Abstract
Assessment of foreign chemical inducibility on CYP3A4 is necessary to optimize drug therapies. The properties of chemicals such as pesticides, however, are not well investigated. In the present study, properties of various pesticides on human CYP3A4 induction have been tested using HepG2-derived cells stably expressing the CYP3A4 promoter/enhancer (3-1-10 cells) and the human pregnane X receptor (hPXR)-small interfering RNA (siRNA) system. Among the examined pesticides, 13 pesticides were observed to activate the CYP3A4 gene. Surprisingly, pyributicarb was found to increase the CYP3A4 reporter activity at 0.1 to 1 μM more strongly than typical CYP3A4 inducer rifampicin. Expression of hPXR-siRNA clearly diminished the pyributicarb-stimulated CYP3A4 reporter activity in 3-1-10 cells and decreased the endogenous CYP3A4 mRNA levels in HepG2 cells. Pyributicarb caused enhancement of CYP3A4-derived reporter activity in mouse livers introduced with hPXR by adenovirus. These results indicate pyributicarb as a potent activator of CYP3A4 gene, suggesting the existence of pesticides leading to CYP3A4 induction in our environment.
Activities of drug-metabolizing enzymes often show large interindividual differences, which leads to interindividual variation in the efficacy of therapeutic drugs and/or severities of their adverse effects. The variation results not only from genomic backgrounds but also physiological and environmental factors such as nutrition and disease state and exposure to foreign chemicals, including therapeutic drugs. CYP3A4 is an enzyme showing large interindividual differences (Wolbold et al., 2003). This form is a major cytochrome P450 contained in human livers (Gonzalez, 1992) and involved in metabolisms of about a half of clinically used drugs (Li et al., 1995). The exact mechanism causing the differences in the expression levels of CYP3A4 in human livers remains unclear, although genetic variation is considered as a reason for the difference of CYP3A4 expression. However, endogenous and exogenous chemical-mediated transactivation of the CYP3A4 gene may be one of the underlying mechanisms.
To develop a screening system that detects CYP3A4 gene activators with high sensitivity, we established the HepG2-derived cell lines stably expressing the CYP3A4-luciferase reporter gene (bases –362 to +11 and –7836 to –7008) (Noracharttiyapot et al., 2006). The reporter gene contains pregnane X receptor (PXR) binding sites as reported previously (Hashimoto et al., 1993; Barwick et al., 1996; Goodwin et al., 1999). PXR is widely known as a major transcription factor mediating CYP3A4 induction (Moore and Kliewer, 2000) and interacts with a wide variety of therapeutic drugs, including rifampicin (RIF) and clotrimazole (CTZ) (Bertilsson et al., 1998; Blumberg et al., 1998; Lehmann et al., 1998). Among the established cell lines, 3-1-10 showed the highest response to RIF (Noracharttiyapot et al., 2006). Thus, this cell line was chosen to assess chemical inducibility of the CYP3A4 gene.
In addition to PXR, other receptors such as constitutive androstane receptor (CAR) (Goodwin et al., 2002) and vitamin D receptor (VDR) (Schmiedlin-Ren et al., 2001; Thummel et al., 2001) have also been reported to activate the transcription of CYP3A4 gene. Understanding whether a chemical can activate nuclear receptors is an inevitable step to predict the inducibility of CYP3A4. However, the involvement of PXR in the chemical-mediated CYP3A4 induction is not precisely determined using the PXR binding sites because CAR and VDR are capable of interacting with the sites. Recently, introduction of specific short nucleotides [small interfering RNA (siRNA)] in cells has been shown to specifically knock down the target gene expression (Elbashir et al., 2001). However, human PXR (hPXR)-siRNA is scarcely used to understand the mechanisms of the endogenous and exogenous chemical-mediated activation of the CYP3A4 gene. The development of an effective hPXR-siRNA probe may stimulate clear identification of PXR involvement in chemical-mediated CYP3A4 induction.
Pesticides are among those xenobiotic candidates that may cause interindividual variation of CYP3A4 expression because they are widely used for animals, insects, and plants and are released into the environment. They can be taken into human bodies via food, water, and air. There is a possibility of the existence of a chemical modulating the CYP3A4 expression through hPXR activation. Although pesticides have been reported to activate PXR or CAR (Wyde et al., 2003; Jacobs et al., 2005; Lemaire et al., 2006), pesticide-mediated activations of CYP3A4 gene are not well assessed.
