![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
|
|
|
|
|
||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
IN PRIMARY HUMAN HEPATOCYTES
Departments of Pharmacology/Toxicology (T.L.N., P.R.S., J.F.S.), Microbiology/Immunology (R.C.N., J.M.J.), Medicine (R.C.N.), Biochemistry (H.W.T., J.M.J., P.R.S., J.F.S.), Dartmouth Medical School, Hanover, New Hampshire; Veterans Administration Medical Center, White River Junction, Vermont (T.L.N., S.G.W., R.C.N., H.W.T., J.M.J., P.R.S., J.F.S.); Department of Safety Sciences, Pfizer, Ann Arbor, Michigan (V.E.K.); Department of Pathology, University of Pittsburgh Medical Center, Pittsburgh, Pennsylvania (S.C.S.); Lilly Research Laboratories, Indianapolis, Indiana (S.A.W.); and Howard Hughes Medical Institute, Salk Institute for Biological Studies, San Diego, California (R.M.E.)
(Received January 31, 2005; accepted April 13, 2005)
 |
Abstract |
|---|
 Top Abstract Materials and MethodsResultsDiscussionReferences |
|---|
(RXR) and the pregnane X receptor (PXR). We found that arsenite failed to affect expression of PXR or the transcription factor Sp1, yet caused a significant decrease in PXR responsiveness to Rif. Arsenite caused a large decrease in nuclear RXR protein and, to a lesser extent, RXR mRNA. These results suggest that arsenite inhibits both untreated and induced CYP3A4 transcription in primary human hepatocytes by decreasing the activity of PXR, as well as expression of the nuclear receptor RXR.
). Arsenic is a naturally occurring element found within the Earth's crust and, therefore, is an environmental contaminant in many areas throughout the world. In recent years, environmental exposure to arsenic has been implicated in the occurrence of various cancers and numerous health problems (Yoshida et al., 2004). One target of arsenic appears to be the liver, as hepatocellular carcinoma, fibrosis, and cirrhosis, as well as biliary occlusion are associated with arsenic exposure (Ferrini et al., 1997; Runge et al., 2000).
A target of arsenite in liver cells may be the P450s, which are gene families of hemoproteins that catalyze the oxidation of many endogenous and exogenous lipophilic chemicals (Nelson et al., 1996). Exposure to organic toxic chemicals can result in increases in P450s involved in their metabolism, a response that may have evolved to protect against the toxicity of such chemicals (Whitlock, 1999). In rodents, acute exposure to arsenite reduces both basal and induced levels of some hepatic P450s, as measured spectrally or enzymatically (Sardana et al., 1981; Albores et al., 1992). In primary cultures of rat hepatocytes, arsenite has been found to decrease induction of CYP1A1/2, 2B1, and 3A23 (Jacobs et al., 1999). The effect of arsenite to decrease P450s can have major implications in human health, by altering the metabolism and elimination of toxic chemicals, as well as drugs, that are substrates for such P450s.
It is generally accepted that CYP3As are the most abundant P450 proteins in the human liver, accounting for between 30% and 60% of the total cytochrome P450 content, with CYP3A4 being the major hepatic CYP3A present (Wrighton et al., 1990; Shimada et al., 1994). In humans, CYP3A proteins are involved in the metabolism of 45 to 60% of all currently used drugs (Li et al., 1995; Evans and Relling, 1999). Therefore, variability in CYP3A4 expression would be expected to have a profound effect on the efficacy and safety of drugs that have a narrow therapeutic index and are metabolized by CYP3As. In addition to being a worldwide contaminant, arsenic, in the form of arsenic trioxide, is a first-line treatment for cancers such as multiple myeloma (Rousselot et al., 2004) and acute promyelocytic leukemia (Evens et al., 2004). Arsenic-mediated decreases in the de novo synthesis of CYP3A may compromise the metabolism or effectiveness of other drugs administered to these patients. On the other hand, arsenic-mediated decreases in CYP3A could prove advantageous in prolonging the half-life of coadministered drugs in which the parent drug is active.
