification of the latter two compounds, this is not the case for CCl₄. The formation of a CCl₄-derived free radical adduct with GSH has been demonstrated in vitro (Connor et al., 1990), but this interaction is not believed to be quantitatively important in vivo. GSH does impede CCl₄-induced lipid peroxidation by GSHPx-catalyzed metabolism of hydrogen peroxide and lipid peroxides (Harisch and Meyer, 1985; Nishida et al., 1996). The latter group of investigators found that the GSH redox cycle protected against early stages of hepatocellular injury and lipoperoxidation but not against more pronounced changes 12 to 24 h after i.p. injection of mice with 1.6 g of CCl₄/kg. Harris and Anders (1980) have demonstrated that marked (i.e., 83%) depletion of liver GSH by diethyl maleate potentiated CCl₄ hepatotoxicity in S-D rats. The diurnal NPSH decreases we observed (i.e., ~30%) were less pronounced. Mitchell et al. (1985) observed marked enzyme leakage from primary cultures of rat hepatocytes when GSH was reduced below a threshold of 20% of control, irrespective of the chemical used to deplete GSH. Harris and Anders (1980) noted that fasted rats exhibited liver injury that was twice as severe as their diethyl maleate-pretreated counterparts, indicative of involvement of other factors. Our own results demonstrate that CYP2E1 induction is an important factor in fasting-elicited and in diurnal changes in susceptibility of rats to CCl₄. The relatively modest decrease in liver NPSH that we saw would appear to be a less important diurnal variable than CYP2E1 induction, although it is not possible to determine their relative contribution from findings in the current investigation.

Findings with rats in the current investigation may be directly relevant to humans. Chronological variations in the pharmacokinetics of nicotine (Gries et al., 1996), ethanol (Sturtevant et al., 1978), aminopyrine (Poley et al., 1978), and a variety of other well metabolized drugs have been reported in human subjects. Interestingly, maximum clearance rates are usually observed during the first part of the subjects' active cycle. Ketone bodies in the blood exhibit diurnal variations and exhalation of acetone peaks before breakfast (Wildenhoff, 1972). Thus, people may be more susceptible to CCl₄ and other chemicals that undergo CYP2E1-mediated activation when exposed to the chemicals during their initial waking hours.

Information gained from the current study of the circadian rhythmicity of CCl₄ hepatotoxicity may have a number of applications. The phenomenon may well apply to a variety of VOCs and other lipophilic, low molecular weight compounds that are metabolically activated by CYP2E1 to cytotoxic or mutagenic metabolites. The toxic and carcinogenic potency of such chemicals could vary with the (clock) time of exposure. Thus, such chemicals should be evaluated at the time(s) of maximum sensitivity of the test animal. Light/dark cycles and treatment times should be considered in designing experimental protocols. P450 isozymes other than CYP2E1 may also be found to exhibit diurnal rhythms.