In this study, we have developed an adenovirus vector expressing hPXR-siRNA that is able to specifically knock down hPXR expression. A combination of the 3-1-10 cell line and the hPXR-siRNA system enabled us to efficiently identify a potential hPXR activator causing CYP3A4 induction. These results suggest the possibility that some pesticides contaminated in the environment potentially induce CYP3A4 expression through hPXR.
Materials and Methods
Materials. Protein assay dye reagent concentrate was purchased from Bio-Rad Laboratories (Hercules, CA). Restriction enzymes were purchased from New England BioLabs (Beverly, MA). RIF, CTZ, and 1α,25-dihydroxyvitamin D3 (VD3) were purchased from Sigma-Aldrich (St. Louis, MO). All the pesticides were purchased from Wako Pure Chemical Industries, Ltd. (Osaka, Japan).
Animal Treatment and Cell Culture. Male ICR mice (5 weeks old) were purchased from Charles River Laboratories Japan, Inc. (Yokohama, Japan) and fed standard rodent chow (CE-2; CLEA, Tokyo, Japan) and water ad libitum. After 18-h fasting, mice were injected i.v. with adenovirus [4.0 × 109 50% titer culture infectious dose (TCID50)/mouse]. Three days after the infection, vehicle (0.5% methyl cellulose/saline) or pyributicarb (100 mg/kg/day) was administered p.o. for 2 consecutive days. Animals were killed 20 h after the last dose. Cell culture and adenovirus infection were carried out as described previously (Furukawa et al., 2002). Methylthiazole tetrazolium incorporation assay was carried out according to the method described by Jaiswal et al. (2004).
Construction of Recombinant Adenovirus.AdhPXR-siRNA. Human U6 small nuclear RNA gene promoter was used for the expression of siRNA. The U6 small nuclear RNA promoter was amplified by polymerase chain reaction (PCR) with primers 5′-CGCTCGAGCCGACGCCGCCATCTC-3′ and 5′-GCGTCTAGAGTTAACAAGGCTTTTCTCCAAGGG-3′. The PCR product was digested with XhoI and XbaI and cloned into the same restriction sites of the promoterless vector pShuttle (pShuttle-U6). The DNA encoding hPXR-specific siRNA was amplified by PCR with primers 5′-GCGGTCGACGAGCTGTGTCAACTGAGATTCTTCAAGAGA-3′ and 5′-GCGAGATCTAAAAAGAGCTGTGTCAACTGAGATTCTCTCTTGAAGA-3′, and the PCR product was digested with SalI and BglII and ligated into the same restriction sites of pShuttle-U6 (pShuttle-U6-hPXR-siRNA). To obtain an adenovirus expressing hPXR-siRNA, the homologous recombination was utilized in BJ5183 cells transfected with pShuttle-U6-hPXR-siRNA linearized with PmeI and adenoviral backbone plasmid pAdEasy-1 (Quantum Biotechnologies, Laval, QC, Canada). The resultant DNA was linearized with PacI, and human embryonic kidney 293 cells were transfected with the DNA using CellPhect Transfection Kit (GE Healthcare, Piscataway, NJ). Target sequences for hPXR-siRNA were selected manually on a random basis. The adenovirus expressing hPXR-siRNA used in this study showed the highest efficiency among four siRNA tested on the knockdown of hPXR mRNA levels (data not shown).
AdCYP3A4-362-7.7k. The CYP3A4 enhancer region (from –7836 to –7208) was amplified by PCR with oligonucleotides 5′-CGACGCGTCTAGAGAGATGGTTCATTCC-3′ and 5′-GCAGATGTAATGATCTCGTCAACAGG-3′. The PCR product was digested with MluI and BglII and ligated into the same restriction sites of the pGLCYP3A4-362 (Furukawa et al., 2002). AdCYP3A4-362-7.7k was then obtained using the method reported previously (Furukawa et al., 2002).
