Abstract
The role of the glucocorticoid receptor (GR) and pregnane X receptor (PXR) in the regulation of female-predominant expression of mouse CYP3A44 by glucocorticoid hormones was evaluated using a primary culture of female mouse hepatocytes, as the expression was suppressed in adrenalectomized female mice, restored by dexamethasone (DEX) treatment and was not detected in male mouse livers. Glucocorticoid hormones, such as DEX, hydrocortisone, and corticosterone, 11β-[4-dimethylamino] phenyl-17β-hydroxy-17-[1-propynyl]estra-4,9-diene-3-one (RU486), antagonists for GR and an agonist for PXR, and rifampicin, an agonist for PXR, were chosen to investigate the relationship of GR/PXR activation and Cyp3a44 gene expression. Glucocorticoid-inducible expression of CYP3A44 was not suppressed but rather was increased by RU486. Treatment of GR expression plasmid-transfected hepatocytes with DEX concentration dependently enhanced the expression of PXR as well as CYP3A44 mRNAs. A synergistic effect of DEX at submicromolar concentrations and rifampicin is observed. Furthermore, transfection of PXR and retinoid X receptor-α (RXRα) also showed prominent induction of CYP3A44 mRNA by DEX. These results suggest that DEX plays a dual role in CYP3A44 expression: first, direct activation of the Cyp3a44 gene by the PXR-RXRα complex, and, second, indirect activation of the Cyp3a44 gene through the induction of PXR gene expression by the GR pathway.
Cytochrome P450 proteins form a superfamily of heme-containing enzymes involved in the oxidative metabolism of both endogenous and exogenous compounds; the former include steroids, fatty acids, retinoids, bile acids, and others, and the latter include drugs and environmental pollutants (Gonzalez, 1991). The CYP3A subfamily represents the most abundant cytochromes P450 in adult human liver, comprising approximately 30% of the total content. The human CYP3A subfamily comprises four isoenzymes, CYP3A4, CYP3A5, CYP3A7, and CYP3A43, which show variable expressions in the population. Among them, the CYP3A4 isoform is the most prevalent in adults. It is estimated that approximately 50% of currently marketed drugs are metabolized by CYP3A4 (Bertz and Granneman, 1997).
Recently, several members of the nuclear hormone receptor superfamily such as constitutive androstane receptor (CAR), pregnane X receptor (PXR) (Bertilsson et al., 1998; Blumberg et al., 1998; Kliewer et al., 1998; Lehmann et al., 1998), vitamin D receptor (Drocourt et al., 2002), and glucocorticoid receptor (GR) (Schuetz et al., 1996; Pereira et al., 1998) have been shown to be responsible for endobiotic- and xenobiotic-mediated expression of CYP3A genes. The GR is activated upon binding of glucocorticoids and then regulates gene transcription either actively or repressively (Beato et al., 1995; Gupta and Lalchhandama, 2002). Several lines of evidence support the role of the GR in CYP gene regulation. For instance, the human CYP3A5 gene promoter contains two glucocorticoid response elements (GRE), separated by 160 bp, which confer the responsiveness of the reporter gene to glucocorticoid in HepG2 cells (Schuetz et al., 1996). Furthermore, it has been reported that GR binds to GRE present in the rat CYP3A1 gene, suggesting that cooperation of the upstream GRE and downstream elements may be required for the maximal response of CYP3A to glucocorticoids (Pereira et al., 1998). In addition to the GR, the increased expression of CYP3A mRNA is also mediated via the PXR. On activation by a xenobiotic ligand, the PXR dimerizes with retinoid X receptor-α (RXRα), and the heterodimer formed binds to their respective response elements to induce CYP3A expression (Mangelsdorf et al., 1995; Bourguet et al., 2000). Functional cross-talk between the GR- and PXR-signaling pathways has been reported in human CYP3A4 and rat CYP3A23 gene expression (Huss and Kasper, 2000; Pascussi et al., 2000, 2001); however, the identity of virtual controlling of the activity of PXR at low ligand concentration is still unclear. If a similar regulation pathway were found in other laboratory animal species, the model would be valuable for more comprehensive understanding of orthologous cytochrome P450 in humans.
