Search

Long-term administration of some antiepileptic drugs often increases blood lipid levels. In this study, we investigated its molecular mechanism by focusing on the nuclear receptors constitutive active/androstane receptor (CAR) and peroxisome proliferator–activated receptor α(PPARα), which are key transcription factors for enzyme induction and lipid metabolism, respectively, in the liver. Treatment of mice with the CAR activator phenobarbital, an antiepileptic drug, increased plasma triglyceride levels and decreased the hepatic expression of PPARαtarget genes related to lipid metabolism. The increase in PPARαtarget gene expression induced by fenofibrate, a PPARαligand, was inhibited by cotreatment with phenobarbital. CAR suppressed PPARα-dependent gene transcription in HepG2 cells but not in COS-1 cells. The mRNA level of peroxisome proliferator–activated receptor γcoactivator 1α(PGC1α), a coactivator for both CAR and PPARα, in COS-1 cells was much lower than in HepG2 cells. In reporter assays with COS-1 cells overexpressing PGC1α, CAR suppressed PPARα-dependent gene transcription, depending on the coactivator-binding motif. In mammalian two-hybrid assays, CAR attenuated the interaction between PGC1αand PPARα. Chemical inhibition of PGC1αprevented phenobarbital-dependent increases in plasma triglyceride levels and the inhibition of PPARαtarget gene expression. These results suggest that CAR inhibits the interaction between PPARαand PGC1α, attenuating PPARα-dependent lipid metabolism. This might explain the antiepileptic drug–induced elevation of blood triglyceride levels.SIGNIFICANCE STATEMENTConstitutive active/androstane receptor activated by antiepileptic drugs inhibits the peroxisome proliferator–activated receptor α–dependent transcription of genes related to lipid metabolism and upregulates blood triglyceride levels. The molecular mechanism of this inhibition involves competition between these nuclear receptors for coactivator peroxisome proliferator–activated receptor γcoactivator-1αbinding.

Introduction:Glutathione S-transferases (GSTs) are Phase II drug-metabolizing enzymes that catalyze the conjugation of electrophilic compounds to glutathione. This reaction generally detoxifies reactive metabolites of xenobiotics, such as drugs and environmental chemicals, and therefore, GSTs are considered toxicologically important enzymes. Human GSTgenes display genetic polymorphisms, and the impact of null genotypes of GST Mu 1(GSTM1) and GST Theta 1(GSTT1) have been extensively studied. Recently, corresponding Gstm1- and Gstt1-null mice have been developed.Areas covered:This review covers GSTM1- and GSTT1-null genotypes in humans and utilization of Gstm1- and Gstt1-null mice as animal models for absorption, distribution, metabolism, excretion and toxicity (ADMET) studies. Furthermore, the possibility of Gstm1- and Gstt1-null mice as models of humans with GSTM1- and GSTT1-null genotypes is discussed.Expert opinion:The experimental approach to assess the impact of GSTM1- and GSTT1- null genotypes on ADMET is very limited so far. Gstm1- and Gstt1-null mice have potential as human models, since null genotypes of GSTM1/Gstm1and GSTT1/Gstt1decreased hepatic GST activities toward p-nitrobenzyl chloride and dichloromethane, respectively, in both humans and mice. This suggests functional similarity of GSTM1 and GSTT1 toward some substrates between humans and mice.