AdhPXR was reported previously (Noracharttiyapot et al., 2006). AdCont (AxCALacZ), which expresses β-galactosidase, was provided by Dr. Izumi Saito (Tokyo University, Tokyo, Japan). We showed that this adenovirus did not affect either hPXR mRNA levels or the CYP3A4 induction by RIF in HepG2 cells as an adenovirus expressing nontargeting siRNA, mouse PXR (mPXR)-targeting siRNA, which does not match to any human mRNA (data not shown). Thus, we used AdCont as a control for both overexpression and RNA knockdown experiments. The titer of adenoviruses, TCID50, was determined in human embryonic kidney 293 cells. The value of TCID50 was reported to be almost equivalent to that of plaque-forming unit (Kanegae et al., 1994). Multiplicity of infection (MOI) was calculated by dividing TCID50 with the number of cells.
Detection of mRNA. Total RNA were extracted from HepG2 cells using the acid guanidine thiocyanate-phenol-chloroform method. The cDNA was reverse-transcribed from total RNA (3 μg) with Ready-To-Go (GE Healthcare). PXR, CYP3A4, and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) mRNA levels were determined by conventional PCR with TaqDNA polymerase (ABgene, Epsom, UK) as follows. After initial denaturation at 94°C for 5 min, DNA were amplified for 30 (PXR and GAPDH) or 39 cycles at 94°C for 30 s, 57°C for 15 s (PXR and GAPDH) or 63°C for 5 s (CYP3A4), 72°C for 30 s, with a final extension period at 72°C for 7 min. CYP3A4 and GAPDH mRNA levels were also determined by real-time PCR using Platinum qPCR (Invitrogen, Carlsbad, CA) with ABI PRISM 7000 (Applied Biosystems, Foster City, CA) following the manufacturers' instructions. The sequences of primers and probes are shown in Table 1.
Luciferase Assay.In vitro assay. The luciferase assay with cells was performed as reported previously (Furukawa et al., 2002). In vivo assay: Mouse livers were homogenized in a double volume of 25 mM Tris-HCl buffer (pH 7.4), and the homogenate was centrifuged at 9000g for 20 min. The supernatant was centrifuged at 105,000g for 60 min. The resultant supernatant was used for the luciferase assay with luciferase assay system (Promega, Madison, WI).
Preparation of PXR Antibody. A portion of hPXR (amino acids 129–235) was expressed as a histidine-tagged protein with QIAexpressionist (Qiagen, Chatsworth, CA) according to the manufacturer's protocol. The expressed protein was purified with Ni-NTA column (Qiagen) under the denatured condition, and then the histidine tag was removed with enterokinase. Rabbit was immunized with the recombinant protein, and antisera were affinity-purified with the same antigen immobilized to CNBr-activated Sepharose-4B (GE Healthcare). The purified antisera detected both the antigen and hPXR protein expressed in COS-1 cells at expected molecular sizes. Apparent molecular mass of the detected hPXR proteins in the HepG2 extract is 51.8 kDa (cf. Fig. 3B), which is similar to both that calculated from amino acids and that reported previously (Lehmann et al., 1998).
Immunoblot Analysis. Nuclear extracts were prepared from 3-1-10 cells according to the method reported previously (Ogino et al., 1999), separated by SDS-polyacrylamide gel electrophoresis, and transferred to a nitrocellulose membrane. The membrane was immunostained with the anti-PXR antibody and alkaline phosphatase-conjugated goat anti-rabbit IgG, and signals were visualized with 5-bromo-4-chloro-3-indolylphosphate and nitro blue tetrazolium.
Statistical Analysis. GraphPad Prism version 4.00 (GraphPad Software, San Diego, CA) was used for all the statistical analyses.
