Arch Toxicol (2006) 80: 187–195 DOI 10.1007/s00204-005-0027-3
T OX I C OK I N ET I C S A ND M ET A B OL I SM
Y. Gue´guen Æ M. Souidi Æ C. Baudelin Æ N. Dudoignon S. Grison Æ I. Dublineau Æ C. Marquette Æ P. Voisin P. Gourmelon Æ J. Aigueperse
Short-term hepatic effects of depleted uranium on xenobiotic and bile acid metabolizing cytochrome P450 enzymes in the rat Received: 27 June 2005 / Accepted: 23 August 2005 / Published online: 18 October 2005 Springer-Verlag 2005
Abstract The toxicity of uranium has been demonstrated in different organs, including the kidneys, skeleton, central nervous system, and liver. However, few works have investigated the biological effects of uranium contamination on important metabolic function in the liver. In vivo studies were conducted to evaluate its effects on cytochrome P450 (CYP) enzymes involved in the metabolism of cholesterol and xenobiotics in the rat liver. The effects of depleted uranium (DU) contamination on Sprague–Dawley were measured at 1 and 3 days after exposure. Biochemical indicators characterizing liver and kidney functions were measured in the plasma. The DU affected bile acid CYP activity: 7a-hydroxycholesterol plasma level decreased by 52% at day 3 whereas microsomal CYP7A1 activity in the liver did not change significantly and mitochondrial CYP27A1 activity quintupled at day 1. Gene expression of the nuclear receptors related to lipid metabolism (FXR and LXR) also changed, while PPARa mRNA levels did not. The increased mRNA levels of the xenobiotic-metabolizing CYP3A enzyme at day 3 may be caused by feedback up-regulation due to the decreased CYP3A activity at day 1. CAR mRNA levels, which tripled on day 1, may be involved in this up-regulation, while mRNA levels of PXR did not change. These results indicate that high levels of depleted uranium, acting through modulation of the CYP enzymes and some of their nuclear receptors, affect the hepatic metabolism of bile acids and xenobiotics.
Y. Gue´guen and M. Souidi participated equally to this work. Y. Gue´guen (&) Æ M. Souidi Æ C. Baudelin Æ N. Dudoignon S. Grison Æ I. Dublineau Æ C. Marquette Æ P. Voisin P. Gourmelon Æ J. Aigueperse Institut de Radioprotection et de Suˆrete´ Nucle´aire, Direction de la RadioProtection de l’Homme, Service de Radiobiologie et d’Epide´miologie. IRSN, B.P. No. 17, F 92262 Fontenay-aux-Roses Cedex, France E-mail:
[email protected] Tel.: 33-1-58359978 Fax: 33-1-58358467
Keywords Depleted uranium Æ Cytochromes P450 Æ Bile acids Æ Xenobiotics Æ Liver
Introduction Beyond its natural presence in the environment (Durakovic 1999), uranium is dispersed by civil and military applications (Betti 2003; Bleise et al. 2003). These populations could be exposed to acute contamination in the case of accident or armed conflicts and to chronic contamination via the presence of uranium in the food chain or well water. The toxic action of uranium and its compounds have been reported in vitro (Miller et al. 2004) and in vivo (Leggett 1989). Recently, effects on physiological systems such as reproduction and development have been described (Domingo 2001; Linares et al. 2005) Specifically, its chemical and radiological toxicity has been demonstrated in a variety of organs, including the kidneys, lungs, skeleton, and central nervous system. Among these, the kidney is the most sensitive to uranium. Nevertheless its accumulation in the liver has been described in some experimental studies (McClain et al. 2001; Leggett and Pellmar 2003), although few have investigated its precise biological effect in this tissue. Previous works report histological and functional alterations to the cytochrome P450 (CYP) hepatic system after acute uranium contamination (Pasanen et al. 1995; Chung et al. 2003; Moon et al. 2003): uranium modulates liver cytochrome P450 (CYP) activity and expression at short term. This finding indicates that uranium may target key hepatic CYP involved in the metabolism of cholesterol or bile acids or in the detoxification of xenobiotics. The CYP monooxygenase system is critical for the metabolism of both endogenous and exogenous lipophilic substrates. Cholesterol degradation to bile acids in the liver can be initiated either by cholesterol 7ahydroxylase (CYP7A1) in the classic pathway or by mitochondrial 27-hydroxylase (CYP27A1) in the alternative pathway (Chiang 2004). Nuclear receptors
188