With regard to the CYP3A subfamily in mice, CYP3A44, female-predominant CYP3A mRNA, was isolated, and it was found that the expression is dependent on the feminine plasma growth hormone profile (Sakuma et al., 2002). Furthermore, we also found that glucocorticoids increased CYP3A44 expression in cultured hepatocytes (unpublished results); however, the role of nuclear hormone receptors in the regulation of Cyp3a44 gene expression has not been extensively determined. Observation suggests that pregnenolone-16α-carbonitrile- and dexamethasone (DEX)-induced CYP3A44 mRNA expression is PXR-dependent in male mice. On the other hand, pregnenolone-16α-carbonitrile and DEX down-regulated CYP3A44 expression in female PXR-null mice (Anakk et al., 2007). Given these findings, it is suggested that gender, the xenobiotic activator, and the nuclear receptor comprehensively act to control Cyp3a44 gene expression; therefore, the role of the GR or the PXR in the overall regulation of mouse Cyp3a44 gene expression was independently explored.
In the present study, we investigated the role of the GR and the PXR in the regulation of Cyp3a44 gene expression using a primary culture of mouse hepatocytes. The results suggest that DEX plays a dual role in CYP3A44 expression: first, direct activation of the Cyp3a44 gene by the PXR-RXRα complex, and, second, indirect activation of the Cyp3a44 gene through induction of PXR gene expression by the GR pathway.
Materials and Methods
Materials. Materials for culturing hepatocytes were purchased from Wako Pure Chemicals (Osaka, Japan), Invitrogen (Carlsbad, CA), and Sigma-Aldrich (St. Louis, MO). Percoll was obtained from GE Healthcare (Little Chalfont, Buckinghamshire, UK). TransPass D1 Transfection Reagent was from New England Biolabs (Hercules, CA). Dexamethasone, hydrocortisone, corticosterone, and RU486 were obtained from Sigma-Aldrich. Rifampicin was obtained from Wako Pure Chemicals (Osaka, Japan). A TaKaRa RNA PCR Kit (AMV) version 3.0 was obtained from TaKaRa Shuzo (Kyoto, Japan). All other laboratory chemicals were of the highest grade commercially available.
Animals. Mice were housed in the University of Toyama's Animal Center under the supervision of certified laboratory veterinarians and were treated according to a research protocol approved by the University's Institutional Animal Care and Use Committee. Animals were allowed food and water ad libitum and were subjected to a 12-h light/dark cycle. C57BL/6 mice of both sexes were purchased from Sankyo Experimental Animals (Tokyo, Japan). Four-week-old C57BL/6 mice were adrenalectomized or sham-operated and killed 5 days later. Some adrenalectomized or sham-operated mice received a s.c. administration of DEX at 10 mg/kg/day for the last 3 days. The liver was excised immediately after death and used for the preparation of total RNA.
Preparation of Primary Hepatocyte Cultures. Eight-week-old female ddY mice were purchased from Japan SLC, Inc. (Shizuoka, Japan). The livers were perfused with collagenase-containing Hanks' solution, and viable hepatocytes were isolated by Percoll isodensity centrifugation as described previously (Nemoto and Sakurai, 1995) and seeded in dishes at a density of 2 × 106 cells/4 ml/60 mm. The Waymouth medium did not contain phenol red, a pH indicator, to exclude estrogen-like action. The culture dishes were maintained at 37°C in a CO2-humidified incubator. The medium was renewed 24 h after seeding, and treatment with either DEX, hydrocortisone, corticosterone, RU486, or rifampicin was started 1 day after the medium change. Each chemical was dissolved in dimethyl sulfoxide to a 0.1% final concentration. The cells were harvested 24 h later to prepare the total RNA fraction.
Plasmids. The GR expression plasmid was generated by replacing the DNA fragment between NheI and XbaI sites containing the coding sequence of Renilla luciferase of the pRL-SV40 vector (Promega, Madison, WI) with the 2385-bp cDNA fragment involving the entire coding region (2379 bp) and 6-bp 3′-noncoding regions of mouse GR. Both expression plasmids of PXR and the RXRα were constructed using the same strategy. The PXR expression plasmid contains the 1312-bp cDNA fragment with the entire coding region (1296 bp) and both 6-bp 5′- and 10-bp 3′-noncoding regions of mouse PXR. The RXRα expression plasmid contains the 1409-bp cDNA fragment with the entire coding region (1404 bp) and 5-bp 5′-noncoding regions of mouse RXRα.