Results
Transcriptional Activation of CYP3A4 Reporter Gene by Pesticides. We investigated whether pesticides activate the CYP3A4 transcription using the HepG2-derived cells stably expressing the CYP3A4 reporter gene (3-1-10 cells). The cells were treated with 0.3 to 30 μM pesticides for 48 h. Then reporter activities were determined (Fig. 1). Seventeen pesticides were tested in this study, which included five insecticides [chlorpyrifos, O,O-diethyl O-(3-methyl-4-nitrophenyl) phosphorothioate, etoxazole, O-ethyl O-4-nitrophenyl phenylphosphonothioate (EPN), and isoxathion], five fungicides (tetrachloroisophthalonitrile, echlomezol, isoprothiolane, mepronil, and flutolanil) and seven herbicides (chloroneb, 2-chloro-4,6-bis(ethylamino)-1,3,5-triazin, propyzamide, pyributicarb, dithiopyr, isofenphos, and dymron). RIF was used as a positive control. Among the insecticides, EPN and isoxathion showed strong activation of the CYP3A4 reporter (the maximum activation by EPN and isoxathion was 30.2- and 16.2-fold, respectively, both at 10 μM). For the fungicides, mepronil and flutolanil increased the activity (the maximum activation by mepronil and flutolanil was 14.6- and 20.5-fold, respectively, both at 30 μM). In herbicides, pyributicarb, dithiopyr, isofenphos, and dymron were found to activate strongly the CYP3A4 reporter (the maximum activation by pyributicarb, dithiopyr, isofenphos, and dymron was 22.1-fold at 3 μM, 11.4-fold at 10 μM, 18.5-fold at 10 μM, and 12.2-fold at 10 μM, respectively). Surprisingly, the CYP3A4 reporter activities were higher with pyributicarb than with RIF at 0.3 and 1 μM (Fig. 1). Thus, we compared the dose-dependent change of the activation of the CYP3A4 reporter gene by pyributicarb with those by typical hPXR activators (RIF and CTZ) in HepG2 cells (Fig. 2). Pyributicarb and RIF showed EC50 values of 0.21 and 2.2 μM, respectively, for CYP3A4 reporter activities in the transiently transfected HepG2 cells. Pyributicarb was a more potent activator of the CYP3A4 reporter than RIF and CTZ.
Role of hPXR in the Pesticide-Mediated Activation of the CYP3A4 Promoter. To verify the mechanism that pesticides activate the CYP3A4 reporter gene, we have constructed a recombinant adenovirus expressing hPXR-siRNA. The 3-1-10 cells were infected with AdhPXR-siRNA and treated with RIF, CTZ, or VD3 to confirm the specificity of this system. As shown in Fig. 3A, introduction of the adenovirus drastically diminished the RIF- and CTZ-mediated activation of the reporter gene but not that induced by VD3. Infection with AdhPXR-siRNA drastically decreased the amount of nuclear hPXR proteins (Fig. 3B). Effects of hPXR-siRNA expression on the endogenous CYP3A4 mRNA levels in HepG2 cells were also investigated (Fig. 3C). Introduction of hPXR-siRNA decreased hPXR mRNA levels. This treatment also inhibited strongly the RIF-stimulated, but not the VD3-stimulated, increase of the endogenous CYP3A4 mRNA levels.
Using the hPXR-siRNA system, we found that the pyributicarb-, isoxathion-, and EPN-mediated activation of the CYP3A4 reporter was decreased by the introduction of AdhPXR-siRNA as in the case of RIF in 3-1-10 cells (Fig. 4A). Effects of the hPXR-siRNA expression on endogenous CYP3A4 mRNA levels in HepG2 cells were also investigated (Fig. 4B). Pyributicarb enhanced levels of CYP3A4 mRNA in HepG2 cells as RIF did. The introduction of hPXR-siRNA attenuated the pesticide-mediated increase of the mRNA levels. Similar decreases were also observed with isoxathion and EPN.
We further investigated whether pyributicarb activated the CYP3A4 promoter by in vivo reporter assay. Mice were infected with AdCYP3A4-362-7.7k and AdCont or AdhPXR, and then treated with pyributicarb. As shown in Fig. 5, mPXR activator pregnenolone 16α-carbonitrile (PCN) treatment increased CYP3A4 reporter activity (53-fold). On the other hand, pyributicarb treatment did not increase reporter activity. Introduction of hPXR increased CYP3A4 reporter activity (153-fold). Pyributicarb treatment increased CYP3A4 reporter activity (623-fold) with introduction of hPXR, although it did not increase reporter activity without hPXR (Fig. 5).
Discussion
With the advantage of the 3-1-10 cell, which stably expresses a CYP3A4 reporter gene, we have examined properties of 17 functionally and structurally different pesticides for induction of CYP3A4 expression. Among them, 13 pesticides activated the CYP3A4 reporter gene (Fig. 1). EPN, isoxathion, and pyributicarb also increased the levels of endogenous CYP3A4 mRNA in HepG2 cells (Fig. 4B). Pyributicarb was found to activate the CYP3A4 reporter more strongly than RIF at submicromolar concentrations in both 3-1-10 and HepG2 cells (Figs. 1 and 2), indicating a possibility that this pesticide functions as a potent hPXR activator in vivo. However, at high concentrations, pyributicarb-induced activation was diminished. This may result from its characteristics as an herbicide. Viability of pyributicarb-treated cells, estimated by methylthiazole tetrazolium incorporation assay, was decreased to 37% that of vehicle-treated cells after 48-h treatment with 30 μM pyributicarb (data not shown). Nevertheless, utilization of a stable cell line 3-1-10 cells enabled us to identify herbicides that may cause CYP3A4 induction.
