O6-methylguanine-DNA methyltransferase (MGMT) plays a crucial role in the defense against the alkylating agent-induced cytotoxic lesion O6-alkylguanine in DNA. Although a significant circadian variation in MGMT activity has been found in the liver of mice, the exact mechanism of the variation remains poorly understood. In this study, we present evidence that glucocorticoids were required for the 24-h oscillation of MGMT expression in mouse liver. The exposure of mouse hepatic cells (Hepa1-6) to dexamethasone (DEX) significantly increased the mRNA levels of MGMT in a dose-dependent manner. The DEX-induced increase in MGMT expression was reversed by concomitant treatment with RU486 [11β-[p-(dimethylamino) phenyl]-17β-hydroxy-17-(1-propynyl)estra-4,9-dien-3-one], a glucocorticoid receptor antagonist. The mRNA levels of MGMT and its enzymatic activity in the liver of mice showed significant 24-h oscillations, which were not observed in adrenalectomized mice. A single administration of DEX to adrenalectomized mice significantly increased the mRNA levels of MGMT in the liver. These findings suggest that the 24-h oscillation in the hepatic expression of MGMT is caused by the endogenous rhythm of glucocorticoid secretion. Dacarbazine (DTIC), a potent O6-guanine-alkylating agent, causes serious hepatotoxicity accompanied by hepatocellular necrosis and hepatic vein thrombosis. DTIC-induced hepatotoxicity in mice was attenuated by administering the drug at the time of day when MGMT expression was abundant. The present findings suggest that glucocorticoid-regulated oscillation in the hepatic MGMT expression is the underlying cause of dosing time-dependent changes in DTIC-induced hepatotoxicity.
Cytotoxic alkylating agents, cystemustine, dacarbazine (DTIC), and temozolomide, are highly effective against lymphatic and several types of solid tumor (Marchesi et al., 2007). They damage DNA by forming adducts with bases, the most common of which occur at the N7 position of guanine and the N3 position of adenine, which are repaired by the base excision repair mechanism. On the other hand, another important lesion formed, in terms of mutagenesis and cytotoxicity, is O6-alkylguanine (Marchesi et al., 2007). This adduct can be removed by O6-methylguanine-DNA methyltransferase (MGMT), which thus plays a key role in DNA repair from O6-guanine-alkylating agent exposure (Kaina et al., 2007). In fact, the lethal effect of DTIC in homozygous MGMT knockout mice was more severe than it was in wild-type mice (Shiraishi et al., 2000).
It has been reported that MGMT activity in the liver of mice shows significant circadian variation, and this rhythmicity could account for the dosing time-dependent change in cystemustine-induced toxicity (mortality and myelosuppression) in mice (Martineau-Pivoteau et al., 1996a,b). Time-dependent variations in MGMT activity are also observed in human circulating mononuclear cells (Marchenay et al., 2001), suggesting that the time-dependent change in MGMT activity affects the toxicity of alkylating agents; however, the exact regulation mechanism of circadian oscillation of MGMT activity remains to be clarified.
The effectiveness, toxicity, or both of many drugs vary according to their administration time (Levi and Schibler 2007; Ohdo 2007). Daily variations in biological functions, such as gene expression and protein synthesis, are thought to be important factors affecting the efficacy of drugs. Recent molecular studies of the circadian biological clock system have revealed that oscillations in the transcription of specific clock genes play a central role in the generation of 24-h rhythms (Schibler 2007). In mammals, 24-h rhythms in different tissues are coordinated by a master clock located in the suprachiasmatic nuclei of the anterior hypothalamus. The master circadian clock follows a daily light/dark cycle and, in turn, synchronizes subsidiary oscillators in other brain regions and many peripheral tissues through neural and/or hormonal signals (Kalsbeek et al., 1996; Terazono et al., 2003). These subsidiary oscillators coordinate a variety of biological processes, producing 24-h rhythms in physiology and behavior. Consequently, variations in biological function are associated with dosing time-dependent changes in the efficacy and/or toxicity of many drugs.