Transfection of the Nuclear Receptor Expression Plasmid into Hepatocytes in Cultures. Mouse hepatocytes were cultured in Waymouth medium and transfected using TransPass D1 Transfection Reagent (New England Biolabs). Transfection mixtures consisted of Waymouth medium, empty plasmid, or nuclear receptor expression plasmid and TransPass D1 at 2 ml, 5 μg, and 5 μl, respectively. Transfection continued for 3 h, and the medium was changed. The cells were treated with DEX at various concentrations after a further 24-h incubation. Total RNA was prepared from other 24-h treated cells.
Real-Time RT-PCR. Hepatic total RNA was prepared from hepatocytes as described previously (Nemoto and Sakurai, 1995). Semiquantitative RT-PCR of CYP3A44 and GAPDH with 32P-radiolabeled primers was performed using a TaKaRa RNA PCR Kit (AMV) version 3.0 as described previously (Sakuma et al., 2000, 2002). Quantitative real-time RT-PCR was performed using a TaKaRa RNA PCR Kit (AMV) version 3.0 in combination with a gene-specific TaqMan MGB Gene Expression Detection Kit or SYBR Green reagent. The forward primer, reverse primer, and TaqMan MGB probe of the TaqMan MGB Gene Expression Detection Kit for CYP3A44, designed by us with the assistance of Primer Express software, were 5′-GAAACTGCAGGCAGAGACCATA-3′, 5′-TTTCTTACAGACTCTCTCTCAAGTCTAGTAACAAT-3′, and 5′-FAM-AATAAGGCAACTCCCACCTG-MGB-3′, respectively. For CYP3A41, the forward primer, reverse primer, and TaqMan MGB probe of the TaqMan MGB Gene Expression Detection Kit, designed by us, were 5′-GCCAAAGGGATTTTAAGAGTTGACT-3′, 5′-GGTGTCAGGAATGGAAAAAGTACA-3′, and 5′-FAM-ATCCTTTGGTCTTCTCAG-MGB-3′, respectively. PXR or GAPDH cDNA was detected with SYBR Green reagent and gene-specific primer sets. The forward and reverse primers for PXR were 5′-GCCAAAGGGATTTTAAGAGTTGACT-3′ and 5′-GGTGTCAGGAATGGAAAAAGTACA-3′, respectively, and for GAPDH were 5′-TCCACTCACGGCAAATTCAACG-3′ and 5′-TAGACTCCACGACATACTCAGC-3′, respectively. PCR conditions were as follows. For Cyp3a44, initial denaturation at 95°C for 10 min, denaturation at 95°C for 15 s, and extension at 62°C for 1 min were performed. PCR, denaturation, and extension were repeated for 60 cycles. For Cyp3a41, initial denaturation at 95°C for 4 min, denaturation at 95°C for 15 s, and extension at 60°C for 1 min were performed. PCR, denaturation and extension were repeated for 50 cycles. For PXR, initial denaturation at 95°C for 4 min, denaturation at 95°C for 15 s, annealing at 64°C for 15 s, and extension at 72°C for 1 min were performed. PCR, denaturation, annealing, and extension were repeated for 40 cycles. For GAPDH, initial denaturation at 95°C for 4 min, denaturation at 95°C for 15 s, annealing at 64°C for 15 s, and extension at 72°C for 1 min were performed. PCR, denaturation, annealing, and extension were repeated for 40 cycles. The mRNA quantitation of CYP3A44, CYP3A41, or PXR was normalized to GAPDH mRNA and expressed as fold induction and compared with control mRNA expression as 1. Amplification and detection were performed using the ABI PRISM 7000 Sequence Detection System (Applied Biosystems, Foster City, CA) with ABI Prism 7000 SDS software.
Results
To investigate the role of glucocorticoids in regulation of the Cyp3a44 gene in vivo, adrenalectomy and then treatment with DEX were carried out. Figure 1 shows CYP3A44 mRNA expression in male and female mouse livers. In females, adrenalectomy drastically decreased the expression and DEX restored it; however, CYP3A44 mRNA expression was not detected in males, consistent with our previous report (Sakuma et al., 2002).