P450 gene activation is induced by many natural and environmental compounds. The xenobiotic-mediated induction of CYP3A4 is regulated primarily by PXR (Bertilsson et al., 1998; Blumberg et al., 1998; Lehmann et al., 1998). However, Goodwin et al. (2002a) have recently reported that the constitutive androstane/active receptor (CAR) can transactivate expression of the CYP3A4 gene in HepG2 cells. Rif has been shown to activate human PXR, but not CAR, in transfection studies using CV-1 cells (Moore et al., 2000). CAR and PXR are members of the nuclear receptor superfamily and form heterodimers with RXR. PXR is activated by a number of xenobiotics and steroids (Goodwin et al., 2002b) and regulates CYP3A gene induction by targeting specific response elements present in the regulatory region of CYP3A (Sueyoshi and Negishi, 2001; Goodwin et al., 2002b). Although PXR and CAR function as heterodimers with RXR, little is known about the role of RXR or RXR-selective ligands in P450 regulation.
We have previously reported that 5 µM arsenite decreases phenobarbital (PB)-induced CYP3A23 protein, with little to no decrease in CYP3A23 mRNA, in primary cultures of rat hepatocytes (Jacobs et al., 1999). Arsenite has also been shown to decrease CYP1A1 and CYP1A2 expression in primary human hepatocytes (Vakharia et al., 2001), benzo[a]pyrene-induced CYP1A1 and CYP1B1 expression in T-47D human breast cancer cells (Spink et al., 2002), and benzo[k]-fluoranthene-mediated induction of CYP1A1 mRNA in HepG2 cells (Bessette et al., 2005). Although the mechanisms underlying arsenite-mediated decreases in P450s have yet to be identified, several hypotheses have been generated, including transcriptional and posttranslational events (Spink et al., 2002; Vernhet et al., 2003).
In the present study, we examined the effect of low concentrations of arsenite (2.5 or 5 µM) on untreated and Rif- or PB-induced expression of CYP3A4, as well as on several transcription factors involved in CYP3A4 expression. We found that arsenite abolished the untreated and induced expression of CYP3A4 mRNA and protein as well as associated catalytic activity. Furthermore, arsenite caused a dramatic decrease in nuclear RXR protein, while having no effect on expression of PXR, or the transcription factor Sp1. Arsenite also caused a significant decrease in the responsiveness of PXR to inducer. These results suggest that arsenite is acting at the level of CYP3A4 transcription, by decreasing expression of RXR and activity of PXR.
|
Materials and Methods |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
was purchased from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA). Alkaline phosphatase-conjugated goat anti-rabbit antibody was purchased from Bio-Rad (Hercules, CA).
Human Hepatocyte Culture. Human hepatocytes were prepared from livers not used for whole organ transplant within 24 h of procurement. Information on the liver donors is provided in Table 1. Hepatocytes were isolated by a three-step collagenase perfusion technique, as described previously (Strom et al., 1996), and plated at a cell density of 2 x 106 cells per well in six-well plates previously coated with type I collagen. The hepatocytes were maintained in Williams E medium supplemented with 10–7 M dexamethasone (DEX), 10–7 M insulin, 100 units/ml penicillin G, 100 µg/ml streptomycin, 1 mM ascorbate, 10–7 M sodium selenite, 20 mM Hepes and were kept at 37°C in a humidified incubator with 95% air/5% CO2. After 24 to 48 h in culture, cells either remained untreated or were exposed to 10 µM Rif or 2 mM PB for 24 h, with or without 2.5, 5, or 10 µM arsenite or 20 µM ferric nitrilotriacetic acid (FeNTA), as indicated in the figure legends. Nuclear and cytoplasmic extracts were prepared from hepatocytes, as described (Pascussi et al., 2000a). PB was dissolved in saline. As (made fresh) was dissolved in water. Rif and testosterone were dissolved in dimethyl sulfoxide. The final concentration of dimethyl sulfoxide in culture medium was below 1 µl/ml. FeNTA was prepared as described (Shedlofsky et al., 1983).