We investigated the impact of glutathione transferases Mu 1 (GSTM1)- and glutathione transferase Theta 1 (GSTT1)-null genotypes on hepatic GST activities in humans and compared the results with those of Gstm1- and Gstt1-null mice. In liver with GSTM1/Gstm1-null genotype, GST activity toward p-nitrobenzyl chloride (NBC) was significantly decreased in both humans and mice. In addition, in liver with GSTT1/Gstt1-null genotype, GST activity toward dichloromethane (DCM) was significantly decreased in both humans and mice. Therefore, null genotypes of GSTM1/Gstm1 and GSTT1/Gstt1 are considered to decrease hepatic GST activities toward NBC and DCM, respectively, in both humans and mice. This observation shows the functional similarity between humans and mice for GSTM1 and GSTT1 toward some substrates. In the case of NBC and DCM, Gst-null mice would be relevant models for humans with GST-null genotype. In addition, decreases in GST activities toward 1,2-dichloro-4-nitrobenzene, trans-4-phenyl-3-buten-2-one, and 1-chloro-2,4,-dinitrobenzene were observed in Gstm1-null mice, and a decrease in GST activity toward 1,2-epoxy-3-(p-nitrophenoxy)propane was observed in Gstt1-null mice. However, an impact of GST-null genotypes on GST activities toward these substrates was not observed in humans. In the case of these mouse-specific substrates, Gst-null mice may be relevant models for humans regardless of GST genotype, because GST activities, which are higher in wild-type mice than in humans, were eliminated in Gst-null mice. This study shows that comparison of hepatic GST activities between humans and mice using genotype information would be valuable in using Gst-null mice as human models.

Glutathione S-transferase (GST) theta 1 (GSTT1) has been regarded as one of the key enzymes involved in phase II reactions because of its unique substrate specificity. In this study, we generated mice with the disrupted Gstt1 gene (Gstt1-null mice) by gene targeting and analyzed the metabolic properties in cytosolic and in vivo studies. The resulting Gstt1-null mice failed to express the Gstt1 mRNA and GSTT1 protein by reverse transcriptase-polymerase chain reaction analysis and two-dimensional fluorescence difference gel electrophoresis/mass spectrometry analysis, respectively. However, the Gstt1-null mice appeared to be normal and were fertile. In an enzymatic study using cytosolic samples from the liver and kidney, GST activity toward 1,2-epoxy-3-(p-nitrophenoxy)propane (EPNP), dichloromethane (DCM), and 1,3-bis(2-chloroethyl)-1-nitrosourea (BCNU) was markedly lower in Gstt1-null mice than in the wild-type controls, despite there being no difference in GST activity toward 1-choloro-2,4-dinitrobenzene between Gstt1-null mice and the wild-type controls. Gstt1-null mice had GST activity of only 8.7 to 42.1% of the wild-type controls to EPNP, less than 2.2% of the wild-type controls to DCM, and 13.2 to 23.9% of the wild-type controls to BCNU. Plasma BCNU concentrations after a single i.p. administration of BCNU to Gstt1-null mice were significantly higher, and there was a larger area under the curve5–60min (male, 2.30 times; female, 2.28 times, versus the wild-type controls) based on the results. In conclusion, Gstt1-null mice would be useful as an animal model of humans with the GSTT1-null genotype.

Glutathione S-transferase Mu 1 (GSTM1) has been regarded as one of the key enzymes involved in phase II reactions in the liver, because of its high expression level. In this study, we generated mice with disrupted glutathione S-transferase Mu 1 gene (Gstm1-null mice) by gene targeting, and characterized the phenotypes by cytosolic and in vivo studies. The resulting Gstm1-null mice appeared to be normal and were fertile. Expression analyses for the Gstm1-null mice revealed a deletion of Gstm1 mRNA and a small decrease in glutathione S-transferase alpha 3 mRNA. In the enzymatic study, GST activities toward 1,2-dichloro-4-nitrobenzene (DCNB) and 1-chloro-2,4-dinitrobenzene (CDNB) in the liver and kidney cytosols were markedly lower in Gstm1-null mice than in the wild-type control. Gstm1-null mice had GST activities of only 6.1 to 21.0% of the wild-type control to DCNB and 26.0 to 78.6% of the wild-type control to CDNB. After a single oral administration of DCNB to Gstm1-null mice, the plasma concentration of DCNB showed larger AUC0–24 (5.1–5.3 times, versus the wild-type control) and higher Cmax (2.1–2.2 times, versus the wild-type control), with a correspondingly lower level of glutathione-related metabolite (AUC0–24, 9.4–17.9%; and Cmax, 9.7–15.6% of the wild-type control). In conclusion, Gstm1-null mice showed markedly low ability for glutathione conjugation to DCNB in the cytosol and in vivo and would be useful as a deficient model of GSTM1 for absorption, distribution, metabolism, and excretion/toxicology studies.