The effect of adrenalectomy and restoration by synthetic glucocorticoid treatment suggests a role for glucocorticoids in the control of Cyp3a44 gene expression in mouse liver. In the next experiment, the effect of glucocorticoids on CYP3A44 mRNA expression in cultured hepatocytes was investigated. Hepatocytes were treated with DEX, hydrocortisone, or corticosterone at 10-5 or 10-7 M, either in the absence or presence of 10-5 M RU486, which functions as an antagonist for GR (Cadepond et al., 1997) and as an agonist for PXR (Kliewer et al., 1998). As shown in Fig. 2A, the expression of CYP3A44 mRNA was induced by DEX at both 10-7 and 10-5 M and by hydrocortisone and corticosterone at 10-5 M; however, combined treatment with RU486 did not suppress glucocorticoid-induced CYP3A44 mRNA expression, which was a different result for RU486 than that with other mouse female-predominant CYP3A41 mRNA expression (Fig. 2B). On the other hand, treatment with RU486 alone significantly induced the mRNA expression of CYP3A44, and combined treatment of hydrocortisone at 10-5 M with RU486 significantly enhanced CYP3A44 mRNA expression over the level attained by treatment with hydrocortisone alone.
As DEX and natural glucocorticoids act as a common ligand for GR in various animal species, we next investigated the role of these glucocorticoids in the induction of CYP3A44 mRNA by a GR-mediated pathway. In our cell culture system, the expression of GR mRNA declined to 10% in the liver during the initial 24-h cultivation (data not shown); therefore, experiments were performed in primary cultured mouse hepatocytes transfected with GR expression plasmid in the presence of 10-7 to 10-5 M DEX or corticosterone. As shown in Fig. 3, in the absence of an expressed receptor, concentration-dependent effects of DEX on CYP3A44 mRNA were observed. As expected, when the GR expression plasmid was transfected, DEX treatment at 10-6 up to 10-5 M caused an increase in CYP3A44 expression over the level attained by empty plasmid; however, no significant increase in expression was seen at 10-7 M DEX in the presence of the GR expression plasmid. Similarly, induction by corticosterone was enhanced by transfection of the GR expression plasmid, but it was observed only at 10-5 M. These results demonstrate the role of DEX and corticosterone for the GR in the induction of Cyp3a44 gene expression at supramicromolar concentrations.
It is known that a submicromolar concentration of DEX results in the glucocorticoid receptor-mediated expression of the human PXR gene, which, in turn, is able to transactivate the human CYP3A4 gene, although the identity of the virtual activator of PXR at a submicromolar concentration of DEX is unclarified (Pascussi et al., 2001). We wondered whether the expression of mouse PXR also could be increased by DEX in primary cultured mouse hepatocytes. We anticipated that, if this were the case, one mechanism of DEX induction of the Cyp3a44 gene might act through GR-mediated expression of PXR. The results reported in Fig. 4 show that endogenous mouse PXR mRNA is induced significantly by DEX treatment at 10-5 M. Furthermore, transfection of the GR expression plasmid considerably induced PXR mRNA expression up to 2500-fold.
In consideration of these findings, together with activation of GR causing an increase in endogenous PXR mRNA at a low concentration of DEX, the next experiment was performed. To observe the potentiation of PXR transactivation in expression of the Cyp3a44 gene at a low concentration of DEX, rifampicin, which is a ligand for PXR, was used in the absence or presence of 10-7 M DEX, a concentration sufficient to activate the GR but not the PXR, and also at 10-5 M DEX, a concentration that activates both the GR and the PXR (Lehmann et al., 1998). As shown in Fig. 5, DEX at 10-7 M or rifampicin alone did not induce CYP3A44 mRNA. Concomitant addition of DEX at 10-7 M and rifampicin, however, enhanced Cyp3a44 gene expression. As suspected, DEX at 10-5 M in association with rifampicin increased Cyp3a44 expression over the level attained by DEX at 10-5 M alone. These observations serve as evidence for the synergistic effect of DEX at a low concentration and the PXR activator on CYP3A44 induction. Taken together with DEX increasing in PXR expression via a GR-mediated mechanism and this synergistic effect, the possibility that GR/PXR-dependent regulation for basal Cyp3a44 gene expression is suggested; however, no significant increase in Cyp3a44 expression was seen at 10-7 M DEX in the presence of the GR expression plasmid (Fig. 4). This observation might result from the lack of some PXR ligand(s) in primary cultured mouse hepatocytes, which is necessary for full induction.