|
Rat Hepatocyte Culture and Transfection. Primary cultures of rat hepatocytes were transfected with pGL3CYP3A23, a luciferase reporter construct containing –1360 to +82 base pairs of the 5' upstream region of rat CYP3A23 (Burger et al., 1992) and pCMX-PL2-hSXR, an expression vector containing the complete protein-coding region of human PXR [steroid and xenobiotic receptor (SXR)] (Blumberg et al., 1998). Cells were prepared from mature male Fisher 344 rats (200 to 250 g), as previously described (Sinclair et al., 1990). The yield was approximately 350 x 106 cells per liver. Cells were plated onto 3.5-cm tissue culture plates, previously coated with collagen, and maintained in Williams E medium containing 10–7 M DEX, 10–7 M insulin, 10–7 M sodium selenite, 100 units/ml penicillin G, 100 µg/ml streptomycin, and 0.3 mM ascorbate. Forty-eight hours after isolation, transfection of the plasmid DNA was performed with Lipofectamine 2000 (Invitrogen), according to the manufacturer's instructions. Transfections were performed in fresh medium using 2 x 106 cells, 500 ng of pGL3CYP3A23, 100 ng of pCMX-PL2-hSXR, and 50 ng of pRL-tk (Promega, Madison, WI) as an internal control. After 6 h, the medium was changed, and fresh medium containing Matrigel (0.2 mg/ml) was added. The following day, chemicals were added in fresh medium. Cells were harvested in Passive Lysis Buffer (Promega) 18 h after treatments and firefly and Renilla luciferase levels in cell extracts were analyzed according to the manufacturer's directions (Promega).
Protein Synthesis. Protein synthesis was determined in intact cells by the incorporation of [14C]leucine (specific radioactivity, 0.1 mCi/mmol, 0.2 µCi/plate) into acid-precipitable material in 1 h, as described (Kostrubsky et al., 1997). The rates of protein synthesis are expressed relative to cells treated with inducer (Rif) alone.
RNA Isolation and Invader RNA Analysis. Total RNA was extracted from cells using guanidinium isothiocyanate lysis buffer and cesium chloride ultracentrifugation, as described (Hamilton et al., 1988). CYP3A4 mRNA was quantified using the CYP3A4 Invader RNA assay kit, according to instructions provided by the manufacturer (Third Wave Technologies, Madison, WI). This assay is specific for CYP3A4 and uses a fluorescence resonance energy transfer-based signal amplification method that is based on an enzyme-substrate reaction using Cleavase enzymes. These enzymes cut only the specific structure on targeted mRNA that is formed during the Invader process. Primary reactions (20 µl) were performed in 96-well microplates (MJ Research, Watertown, MA), and to each well, 5 µl of primary reaction components (10 mM MOPS, pH 7.5, 0.05% Tween 20, 0.05% Nonidet P-40, 4% polyethylene glycol, 100 mM KCl, 12.5 mM MgSO4, 2 ng/µl Tth 5'-nuclease, 0.8 µM probe, 0.5 µM invasive oligonucleotide, and 0.3 µM stacking oligonucleotide), 5 µl of standard or RNA sample, and 10 µl of mineral oil were added. Microplates were incubated at 60°C for 1 h. To initiate the secondary reaction, 5 µl of secondary reaction components (10 mM MOPS, pH 7.5, 0.05% Tween 20, 0.05% Nonidet P-40, 20 mM MgSO4, 0.67 µM fluorescence resonance energy transfer oligonucleotide, 0.1 µM secondary reaction template, and 2.1 µM arrestor oligonucleotide) were added to each well and incubated at 60°C for 1 h. Microplates were read directly in a SpectraMax Gemini XS (Molecular Devices, Sunnyvale, CA) using SOFTmax Pro Software (Molecular Devices) at 485 nm excitation and 530 nm emission. A standard curve for CYP3A4 mRNA was generated in each assay using purified CYP3A4 mRNA provided by the manufacturer. Expression levels (CYP3A4 mRNA molecules/ng total RNA) were calculated using net signal values from the CYP3A4 standard curve equation and reported as fold induction over control. Each sample was assayed in triplicate using a range of 1 to 50 ng of total RNA per reaction.