The daily rhythm of glucocorticoid secretion from the adrenal cortex is regulated by the hypothalamus-pituitary-adrenal axis, which in turn is controlled by the suprachiasmatic nuclei (Moore and Eichler, 1972). Recent study also suggests that circadian clock genes expressed in adrenal cortex promote the rhythmic steroid generation (Son et al., 2008). As glucocorticoids are involved in the regulation of various physiological functions, such as energy metabolism and immunity function, 24-h changes in circulating glucocorticoid levels also affect the efficacy of many drugs (Koyanagi et al., 2006; Kuramoto et al., 2006). It has been reported that glucocorticoids can induce the expression of the MGMT gene in human cell lines (Marchenay et al., 2001). In this study, we also confirmed the induction of the mouse MGMT gene in response to glucocorticoid stimuli. We thus investigated the role of endogenous glucocorticoid in the regulation of 24-h oscillation of MGMT gene expression in the liver of mice and explored the relationship between the circadian change in MGMT activity and DTIC-induced hepatotoxicity.
Materials and Methods
DTIC was kindly provided by Kyowa Hakko Kirin Co., Ltd. (Tokyo, Japan). The compounds were stored at 4°C and diluted in sterile saline before use. For the treatment of animals, the drug was injected intraperitoneally. Dexamethasone (DEX) was purchased from Wako Pure Chemicals (Osaka, Japan). DEX-phosphate was used instead of free-DEX for solubility reasons. The compound was dissolved in sterilized saline containing 1% ethyl alcohol.
Cell Cultures and Treatments.
Mouse hepatoma (Hepa1–6) cells were purchased from the American Type Culture Collection (Manassas, VA) and cultured in Dulbecco's modified Eagle's medium (Invitrogen, Carlsbad, CA) supplemented with 10% fetal bovine serum and 100 unit/ml penicillin/streptomycin. The cells were routinely grown in monolayer cultures at 37°C in 5% CO2. For each experiment, cells were plated 24 h before treatment so that they were at approximately 80 to 90% confluence at the time of use.
Animals and Treatments.
Male ICR mice (5 weeks old) were purchased from Charles River Japan Inc. (Kanagawa, Japan). They were housed in a light-controlled room (lights on 7:00 AM to 7:00 PM) at room temperature of 24 ± 1°C and humidity of 60 ± 10% with food and water available ad libitum. The animals were treated in accordance with the guidelines stipulated by the animal care and use committee of Kyusyu University.
To investigate the role of the glucocorticoid receptor in the regulation of MGMT expression, Hepa1–6 cells were treated with 5, 10, and 25 nM DEX for 18 h in the presence or absence of RU486 [11β-[p-(dimethylamino)phenyl]-17β-hydroxy-17-(1-propynyl)estra-4,9-dien-3-one], a glucocorticoid receptor antagonist. Sterilized saline containing 1% ethyl alcohol was used as vehicle control. To examine the influence of eliminating glucocorticoid secretion rhythm on the expression of MGMT mRNA and its enzymatic activity, livers were removed from adrenalectomized (Adx) mice at one of six times: 9:00 AM, 1:00 PM, 5:00 PM, 9:00 PM, 1:00 AM, or 5:00 AM. The adrenals were removed via a dorsal approach using an aseptic technique under sodium pentobarbital anesthesia. The adrenalectomized mice were given 0.9% NaCl to drink during the experiment. Sham adrenalectomy was conducted by the same procedure to expose the adrenals without their removal. Seven days after the operation, plasma corticosterone levels in adrenalectomized or sham-operated mice were determined by the enzyme-linked radioimmunoassay method. The adrenalectomized and sham-operated mice were used for each experiment after checking the plasma corticosterone levels. The mRNA levels of MGMT in the liver of adrenalectomized or sham-operated mice were assessed by the reverse transcription-polymerase chain reaction (RT-PCR) method. MGMT activity was determined by detecting the removal of O6-methylguanine by a restriction endonuclease inhibition assay according to the manufacturer's protocol (Sigma-Aldrich, St. Louis, MO). To explore the effect of DEX on mRNA levels of MGMT in the liver, adrenalectomized mice were given an intraperitoneal injection of 3 mg/kg DEX or vehicle. The dosage of DEX was selected based on a previous study (Koyanagi et al., 2006; Kuramoto et al., 2006). Total RNA from the liver was extracted 0, 1, 3, 6, 9, and 12 h after DEX injection. To examine the dosing time dependence of DTIC-induced hepatotoxicity, mice were injected intraperitoneally with 300 mg/kg DTIC or saline on two occasions: 9:00 AM or 9:00 PM. Blood samples were collected at 4 h after DTIC injection and placed in collection tubes containing 20 μl of 4% EDTA solution. The activity of alanine aminotransferase (ALT) in plasma was assessed as an index of DTIC-induced hepatotoxicity. To examine the protective effect of glucocorticoids against DTIC-induced hepatotoxicity, mice were injected intraperitoneally with 3 mg/kg DEX or saline at 5:00 AM. DEX-pretreated mice were also injected with 300 mg/kg DTIC at 9:00 AM. Blood samples were collected at 4 h after DTIC injection, and plasma ALT activity was assessed.