provide the primary, mainly transcriptional, regulation of these pathways. These include the oxysterol receptor LXR and the bile acid receptor FXR, which play a major role in homeostasis of bile acids and cholesterol (Repa and Mangelsdorf 2000; Lu et al. 2001). The PXR (pregnane X receptor) and CAR (constitutive androstane receptor) nuclear receptors also play a role in regulating the CYP genes involved in bile acid metabolism, in addition to several CYP that metabolize xenobiotics (Handschin and Meyer 2003; Maglich et al. 2002). The CYP3A is one of the most important enzymes of the latter enzymes, while PXR has been described as a steroid and xenobiotic ‘‘sensing’’ receptor that responds to a broad range of compounds to which humans are exposed in the environment (Blumberg and Evans 1998). The CAR receptor is a close relative of PXR, with which it shares several properties including regulation of CYP3A expression (Handschin and Meyer 2003; Honkakoski et al. 2003). Recent studies indicate that CAR mediates several different metabolic stresses in addition to xenobiotic stresses (Goodwin and Moore 2004). Uranium-induced alterations of these metabolic processes might induce the onset of various pathologies, including hepatitis and cholestasis, due to changes in endogenous metabolism (Nebert and Russell 2002) or alterations of drug pharmacokinetics and drug-drug interactions. The objectives of this work were thus to determine if important liver CYP enzymes were target gene of depleted uranium (DU). In this regards, the enzyme activity and expression of the hepatic CYP involved in the metabolism of cholesterol and bile acids (CYP7A1 and CYP27A1) and in xenobiotic detoxification (CYP3A1 and CYP3A2), as well as on the related nuclear receptors (PXR, CAR, LXR, FXR, and PPARa) were analysed. Plasma levels of cholesterol metabolites were measured simultaneously (7a-hydroxycholesterol for bile acid metabolism and 27-hydroxycholesterol for extrahepatic cholesterol metabolism). The study was performed in rats at 1 and 3 days after administration of DU.c
Materials and methods Animals The experiments were performed in male Sprague– Dawley rats (300 g) obtained from Charles River (L’Arbresle, France), housed at constant room temperature (22C±1) with a 12 h:12 h (light/dark) cycle and provided free access to standard rat pellets (AFE, Augy, France) and water throughout the acclimatation experimental periods. The rats received a single subcutaneous administration of depleted uranium (uranyl hexahydrated, VWR, France) (11.5 mg/kg) dissolved in sterile sodium chloride 0.9% (3.5 mg in 200 ll, pH=4.5). This dose was choosed as a sublethal toxic dose for rats (Domingo et al. 1987). Control animals received the
same volume of saline solution. They were then carefully monitored (food and water intake, body weight) until the end of the experiment. They were euthanized at 1 or 3 days after DU administration. At theses times, samples were collected for further analysis. The experiments were approved by Animals Care Committee of the Institute and conducted in accordance with French regulations for animal experimentation (Ministry of Agriculture Act No. 2001-464, May 2001). Plasma clinical and biochemical parameters We used an automated Technicon RA-XT (Bayer Diagnostics, France) system to measure the levels of plasma creatinin, urea, ALT, AST, GGT, AP, bilirubin, iron, cholesterol, and triglycerides (biological chemistry reagents, Bayer Diagnostics) at days 1 and 3 after acute subcutaneous administration of DU. Chemicals and isotopes [4-14C]Cholesterol was obtained from NEN Products (Les Ulis, France). Cholesterol, 7a-hydroxcholesterol, and 27-hydroxycholesterol were obtained from Sigma Diagnostics (Isle d’Abeau Chesnes, France), as were cholesterol oxidase, testosterone, and cortexolone. Hydroxypropyl-b-cyclodextrin (HPbCD) was kindly donated by Dr. Michel Riottot (University Paris XI, Orsay, France). Plasma 7a-hydroxycholesterol and 27-hydroxycholesterol measurements Blood samples were taken in EDTA-containing tubes; aliquots for analysis of total plasma 7a-hydroxycholesterol and 27-hydroxycholesterol were centrifuged at 4000·g for 10 min at 4C. After adding 10 lg tertiary butylated hydroxytoluene as an antioxidant, the plasma was stored at 80C until analysis. Plasma 7a-hydroxycholesterol and 27-hydroxycholesterol were quantified by high-performance liquid chromatography (HPLC), as described by Chen (Chen et al. 1998) with some modifications. Briefly, 1 ml of plasma was saponified with a mixture of NaOH/ethanol (1:9) for 1 h at 55C in a shaking water bath after adding 3-acetyl-19-hydroxycholesterol (0.3 lg) as an internal standard. Oxysterols were extracted by adding 2 ml of hexane and evaporating the solvent under nitrogen gas. The solute was dissolved by 75 mM potassium phosphate buffer (pH 7.4) containing EDTA (1 mM), DTT (0.5 mM), and MgCl2 (5 mM) with 45% (w/v) HPbCD and then incubated in 200 ll potassium phosphate buffer (pH 7.4) with 3 IU of cholesterol oxidase (cellulomonas sp.) at 37C for 30 min. The reaction was terminated by adding a mixture of methanol–chloroform (2:1 v/v, 2 ml). Sterols were extracted with
189