RT-PCR. Human PXR, Sp1, and glyceraldehyde-3-phosphate dehydrogenase RNAs were measured by RT-PCR using the RNA PCR kit (Takara, Kyoto, Japan), according to the manufacturer's instructions. RXR expression levels were measured by RT-PCR using the SuperScript First-Strand Synthesis (Invitrogen) and Takara PCR Amplification kits. The primer sequences were as follows: PXR sense, 5'-TCCGGAAAGATCTGTGCTCT-3', antisense, 5'-AGGGAGATCTGGTCCTCGAT-3'; Sp1 sense, 5'-AATTCAAGGCCTGCCGTT -3', antisense, 5'-CCATGGAGACCAAGCTGAGC-3'; RXR sense, 5'-TTCGCTAAGCTCTTGCTC-3', antisense, 5'-ATAAGGAAGGTGTCAATGGG-3'; glyceraldehyde-3-phosphate dehydrogenase sense, 5'-GGTCGGAGTCAACGGATTTGGTCG-3', antisense, 5'-CAAAGTTGTCATGGATGACC-3'. Amplification was carried out in a Bio-Rad iCycler with initial denaturation at 94–95°C for 2 min, followed by 16 to 24 cycles of 94°C for 30 s, 60–62°C for 30 s, and 72°C for 30 s. PCR samples were electrophoresed on agarose gels and visualized by ethidium bromide. The number of PCR cycles resulting in PCR products in the linear phase of the amplification curve was determined. The primers were designed to cross exon junctions, and PCR products amplified from genomic DNA templates were not seen. PCR products were sequenced to verify integrity. In some experiments, the resulting gels were quantitated by densitometry, using Adobe Photoshop (Adobe Systems, Mountain View, CA) with an HP Precision Scanner (Hewlett Packard, Palo Alto, CA) and OneDScan software (Scanalytics, Fairfax, VA).
Western Blot. CYP3A4 and Sp1 protein levels were measured in cell sonicates, and PXR and RXR protein levels were analyzed in nuclear and cytoplasmic fractions. Briefly, proteins were separated by electrophoresis in SDS-polyacrylamide gel (10% acrylamide) and transferred to nitrocellulose electrophoretically at 100 V for 1 h. The nitrocellulose sheets were blocked overnight at 4°C in phosphate-buffered saline containing 5% nonfat dry milk and 0.3% Tween 20, incubated for 1 h at room temperature (CYP3A4) or overnight at 4°C (Sp1, PXR, RXR) with primary antibodies, and exposed for 1 h to an alkaline phosphatase-conjugated goat anti-rabbit secondary antibody. The resulting blots were quantitated by densitometry, using Adobe Photoshop with an HP Precision Scanner and OneDScan software (Scanalytics).
CYP3A4 Enzymatic Activity. The activity of CYP3A4 was measured by formation of 6ß-hydroxytestosterone in intact cultured hepatocytes, as described (Kostrubsky et al., 1999). Briefly, after a 24-h exposure to chemicals, the culture medium was replaced with fresh Williams E medium containing 200 µM testosterone. The cells were incubated for 30 min at 37°C, and 6ß-hydroxytestosterone was measured in the medium by HPLC, as described (Kostrubsky et al., 1999).
Additional Assays. Proteins were determined by the method of Lowry et al. (1951) using bovine serum albumin as the standard. Lipid peroxidation in hepatocytes and media was measured by the generation of thiobarbituric acid reactive substances (TBARS), as described (Shedlofsky et al., 1983).