Quantitative RT-PCR Analysis.
Total RNA was extracted by using RNAiso reagent (Takara, Ohtsu, Japan). cDNA was synthesized and amplified by using a RT-PCR system (Takara). The sequence of mouse MGMT primers was as follows: 5′-CCTGTGTTCCAGCAAGATT-3′ and 5′-ACCGTCACTGCGAAC-3′ (GenBank accession no. NM_V008598). The sequence of 18S rRNA primers (internal control) was as follows: 5′-TCAAGAACGAAAGTCGGAGG-3′ and 5′-GGACATCTAAGGGCATCACA-3′ (GenBank accession no. X00686).
Determination of Plasma Corticosterone Concentration.
Plasma samples were obtained by centrifugation at 1500g for 3 min and stored at −20°C until assay. Concentrations of corticosterone in plasma were assessed with a Corticosterone [125I] RIA system (GE Healthcare, Little Chalfont, Buckinghamshire, UK).
Determination of MGMT Activity.
Liver samples were homogenized with lysis buffer [50 mM Tris-HCl (pH 7.5), 1 mM EDTA, 1 mM dithithreitol, 5% glycerol, and 1 mM phenylmethylsulfonyl fluoride]. The lysates were reacted with 0.1 pmol of [γ-32P] dTTP end-labeled, 18-base pair (bp) O6-methylguanine containing DNA substrate at 37°C for 1 h in 150 μl of analysis buffer [50 mM HEPES (pH 7.5), 500 mM KCl, 0.25% Triton X-100, 10 μg/ml bovine serum albumin]. The 23-bp oligonucleotide substrate containing a single methyl lesion at the O6 position of guanine within a PstI site was custom-synthesized (Sigma-Aldrich). The samples were reacted with PstI at 37°C for 3 h, and the reaction was terminated by the addition of 90% formamide loading buffer. The samples were electrophoresed on a 20% denaturing gel. The gel was dried, and DNA oligonucleotides with or without cleavage were visualized by X-ray film. Densitometry analysis was performed by using National Institutes of Health Image software. MGMT activity was expressed as a cleavage index indicating the ratio of cleaved-to-uncleaved oligonucleotide.
Determination of ALT Activity.
ALT activities in the plasma of mice were determined by using the Transaminase C-II test (Wako Pure Chemicals) according to the manufacturer's instructions.
The statistical significance of differences among groups was analyzed by analysis of variance (ANOVA) and post-hoc Bonferroni comparisons. A P value < 0.05 level of probability was considered significant.
Effect of Dexamethasone on the Expression of the MGMT Gene in Cultured Mouse Hepatic Cells.
As reported for human cell lines (Biswas et al., 1999), treatment of Hepa1–6 cells with DEX for 18 h caused a significant increase in MGMT mRNA levels in a concentration-dependent manner. Significant increase in the mRNA levels of MGMT was seen at a concentration of more than 10 nM (Fig. 1A). The induction of MGMT expression by DEX was observed within 6 h after drug exposure (Supplemental Data 1); however, the induction effect of DEX on the expression of MGMT mRNA was reversed by concomitant treatment with RU486, a glucocorticoid receptor antagonist (Fig. 1B). These results suggest that glucocorticoids have the ability to induce MGMT mRNA expression via the glucocorticoid receptor in mouse hepatic cells.