DEX-induced expression of the Cyp3a44 gene was not antagonized by the addition of RU486, and Cyp3a44 was also induced by RU486 alone (Fig. 2A). These findings suggest that DEX-induced CYP3A44 mRNA might be a PXR-dependent process; thus, we examined the effects of transfection of expression plasmids for PXR and RXRα on the expression of the Cyp3a44 gene. Primary cultured mouse hepatocytes were transfected with these plasmids in the presence of 10-7 to 10-5 M DEX. As shown in Fig. 6, when PXR and RXRα were transfected, treatment with DEX at 10-6 to 10-5 M caused a significant increase in Cyp3a44 gene expression over the level attained by an empty plasmid. If increased response of the Cyp3a44 gene to glucocorticoids after transfection of the GR expression plasmid is due to an increased level of GR, not of PXR, it is anticipated that transfection of PXR and RXRα expression plasmids does not cause the increased response to DEX treatment. This result strongly suggests that CYP3A44 mRNA induction with a supramicromolar concentration of DEX is mediated through a direct PXR-dependent mechanism.
Discussion
We have investigated the mechanism by which DEX produces dual effects in CYP3A44 expression. The first is direct activation of the Cyp3a44 gene by the PXR-RXRα complex at a supramicromolar concentration, and the second is indirect activation of the Cyp3a44 gene through the induction of PXR gene expression by the GR pathway.
CYP3A44 has been identified as a female-predominant gene in the livers of C57BL/6 and ddY mice, and its expression was also dependent on the feminine plasma growth hormone profile (Sakuma et al., 2002). A recent study using transgenic human CYP3A mice also revealed the importance of the growth hormone profile in modulating the sex-dependent expression of the mouse Cyp3a44 gene (Cheung et al., 2006). Furthermore, the role of the nuclear receptors PXR and CAR in the regulation of Cyp3a44 gene expression has been defined as showing that gender also influences the critical impact of PXR- and CAR-mediated effects on CYP3A44 expression (Anakk et al., 2007). Anakk et al. reported that DEX induced CYP3A44 mRNA expression in male 129sv/C57BL6 mixed background mice, which we failed to observe in our study (Fig. 1). The difference in the CYP3A44 expression profile might be due to the mouse strains being examined in different laboratories.
In the present study, we demonstrated that the expression of the Cyp3a44 gene in mouse liver is under the control of glucocorticoids (Fig. 1). We propose that in the absence of a xenobiotic inducer glucocorticoids at the physiological level control the basal expression of CYP3A44, whereas in the presence of an inducer Cyp3a44 gene expression is induced.
The results with the GR expression plasmid (Figs. 3 and 4) suggest that regulation of the Cyp3a44 gene might have occurred in a GR-dependent manner; however, in contrast CYP3A44 was shown to have an expression profile that was different from those of CYP3A41 and TAT, a prototypical target gene of the GR (Grange et al., 1989), in which its expression peaked at lower concentrations of DEX (10-7 M) (Fig. 2) (Sakuma et al., 2004). CYP3A44 showed maximum induction by DEX at a higher concentration (10-5 N), and the profile was similar to that of CYP3A11 (Sakuma et al., 2004), a target of PXR (Kliewer et al., 1998). Furthermore, RU486, an antagonist for GR (Cadepond et al., 1997) and an agonist for PXR (Kliewer et al., 1998), did not show any suppressive effect on DEX-induced CYP3A44 mRNA expression, different from its suppressive effect on CYP3A41 expression, which is mediated by the GR (Fig. 2B). Based on these observations, DEX, hydrocortisone, or corticosterone induction of CYP3A44 may not be directly involved in GR activation.