Statistical Analyses. Data are presented for representative experiments that were repeated two or more times. Within each experiment, treatments were performed in duplicate or triplicate, as indicated in the figure legends. In triplicate treatments, results were analyzed either by a Student's t test or by analysis of variance, followed by a Student-Newman-Keuls multiple comparisons test. A p value of <0.05 was taken to indicate significance.
|
Results |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
) found that 5 µM arsenite decreased induction of CYP1A1 and 1A2 in primary cultures of human hepatocytes with no effect on cell viability, as measured by the 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide assay. Therefore, we investigated the effect of 2.5 and 5 µM arsenite on induction of CYP3A4 by Rif and PB. Neither of these doses of arsenite, in the presence of 10 µM Rif, was toxic to the cells, as evidenced by the lack of decrease in total protein synthesis (Fig. 1A). In contrast, a higher dose of arsenite (10 µM) caused a 30% decrease in protein synthesis (results not shown) and, therefore, was not used in later studies.
|
We examined three parameters in investigating the effect of arsenite on induction of CYP3A4 by Rif or PB: mRNA, protein, and enzyme activity. CYP3A4 mRNA levels were measured using the highly sensitive and specific CYP3A4 Invader RNA assay, which does not detect the other forms of human CYP3A (de Arruda et al., 2002). Inducibility of CYP3A4 by Rif and PB was variable among human cultures from different donors, as previously reported for P450s in human hepatocyte cultures (Daujat et al., 1991; Kocarek et al., 1995; Madan et al., 2003). Figure 1B demonstrates that Rif induced expression of CYP3A4 mRNA and that arsenite (2.5 or 5 µM) abolished induction of CYP3A4 mRNA in three separate human hepatocyte cultures. To assess whether these diminished CYP3A4 mRNA levels would have an impact on the amount of immunoreactive protein formed, as well as associated catalytic activity, CYP3A4 protein and activity levels were determined. In a representative experiment, treatment with 5 µM arsenite for 24 h caused a 75% decrease in CYP3A4 protein compared with Rif treatment alone (Fig. 1C). CYP3A4 catalytic activity was measured by determining the 6ß-hydroxylation of testosterone by intact cells (Kostrubsky et al., 1999). The fold increases in CYP3A4 activity did not correlate directly with the increases in CYP3A4 mRNA (Fig. 1, D versus B). However, low-dose arsenite (2.5 or 5 µM) caused significant decreases in CYP3A4 activity in all three cultures (Fig. 1D). Thus, Rif-mediated induction of CYP3A4 mRNA, protein, and enzyme activity was abolished by simultaneous treatment with arsenite.
We next investigated whether arsenite decreases CYP3A4 induction by PB, which is capable of inducing CYP3A4 through PXR or CAR (Goodwin et al., 2002a; Moore et al., 2000), in contrast to Rif, which induces CYP3A4 through PXR (Moore et al., 2000). In two human hepatocyte cultures that were highly responsive to PB-mediated induction of CYP3A4 (Fig. 2A, inset; HH 994 and HH 1051), arsenite decreased the induction of CYP3A4 mRNA (Fig. 2A). Although two other hepatocyte cultures were much less responsive to PB-mediated induction of CYP3A4 (Fig. 2A; HH 845 and HH 888), arsenite still caused a significant decrease in CYP3A4 mRNA. PB-induced CYP3A4 protein (Fig. 2B) and catalytic activity (Fig. 2C) were also decreased when arsenite was included in the treatment. These results show that arsenite decreases both Rif- and PB-mediated induction of CYP3A4 at the level of mRNA, protein, and enzyme activity.
|
Effect of Arsenite on Expression of CYP3A4 mRNA, Immunoreactive Protein, and Enzyme Activity in Untreated Primary Human Hepatocytes. We next investigated whether arsenite affects expression of CYP3A4 in untreated human hepatocytes. Similar to our findings with induced CYP3A4 mRNA levels, arsenite caused a significant decrease in levels of CYP3A4 mRNA in hepatocytes not treated with inducers (Fig. 3A). Likewise, CYP3A4 protein (Fig. 3B) and catalytic activity (Fig. 3C) were diminished after 24 h of treatment with 5 µM arsenite. Therefore, in addition to interfering with the induction of CYP3A4 mRNA, arsenite suppressed the expression of CYP3A4 in cells that were not cotreated with inducers.