Adrenalectomy Eliminates 24-h Oscillations of mRNA Levels of MGMT in Mouse Liver.
Because the treatment of hepatic cells with glucocorticoids induced the expression of MGMT, we next explored the role of endogenous glucocorticoids in the circadian rhythm of MGMT expression. The time-dependent variations in plasma glucocorticoid (corticosterone) levels were eliminated in adrenalectomized mice (Fig. 2A). We thus used these animals to investigate the influence of the circadian oscillation of circulating glucocorticoid levels on MGMT mRNA expression in mouse liver. The mRNA levels of MGMT in the liver of sham-operated mice showed significant 24-h variation, with higher levels around the early dark phase (Fig. 2B). The amplitude of MGMT mRNA rhythm was decreased in the liver of adrenalectomized mice (Fig. 2C). The levels of MGMT mRNA (at 9:00 PM) in adrenalectomized mice were significantly lower than in sham-operated mice (P < 0.05). Furthermore, significant induction of MGMT mRNA expression was observed within 3 h after intraperitoneal injection of 3 mg/kg DEX (P < 0.05; Fig. 2D). The induction of MGMT mRNA expression by DEX was accompanied by elevation of its enzymatic activity (Fig. 2E). Taken together, these results suggest that the oscillations in the hepatic expression of MGMT are caused by circadian changes in circulating glucocorticoid levels.
Consistent with previous observations (Martineau-Pivoteau et al., 1996a), MGMT activity in the liver of sham-operated mice also showed significant 24-h oscillations (Fig. 3A). The amplitude of the oscillations in MGMT activity was decreased in adrenalectomized mice (Fig. 3B). The activity of MGMT at 9:00 PM was significantly lower than in sham-operated mice (P < 0.05). These results also support the notion that endogenous glucocorticoids positively regulates the hepatic expression of MGMT mRNA and its enzymatic activity.
Influence of Adrenalectomy on the Dosing Time Dependence of DTIC-Induced Hepatotoxicity.
Because hepatic activity of MGMT fluctuated rhythmically in a circadian time-dependent manner, we next investigated whether the rhythmicity in MGMT activity affected DTIC-induced hepatotoxicity. To this end, plasma ALT activity after intraperitoneal injection of 300 mg/kg DTIC was assessed as an index of hepatotoxicity. Blood samples were collected 4 h after DTIC injection because plasma ALT activity peaked at 4 h after DTIC injection and was maintained until 12 h after drug injection (Supplemental Data 2). In sham-operated mice, there was a significant dosing time-dependent difference in plasma ALT activity after DTIC injection. Transaminase activity at 4 h after DTIC injection at 9:00 AM was significantly higher than after saline injection (P < 0.05; Fig. 4); however, plasma ALT activity after DTIC injection at 9:00 PM was not significantly different from that measured after saline injection at the same time point. Because these results suggest that the 24-h change in hepatic MGMT activity modulates DTIC-induced hepatotoxicity, we tried to gain further insight into the mechanism of the dosing time dependence of DTIC-induced hepatotoxicity association with the oscillation of MGMT activity. After intraperitoneal injection of 300 mg/kg DTIC to adrenalectomized mice at 9:00 AM or 9:00 PM, plasma ALT activity significantly increased at both time points as the same as the level in sham-operated mice injected with DTIC at 9:00 AM. However, there was no significant dosing time-dependent difference in transaminase activity between the two dosing times (Fig. 4). These findings indicate that glucocorticoid-regulated oscillations in hepatic MGMT activity are the underlying driver of the dosing time-dependent changes in DTIC-induced hepatotoxicity.
Protective Effect of Glucocorticoids against DTIC-Induced Hepatotoxicity.
Finally, we tested whether glucocorticoids exert a protective effect against DTIC-induced hepatotoxicity. To this end, mice were pretreated with DEX or vehicle at 5:00 AM. Four hours after DEX treatment, mice were injected with 300 mg/kg DTIC or saline (at 9:00 AM). As shown in Fig. 5, injection of DEX alone had little effect on plasma ALT activity. Although a single injection of 300 mg/kg DTIC to DEX-pretreated mice resulted in significant elevation of plasma ALT activity, DTIC-induced elevation of transaminase activity in DEX-pretreated mice was significantly lower than in vehicle-pretreated mice. These results suggest that glucocorticoids have a protective role in DTIC-induced hepatotoxicity.