chloroform (2 ml) and water (1.5 ml). The combined extracts were evaporated to dryness under nitrogen and redissolved in 250 ll of acetonitrile. Oxysterols (ketone derivatives) were separated by HPLC (Waters Symmetry, C18, 4.6·250 mm, 5 lm particle size). The mobile phase was acetonitrile/ethanol (98:2). Flow rate was kept at 1 ml/min and absorbency was monitored at 240 nm. Oxysterol peaks were identified by their retention times compared with those of known standards. In our system, 7a-hydroxycholesterol eluted at 14.7 min, 27-hydroxycholesterol at 18.6 min, and the internal standard at 23.5 min. Preparation of liver microsomes and mitochondria Livers were obtained from both control and DU-treated rats (1 and 3 days after subcutaneous administration). Under anesthesia with sodium pentobarbital (100 mg/kg of body weight i.p.), the liver was rapidly removed, chilled in ice-cold buffer, and sliced; 1-g portions were taken and homogenized in buffer (KH2PO4 50 mM, sucrose 300 mM, dithiothreitol 0.5 mM, EDTA 10 mM, NaCl 50 mM, pH 7.4), according to a procedure already described (Souidi et al. 1998). The homogenate was centrifuged for 20 min at 20,000·g and the supernatant centrifuged at 100,000·g for 1 h. The 20,000·g pellet was resuspended in buffer and again centrifuged at 100,000·g for 1 h. The microsomal pellet was homogenized in buffer, sampled and stored at 80C until required. The 20,000·g pellet of homogenate was gently resuspended in 7 ml of buffer and homogenized by ten strokes of a Teflon pestle (motor-driven homogenizer). The homogenate was centrifuged at 2000·g for 10 min. The supernatant was then centrifuged at 9000·g for 10 min and the mitochondrial pellet finally resuspended in 1.2 ml of buffer, fractionated into 200 ll samples, and stored at 80C. We determined microsomal and mitochondrial protein content as previously described, according to the Lowry method with bovine serum albumin as a standard (Lowry et al. 1951). Determination of CYP7A1 and CYP27A1 enzyme activity in the liver Cholesterol 7a-hydroxyase (CYP7A1) in the microsomal fractions was assayed according to a radioisotopic method that used solubilized [4-14C]Cholesterol, with hydroxypropyl-b-cyclodextrin as a carrier (Souidi et al. 1998). Sterol 27-hydroxylase (CYP27A1) in the mitochondrial fractions was assayed with a radioisotopic method that used [4-14C]Cholesterol, solubilized in hydroxypropyl-b-cyclodextrin (Souidi et al. 1999).
modification. Briefly, for a total volume of 500 ll, each assay tube contained KH2PO4 75 mM (pH 7.4), EDTA 1 mM, DTT 0.5 mM, MgCl2 5 mM, NADPH 1 mM, 250 lg of liver microsomal protein and testosterone 200 lM, and was solubilized by 160 ll of potassium phosphate buffer containing 45% (w/v) HPbCD (for an enzyme assay 10 ll of this solution contained 100 nmol of testosterone, 4.5 mg HPbCD). The preincubation time, in which only NADPH was omitted, was 5 min at 37C, and the incubation time following initiation of the assay with NADPH was 15 min. The reaction was terminated by the addition of 2 ml methanol/chloroform (2:1, v/v) and 10 ll cortexolone (10 nmol) as an internal standard. Testosterone and its metabolites in the microsomal samples were extracted with 2 ml choloroform and 1.5 ml water. The tubes were again vortexed for 1 min, and 2 ml of the organic phase was collected and evaporated (nitrogen gas). The dry residue was reconstituted with 250 ll of 26% acetonitrile in water. An aliquot of 15 ll of the reconstituted solution was injected into the HPLC for analysis with a LiChrospher RP-18 (250 mm/ 5 lm particle size, flow rate of 1.3 ml/min, absorbance 247 nm). The testosterone 6b-hydroxylase activity is expressed as picomoles per minute per mg of protein. Real-time quantitative RT-PCR Real-time PCR was used to analyse the mRNA expression of the CYP3A1 and 3A2 and of the CAR, PXR, RXR, FXR, LXRa, and PPARa nuclear receptors. Total RNA was prepared with the RNeasy total RNA isolation Kit (Qiagen, France) according to the manufacturer’s instructions. RNA integrity was confirmed by denaturing agarose gel electrophoresis with ethidium bromide staining. The cDNA was produced from 1 lg of total RNA by reverse transcription with 200 U of Superscript reverse transcriptase (GIBCO) in a 20-ll reaction containing one time Superscript buffer (GIBCO), 1 mM 2-deoxynucleotide 5¢-triphosphate, 20 ng random hexamer, 10 mM DTT, and 20 U Rnase inhibitor. After incubation for 50 min at 42C, the reaction was terminated by a denaturing enzyme for 10 min at 70C. The PCR amplification of the CYPs and nuclear receptors used Syber PCR master mix (PE Applied). Optimized PCR used the Abi Prism 7000 Sequence detection system (Applied Biosystems). The PCR fluorescent signals of each target gene were normalized to the fluorescent signal obtained from the housekeeping gene HPRT (hypoxanthine-guanine phosphoribosyltransferase) for each sample. Sequences for the forward and reverse primers used in the present study are listed in Table 1. Statistical analysis
6b-Testosterone hydroxylase activity in the liver Testosterone hydroxylase activity was determined with a technique adapted from (Li et al. 2002) after minor
Results are reported as means ± SE. Statistical analyses were performed with Student’s t-test. Differences were considered significant when p