|
). Analyses by RT-PCR show that a 24-h treatment with 2.5 or 5 µM arsenite did not affect PXR or Sp1 mRNA levels (Fig. 4A). PXR protein was detected in both the cytoplasm and the nucleus (Fig. 4B). None of the treatments altered the cellular localization of PXR, including arsenite alone, Rif alone, or both in combination (Fig. 4B). These treatments also failed to affect the levels of PXR and Sp1 proteins (Fig. 4B).
|
). The cultures were transiently transfected with a firefly luciferase reporter vector containing nucleotides –1360 to +82 of the 5' upstream region of rat CYP3A23 (pGL3CYP3A23) (Burger et al., 1992). These cultures were cotransfected with pCMX-PL2-hSXR, an expression vector containing the complete protein-coding region of human PXR (Blumberg et al., 1998), as well as a Renilla luciferase plasmid as an internal control for transfection efficiency. Figure 5A shows that Rif is not able to increase reporter gene expression in the absence of human PXR, but upon cotransfection of a vector containing human PXR, Rif caused a 2-fold increase in reporter gene expression. When arsenite was also included, reporter gene expression was decreased by approximately 75% (Fig. 5B), indicating an effect of arsenite to decrease the activity of human PXR.
|
Effect of Arsenite on RXR Expression in Primary Human Hepatocytes. We next investigated the effect of arsenite on expression of the nuclear receptor RXR. In contrast to PXR, RXR protein was detected in nuclear fractions, but not in cytoplasmic fractions (Fig. 6A). Figure 6A shows that a 24-h treatment with arsenite caused a dramatic decrease in nuclear RXR protein levels. Similar results were observed in cultures from five separate livers (results not shown). Arsenite-mediated decreases in RXR were observed in untreated as well as Rif-treated cells. To determine whether arsenite decreased RXR protein by decreasing RXR mRNA, we measured RXR mRNA by RT-PCR. Figure 6B shows that arsenite decreased expression of RXR mRNA; however, the extent of the decrease was less than that of the protein (Fig. 6A). These results suggest that the effect of arsenite to decrease RXR may contribute to diminished transcription of CYP3A4 following treatment with arsenite.
|
). To determine whether 5 µM arsenite causes oxidative stress in primary cultures of human hepatocytes, we measured the generation of lipid peroxides as an indicator of the oxidative response. Exposure of hepatocytes to 20 µM FeNTA, an iron compound known to cause the formation of ROS (Morel et al., 1990), for 24 h caused a 3- to 4-fold increase in lipid peroxidation (Fig. 7), as determined by the generation of TBARS. In contrast, arsenite (2.5 or 5 µM), in combination with Rif (10 µM), did not cause an increase in lipid peroxidation (Fig. 7).
|
|
Discussion |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
), 5 µM arsenite failed to increase lipid peroxidation in human hepatocytes (Fig. 7). Importantly, we show that arsenite caused a major reduction in the activity of PXR (Fig. 5B) and nuclear protein levels of RXR (Fig. 6A), with no effect on PXR or Sp1 expression (Fig. 4). The overall findings suggest that arsenite acts at the level of transcription by inhibiting PXR activity and lowering the amount of RXR available for expression of CYP3A4.