MGMT is known to remove the mutagenic DNA adduct O6-alkylguanine, which is produced by exposure to alkylating antitumor drugs, including DTIC (Marchesi et al., 2007). In this study, we clarified the role of glucocorticoids in the regulation of MGMT gene expression in mouse hepatic cells. The exposure of Hepa1–6 cells to DEX significantly increased the mRNA levels of MGMT. A single administration of DEX to adrenalectomized mice also caused a significant increase in MGMT mRNA expression in mouse liver. These in vitro and in vivo data suggest that glucocorticoids act as positive regulators of hepatic MGMT expression.
DEX-induced expression of MGMT mRNA was reversed by concomitant treatment with the glucocorticoid receptor antagonist RU486. Glucocorticoids exert their action on gene expression though the activation of cytoplasmic glucocorticoid receptors that bind to glucocorticoid response elements (GREs) on DNA. Two putative GREs have been identified in the human MGMT promoter (Marchenay et al., 2001). The distance between the two GREs is approximately 20 bp, and they are located 1.0 kilobase pair upstream of the putative transcription start site; however, the promoter architecture of the mouse MGMT gene has been poorly characterized to date. Computer-aided analysis of the mouse MGMT promoter identified a GRE-like sequence located approximately 1.5 kilobase pair upstream of the transcription start site, suggesting that it may potentially serve in the regulation of mouse MGMT gene expression by glucocorticoids. Further studies are required to investigate the exact molecular mechanism of MGMT transcription by glucocorticoids.
In nocturnal rodents, plasma corticosterone levels increased from the light phase to the early dark phase and decreased around the early light phase (Koyanagi et al., 2006; Kuramoto et al., 2006). The rhythmic phase of the hepatic expression of MGMT mRNA was similar to those of circulating glucocorticoid levels. Although MGMT oscillations disappeared by adrenalectomy, a single injection of 3 mg/kg DEX to adrenalectomized mice could induce the expression of MGMT mRNA in the liver. DNA microarray analysis has revealed that more than 100 circadian genes expressed in mouse liver are eliminated by adrenalectomy (Oishi et al., 2005). These include the MGMT gene and molecules that are targeted by antihyperlipidemia, immunosuppressant, and antitumor drugs. The time-dependent increase and decrease in circulating glucocorticoid levels may be the underlying driver of the 24-h oscillation of MGMT mRNA expression. We demonstrated previously that the expression patterns of clock genes, such as Clock, Bmal1, Per2, and Cry1, in the livers of adrenalectomized mice did not differ significantly from those in the control (sham-operated) mice (Kuramoto et al., 2006). Taken together, these facts suggest that adrenalectomy can eliminate the 24-h oscillation of MGMT mRNA levels in the liver without affecting the rhythmicity of clock gene expression. It is thus unlikely that the molecular components of the circadian clock are involved in the glucocorticoid-dependent regulation of MGMT gene expression.
Like many chemotherapeutic drugs, DTIC also has a variety of side effects, such as vomiting, headache, fatigue, depilation, and immunosuppression (Frosch et al., 1979). In addition to these side effects, DTIC sometimes causes serious hepatotoxicity accompanied by hepatocellular necrosis and hepatic vein thrombosis (Féaux de Lacroix et al., 1983). DTIC is thought to be inactive until metabolized by cytochrome P450 enzymes to form the reactive N-demethylated species 5-[3-hydroxy-methyl-3-methyl-triazen-1-yl]-imidazole-4-carboxamide(HMMTIC) and 5-[3-methyl-triazen-1-yl]-imidazole-4-carboxamide (MTIC) (Reid et al., 1999). The initial product of P450-catalyzed oxidation of DTIC, the carbinolamine HMMTIC, produces MTIC after elimination of formaldehyde. Rapid decomposition of MTIC yields the major metabolite 5-amino-4-imidazole carboxamide and the reactive species methane diazohydroxide, which is believed to be the methylating species. Although several isoforms of P450 are proposed to metabolite from DTIC to HMMTIC, CYP1A2 is responsible mainly for the metabolism process (Reid et al., 1999). Glucocorticoid has marked influence on the hepatic expression of P450s (Dogra et al., 1998), but the mRNA levels of Cyp1a2 in the liver of mice were not significantly changed by a single injection of 3 mg/kg DEX or vehicle (Supplemental Data 3). Similar findings have also been reported previously (Martignoni et al., 2006). The adrenalectomy-induced modulation of hepatotoxic effect of DTIC is not attributable to the alteration of its metabolism.