However, the present data clarified a certain role of the GR in CYP3A44 mRNA expression. We suggest that the GR indirectly controls DEX induction of CYP3A44 mRNA by increasing the expression of the PXR. The following observations support this hypothesis. 1) Increased accumulation of endogenous mouse PXR mRNA was observed after transfection of the GR expression plasmid in the presence of DEX (Fig. 4). 2) Transfection of the GR expression plasmid in primary hepatocytes in the presence of DEX significantly activated CYP3A44 mRNA expression. Nevertheless, no GR transfection effect was observed at 10-7 M DEX, the concentration at which GR target genes such as the TAT gene are efficiently inducible (Fig. 3) (Sakuma et al., 2004). 3) Increased accumulation of endogenous PXR mRNA was observed only in the presence of DEX, whereas other glucocorticoids, such as hydrocortisone and corticosterone, did not affect the accumulation (data not shown). Therefore, it is suggested that DEX may act through the indirect GR-mediated activation of CYP3A44 in the present culture system.
The involvement of the PXR might be confirmed for Cyp3a44 gene expression by DEX (Fig. 6); however, the possibility that DEX only acts at a supramicromolar concentration through direct PXR-mediated activation of the Cyp3a44 gene is strongly suggested. The prototypical model of PXR-mediated gene induction at high concentrations of DEX has been well documented for the expression of the human CYP3A4 gene (Pascussi et al., 2001) or rat glutathione S-transferase A2 gene (Falkner et al., 2001). Studies by Pascussi et al. (2001) and Falkner et al. (2001) have shown that both genes are induced by PXR activation mediated by supramicromolar concentrations of DEX. Recent studies indicate the possibility that the expression of human CYP3A4 in the liver is relatively higher in women than in men (Dhir et al., 2006), suggesting that CYP3A44 has more similar regulation properties than CYP3A11, because the latter shows no significant difference between males and females. Taking those studies and the present findings together, Cyp3a44 could be considered as a relevant murine CYP3A model gene of human CYP3A4 with respect to PXR-mediated induction and sexually dimorphic expression.
In consideration of the fact that the dose of DEX used in this in vivo study was higher than those of therapeutic treatments, the resulting concentration of glucocorticoid in plasma is higher than that seen in physiological status. Furthermore, we recognize that because DEX is a synthetic derivative of glucocorticoids, the inductive effect observed in adrenalectomized mice must be reflected by both pharmacological, i.e., the reaction caused by a high concentration of drug, and the physiological effect, i.e., the response caused by glucocorticoid hormones at physiological concentrations. Because the expression of the Cyp3a44 gene was decreased after adrenalectomy, it is expected that glucocorticoids have a physiological role in the regulation of the Cyp3a44 gene. However, an in vitro study showed no effect of natural glucocorticoids at submicromolar concentrations (physiological condition) (Fig. 3). The reason for this contradiction is not clear at present, but it is likely that it may result from the lack of some factor(s) in primary cultured mouse hepatocytes, which is necessary for high-level expression of the Cyp3a44 gene in vivo. Although the in vivo studies were undertaken in C57BL/6 mice and the cultured hepatocyte (in vitro) studies were undertaken in ddY mice, our unpublished results revealed no differences in Cyp3a44 gene expression between these two strains.
From the above findings, a GR/PXR-dependent regulation mechanism model of CYP3A44 expression is presented in Fig. 7. First, DEX at a high concentration activates PXR, resulting in Cyp3a44 gene expression (direct PXR-dependent mechanism). Second, DEX increases PXR expression via a GR-mediated mechanism (indirect GR-dependent mechanism). At a low DEX concentration, the same mechanism might be possible (Figs. 4 and 5); however, induction of the Cyp3a44 gene by the direct PXR-dependent mechanism is limited to a very low level, because of the relatively low potential of DEX activating the PXR (EC50 = 10 μM) (Lehmann et al., 1998).
Footnotes
-
This work was partly supported by Grants-in-Aid from the Japanese Ministry of Education, Culture, Sport, and Science and the Smoking Research Foundation.
-
doi:10.1124/dmd.107.016832.
-
ABBREVIATIONS: CAR, constitutive androstane receptor; PXR, pregnane X receptor; GR, glucocorticoid receptor; GRE, glucocorticoid responsive element; bp, base pairs; RXRα, retinoid X receptor-α; DEX, dexamethasone; RU486, 11β-[4-dimethylamino] phenyl-17β-hydroxy-17-[1-propynyl] estra-4,9-diene-3-one; RT, reverse-transcriptase; PCR, polymerase chain reaction; GAPDH, glyceraldehyde-3-phosphate.
- Received May 24, 2007.
- Accepted July 18, 2007.
- The American Society for Pharmacology and Experimental Therapeutics