Transcriptional regulation of P450s is complex. PXR and CAR have both been shown to regulate gene expression of CYP3A4, depending on the inducer (Bertilsson et al., 1998; Blumberg et al., 1998; Lehmann et al., 1998; Goodwin et al., 2002a). Both PB and Rif can activate human PXR (Moore et al., 2000). PB has recently been shown to cause translocation of CAR to the nucleus in primary human hepatocyte cultures (Assenat et al., 2004), indicating that PB can also activate human CAR. We investigated whether arsenite affects Rif- and PB-mediated induction of CYP3A4 in primary human hepatocytes. Both Rif- and PB-induced CYP3A4 mRNA levels were abolished by arsenite (Figs. 1B and 2A), suggesting that arsenite affects PXR- and possibly CAR-mediated transcription of CYP3A4. One possibility is that arsenite targets a factor common to both receptors. RXR heterodimerizes with PXR and CAR to transactivate target genes in a ligand-dependent manner (Xie et al., 2000b). Our findings that arsenite decreased nuclear protein levels of RXR in both untreated and Rif-induced hepatocytes (Fig. 6A) may account for the decreased expression of CYP3A4. The decrease in RXR protein (Fig. 6A) was far greater than the decrease in RXR mRNA following treatment with arsenite (Fig. 6B). This suggests that arsenite not only decreases the expression of RXR mRNA, but also has a post-transcriptional effect. One possibility is that arsenite enhances the degradation of RXR protein. For example, it has been reported that tetradecanoylphorbol-1,3-acetate can induce the degradation of RXR in gastric cells via the proteasome pathway (Ye et al., 2003). We are currently investigating whether protein degradation plays a role in the arsenite-mediated decrease in RXR protein.
Our findings that arsenite decreased nuclear RXR protein (Fig. 6A) and CYP3A4 mRNA (Figs. 1B, 2A, and 3A) are similar to a recent report that lipopolysaccharide decreases RXR protein levels in livers of mice, in parallel with decreases in Cyp3a11 mRNA (Ghose et al., 2004). Nuclear translocation affects gene expression by some nuclear receptors. In that study, lipopolysaccharide changed the localization of RXR from the nucleus to the cytoplasm (Ghose et al., 2004). In our studies, arsenite did not change the RXR localization at 24 h (Fig. 6A) but may have had an effect at an earlier time. Nuclear translocation of PXR in response to ligand has been reported to occur following treatment of mice with 5-pregnen-3ß-ol-20-one-16-carbonitrile (Squires et al., 2004). A time course of 5-pregnen-3ß-ol-20-one-16-carbonitrile treatment in mice demonstrated that the increase in hepatic nuclear PXR protein was transient, returning to basal levels by 17 h (Squires et al., 2004). This decrease may have been caused by decreases in the hepatic level of inducer over time. In our studies, a 24-h treatment with Rif had no effect on nuclear localization of PXR (Fig. 4B), even though CYP3A4 mRNA was induced. Since these cells were isolated from livers of humans pre-exposed to numerous drugs (Table 1), including many CYP3A4 inducers, PXR may have already been activated and translocated to the nucleus. Arsenite had no effect on nuclear levels of PXR (Fig. 4B). Since the cultured hepatocytes were continuously exposed to inducer (Rif) for 24 h, arsenite may have had a transient effect on the accumulation of PXR protein in the nucleus. Nevertheless, our results do demonstrate that after 24 h of treatment with inducer, when CYP3A4 mRNA levels are increased, arsenite does not affect CYP3A4 transcription by causing a translocation of PXR or RXR from the nucleus to the cytoplasm.
It is well known that arsenite interacts with critical cysteine residues of many intracellular proteins (Del Razo et al., 2001). Therefore, arsenite may also bind to the nuclear receptors involved in expression of CYP3A4, resulting in decreases in their ability to bind ligands or other transcription factors required for up-regulation of CYP3A4. Our finding that arsenite caused a large decrease in the responsiveness of human PXR to Rif (Fig. 5B) supports this hypothesis. However, if the arsenite-mediated decreases in RXR expression observed in cultured human hepatocytes also occurred in cultured rat hepatocytes, the decrease in human PXR activity (Fig. 5B) might arise from an insufficient amount of RXR for heterodimerization. Therefore, we plan to investigate whether arsenite decreases RXR in primary rat hepatocytes. Furthermore, arsenite has been shown to down-regulate telomerase gene expression in human leukemia cells, partly through inhibition of the DNA binding activity of Sp1, which contains redox-sensitive cysteine residues (Chou et al., 2001). Although we found that arsenite does not decrease levels of Sp1 in cultured human hepatocytes (Fig. 4), arsenite may inhibit the ability of Sp1 to bind to DNA. Additionally, there are numerous other transcription factors that may be affected by arsenite treatment. For example, hepatocyte nuclear factors 3 and 4, and CCAAT/enhancer binding protein have been shown to play a role in the basal or xenobiotic-mediated regulation of CYP3A4 transcription (Jover et al., 2001; Bombail et al., 2004). Therefore, further work is needed to establish whether arsenite affects expression or activity of these or other transcription factors that are involved in CYP3A4 regulation.