Active metabolites of DTIC can induce DNA damage by forming adducts with the O6-position of guanine (Reid et al., 1999), which promote cell death by preventing recombination repair of double-strand breaks. MGMT acts as a protective barrier against cytotoxic DNA damage induced by O6-guanine-alkylating agents (Hansen et al., 2007; Kaina et al., 2007). Consistent with previous observations (Martineau-Pivoteau et al., 1996), hepatic MGMT activity in mice showed significant 24-h oscillations with higher levels during the dark phase. Oscillations in the activity of MGMT may be caused by the rhythmic change in its mRNA expression. The DTIC-induced hepatotoxicity was also attenuated by administering the drug at the time of day when hepatic MGMT activity was increased. These results suggest that 24-h oscillations in MGMT activity are associated with the dosing time dependence of DTIC-induced hepatotoxicity. This notion is also supported by the present findings that the dosing time variation in DTIC-induced hepatotoxicity is attenuated by eliminating the rhythmicity of hepatic MGMT activity.
The present findings in this animal model suggest that toxic side effects of DTIC on normal healthy liver could be attenuated by administering the drug at the time of day when MGMT activity is increased. The rhythmic change in the activity of MGMT seems to be the underlying cause of the dosing time-dependent change in DTIC-induced hepatotoxicity. Time-dependent variations in MGMT activity are also observed in human circulating mononuclear cells (Marchenay et al., 2001), indicating the possibility that varying MGMT activity also affects DTIC toxicity in humans. Our results will provide a clue to understanding how the rhythmic change in hepatic MGMT activity affects the hepatotoxicity induced by the alkylating agent DTIC.
We thank Kyowa Hakko Kirin Co., Ltd. (Tokyo, Japan) for providing the dacarbazine used in this study.
This work was partially supported by a Grant-in-Aid for Scientific Research on Priority Areas “Cancer” from the Ministry of Education, Culture, Sport, Science, and Technology [Grant 20014016] (to S.O.); a Grant-in-Aid for Scientific Research (B) from the Japan Society for the Promotion of Science [Grant 21390047] (to S.O.); a Grant-in-Aid for Challenging Exploratory Research from the Japan Society for the Promotion of Science [Grant 21659041] (to S.O.); and a Grant-in-Aid from the Mochida Memorial Foundation (to S.K.).
Article, publication date, and citation information can be found at http://jpet.aspetjournals.org.
- O6-methylguanine-DNA methyltransferase
- glucocorticoid response element
- alanine aminotransferase
- reverse transcription-polymerase chain reaction
- base pair
- analysis of variance.
- Received January 5, 2010.
- Accepted March 19, 2010.
- Copyright © 2010 by The American Society for Pharmacology and Experimental Therapeutics
- Biswas et al., 1999.↵
- Dogra et al., 1998.↵
- Féaux de Lacroix et al., 1983.↵
- Frosch et al., 1979.↵
- Hansen et al., 2007.↵
- Kalsbeek et al., 1996.↵
- Kaina et al., 2007.↵
- Koyanagi et al., 2006.↵
- Kuramoto et al., 2006.↵
- Levi and Schibler, 2007.↵
- Marchenay et al., 2001.↵
- Marchesi et al., 2007.↵
- Martignoni et al., 2006.↵
- Martineau-Pivoteau et al., 1996a.↵
- Martineau-Pivoteau et al., 1996b.↵
- Moore and Eichler, 1972.↵
- Ohdo, 2007.↵
- Oishi et al., 2005.↵
- Reid et al., 1999.↵
- Son et al., 2008.↵
- Schibler, 2007.↵
- Shiraishi et al., 2000.↵
- Terazono et al., 2003.↵