Chemical-induced activation of CYP3A4 gene is mainly mediated by PXR heterodimerized with retinoid X receptor through their binding to the CYP3A4 5′-flanking region (Bertilsson et al., 1998; Blumberg et al., 1998; Lehmann et al., 1998). In addition to PXR, other nuclear receptors, including CAR and VDR, are also known to activate the CYP3A4 gene. To determine whether hPXR is involved in the pesticide-mediated activation of the CYP3A4 gene, we developed an AdhPXR-siRNA system and applied this to the reporter assays. Introduction of AdhPXR-siRNA drastically decreased the amount of nuclear hPXR proteins and resulted in the diminishment of the RIF- and CTZ-mediated activation but not that mediated by VD3 (Fig. 3). When AdhPXR-siRNA was introduced into the 3-1-10 cells, the pyributicarb-, isoxathion-, and EPN-mediated activation of the CYP3A4 reporter was drastically attenuated (Fig. 4A). Furthermore, the introduction of AdhPXR-siRNA into HepG2 cells prevented the pesticide-induced increase of endogenous CYP3A4 mRNA levels (Fig. 4B). These results suggest that pyributicarb, isoxathion, and EPN enhance the transactivation of the CYP3A4 gene through hPXR activation.
With the use of the adenovirus system, which is applicable to in vivo living organisms, we further investigated whether pyributicarb activated the CYP3A4 reporter gene in mice. Because there are differences of ligand specificity between hPXR and mPXR (Ostberg et al., 2002), we determined the reporter activity both in the absence and presence of hPXR. In the absence of exogenous hPXR, pyributicarb did not increase the CYP3A4 reporter activity. Introduction of hPXR itself increased CYP3A4 reporter activity, and pyributicarb treatment further enhanced reporter activity (Fig. 5). In addition, a potent mPXR ligand, PCN, activated the CYP3A4 reporter (Fig. 5), which excludes the possibility that the ligand-activated mPXR may not bind to the CYP3A4 promoter. These results indicate that pyributicarb is a potent hPXR activator and a weak, if any, ligand of mPXR.
In conclusion, we have investigated the pesticide inducibility of CYP3A4 gene expression using a HepG2-derived cell line expressing the CYP3A4 reporter gene and the AdhPXR-siRNA system, and identified plural hPXR activators among pesticides. Further studies, including the determination of pesticide exposure levels, are necessary to assess the in vivo significance of their human CYP3A4 induction. The screening system presented in this study is useful to understand the molecular mechanism of the xenobiotic-induced expression of CYP3A4 gene, as well as to estimate the properties of chemicals on CYP3A4 induction.
Acknowledgments
We thank Dr. Izumi Saito (Tokyo University, Tokyo, Japan) for providing the AxCALacZ.
Footnotes
-
This work was supported in part by grant-in-aid from the Ministry of Education, Culture, Sports, Sciences, and Technology and from the Ministry of Health, Labour, and Welfare of Japan, and by Research on Health Sciences focusing on Drug Innovation from The Japan Health Sciences Foundation.
-
Article, publication date, and citation information can be found at http://dmd.aspetjournals.org.
-
doi:10.1124/dmd.106.013144.
-
ABBREVIATIONS: PXR, pregnane X receptor; RIF, rifampicin; CTZ, clotrimazole; CAR, constitutive androstane receptor; VDR, vitamin D receptor; siRNA, small interfering RNA; hPXR, human pregnane X receptor; VD3,1α,25-dihydroxyvitamin D3; TCID50, 50% titer culture infectious dose; PCR, polymerase chain reaction; mPXR, mouse pregnane X receptor; MOI, multiplicity of infection; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; EPN, O-ethyl O-4-nitrophenyl phenylphosphonothioate; PCN, pregnenolone 16α-carbonitrile; DMSO, dimethyl sulfoxide; ANOVA, analysis of variance; RT-PCR, reverse transcription-polymerase chain reaction.
- Received September 28, 2006.
- Accepted February 7, 2007.
- The American Society for Pharmacology and Experimental Therapeutics