In human hepatocyte cultures, glucocorticoids increase the level of CAR and PXR mRNAs and proteins, leading to the potentiation of xenobiotic-mediated induction of CYP3A4 (Pascussi et al., 2000a,b). Since we found that arsenite does not affect PXR mRNA or protein levels (Fig. 4), it is likely that the action of arsenite on expression of CYP3A4 is independent of the glucocorticoid receptor (GR). In contrast, arsenite has been shown to inhibit GR-mediated transcription in hepatoma cells (Kaltreider et al., 2001; Bodwell et al., 2004). However, another study demonstrated that exposure to arsenite leads to transactivation of GR elements in mouse epidermal JB6 cells (Huang et al., 2001). Therefore, the effect of arsenite may depend on the cell type.
In our previous studies, using primary rat hepatocytes, treatment with arsenite (5 µM) decreased DEX- but not PB-mediated induction of CYP3A23 mRNA, even though arsenite decreased both DEX- and PB-mediated increases in CYP3A protein (Jacobs et al., 1999). However, the extent of the decrease in CYP3A23 mRNA induced by DEX was small (25%) compared with the much larger decreases (up to 95%) observed with CYP3A4 mRNA in human hepatocytes (Figs. 1B and 2A). Therefore, it appears that CYP3A transcriptional regulation in human hepatocytes is more sensitive to arsenite than in rat hepatocytes.
In summary, we have shown that low doses of arsenite (2.5 or 5 µM) prevented induction of CYP3A4 mRNA, protein, and catalytic activity in primary cultures of human hepatocytes. Arsenite also decreased expression of CYP3A4 in hepatocytes not treated with inducers. The substantial decreases in CYP3A4 mRNA suggest that the mechanism of action of arsenite involves suppressed transcription of the CYP3A4 gene. This possibility is supported by our findings that arsenite caused a dramatic decrease in nuclear protein levels of RXR and responsiveness of PXR to inducer. We are currently evaluating the mechanism by which arsenite decreases RXR and the effect of arsenite on other genes regulated by this critical nuclear receptor.
|
Footnotes |
|---|
Article, publication date, and citation information can be found at http://dmd.aspetjournals.org.
ABBREVIATIONS: P450, cytochrome P450; Rif, rifampicin; PB, sodium phenobarbital; RXR, retinoid X receptor ; PXR, pregnane X receptor; CAR, constitutive androstane/active receptor; As, sodium arsenite; DEX, dexamethasone; FeNTA, ferric nitrilotriacetic acid; SXR, steroid and xenobiotic receptor; MOPS, 3-(N-morpholino)propanesulfonic acid; PCR, polymerase chain reaction; RT-PCR, reverse transcription-PCR; PAGE, polyacrylamide gel electrophoresis; ROS, reactive oxygen species; TBARS, thiobarbituric acid reactive substances; GR, glucocorticoid receptor.
Address correspondence to: Dr. Jacqueline Sinclair, Research 151, Veterans Administration Medical Center, White River Junction, VT 05009. E-mail: JSINC{at}dartmouth.edu
|
References |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
, HNF3 and PXR in the basal- and xenobiotic-mediated regulation of the CYP3A4 gene. Drug Metab Dispos 32: 525–535.-tocopherol in the MGC-803 cells. Eur J Pharmacol 448: 11–18.[CrossRef][Medline]
: a novel mechanism for reduced hepatic gene expression in inflammation. Nucl Recept 2: 4.[CrossRef][Medline]
by TPA through proteasome pathway in gastric cancer cells. World J Gastroenterol 9: 1915–1919.[Medline]
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
