Abstract
Hepatocyte growth factor (HGF), an antimitogenic factor for HepG2 cells, increased mRNA and protein levels of UGT1A1 and CYP2B6, as well as the endogenous cyclin-dependent kinase (CDK) inhibitors p16, p21, and p27 in HepG2 cells but not in HuH6, Caco2, or MCF7 cells. Treatment with 1,4-diamino-2,3-dicyano-1,4-bis(methylthio)butadiene (U0126) (an extracellular signal-regulated kinase inhibitor) suppressed the HGF-induced expression of UGT1A1 and CYP2B6, as well as p16, p21, and p27 in HepG2 cells. The CDK inhibitor roscovitine also enhanced the expression of UGT1A1, CYP2B6, and CYP3A4. Transfection of anti-CDK2 siRNA led to elevated levels of UGT1A1, CYP2B6, and CYP3A4 in HepG2 and SW480 cells, whereas anti-CDK4 small interfering RNA (siRNA) did not significantly enhance the expression of these enzymes. In fact, CDK2 activity was decreased in HGF-treated HepG2 cells. In cells arrested in S phase by a thymidine block and then released into a synchronous cell cycle, there was a clear dissociation among the activation of CDK2 and the expression of UGT1A1, CYP2B6, and CYP3A4. Furthermore, the induction of CYP3A4 but not UGT1A1 or CYP2B6 mRNA expression by roscovitine was repressed in pregnane X receptor (PXR) siRNA-transfected HepG2 cells. Transfection with constitutive androstane receptor siRNA or PXR siRNA in HepG2 cells did not repress the HGF-stimulated expression of UGT1A1 mRNA. Taken together, our results show that the expression of UGT1A1 and CYP2B6 is negatively regulated through a CDK2 signaling pathway linked to cell cycle progression in HepG2 and SW480 cells, the mechanism of which may differ from that of CYP3A4 expression through PXR phosphorylated by CDK2.
The constitutive androstane receptor [(CAR) NR1I3] was originally characterized as a nuclear hormone receptor that interacts with a subset of retinoic acid response elements (Baes et al., 1994). CAR and pregnane X receptor [(PXR) NR1I2] have been recognized as xenobiotic-sensing nuclear receptors that transcriptionally regulate the expression of genes of phase I, II, and III metabolic enzymes and transporters involved in the metabolism and elimination of endogenous and exogenous substances, such as bilirubin, steroid hormones, and xenobiotics (Timsit and Negishi, 2007). UGT1A1 plays a critical role in the detoxification of potentially neurotoxic bilirubin by conjugating it with glucuronic acid for excretion in bile (Ostrow and Murphy, 1970) and conjugates drugs and other xenobiotics (Radominska-Pandya et al., 1999; Tukey and Strassburg, 2000). We identified a phenobarbital (PB)-responsive enhancer module at −3499/−3210 from the transcription start site of UGT1A1, gtPBREM (Sugatani et al., 2001), and showed that the gtNR1 (−3382/−3367) within gtPBREM plays a central role in the expression of UGT1A1 mediated by both CAR and PXR (Sugatani et al., 2005a,b). In addition, we showed that hepatic nuclear factor 1α bound to the proximal promoter motif not only enhances the basal reporter activity of UGT1A1, including the distal (−3570/−3180) and proximal (−165/−1) regions, but also influences the transcriptional regulation of UGT1A1 by CAR, PXR, glucocorticoid receptor (GR), and arylhydrocarbon receptor (AhR) to markedly enhance reporter activities (Sugatani et al., 2008).
We previously showed that CAR expression changes during the cell cycle in HepG2 and SW480 cells and CAR protein accumulates during G1 in both cells (Osabe et al., 2008). In this study, we found that UGT1A1 and CYP2B6 began to accumulate in late mitosis, preceding the CAR's accumulation in the nucleus, and remained at high levels during G1, coinciding with CAR's accumulation. Klinger and Karge (1987) showed that a transient decrease in cytochrome P450 (P450) occurred during liver regeneration following partial hepatectomy. The hepatic growth factors, epidermal growth factor (EGF) and transforming growth factor α, have also been shown to inhibit induction of the mRNA and protein expression of the CYP2B subfamily in primary culture of mouse hepatocytes (Aubrecht et al., 1995). However, it remains to be fully elucidated how the expression of drug-metabolizing enzymes including UGT1A1 and CYP2B6 is regulated by cell signals associated with cell cycle progression.
CAR is predominantly expressed in the liver and is found in the cytoplasm of normal mouse hepatocytes in the absence of stimuli such as drug treatment (Kawamoto et al., 1999). CAR is activated by PB and PB-like inducers, such as 1,4-bis[2-(3,5-dichloropyridyloxy)]benzene (TCPOBOP), that do not bind it directly but activate a signal transduction pathway, resulting in the translocation of CAR from the cytoplasm to the nucleus (Kawamoto et al., 1999). Koike et al. (2007) showed that extracellular signal-regulated kinase (ERK) is an endogenous signal retaining CAR in the cytoplasm in mouse primary hepatocytes, showing that 1) hepatocyte growth factor (HGF) effectively repressed the induction of endogenous Cyp2b10 gene by PB and TCPOBOP in mouse primary hepatocytes; 2) HGF increased phosphorylation of ERK1/2 in the cytosol, decreasing the TCPOBOP-induced nuclear accumulation of CAR; and 3) U0126 dephosphorylated ERK1/2 and increased nuclear CAR accumulation leading to enhanced CAR-regulated promoter activities. Whereas HGF is a potent mitogen for hepatocytes, it is an antimitogenic factor for some tumor cell lines, including HepG2; HGF has been shown to induce p16 and p21 expression and decrease cyclin-dependent kinase (CDK) 2 activity, leading to inhibition of cell growth (Shima et al., 1998; Han et al., 2005). p16 and p21, which are endogenous CDK2/CDK4 inhibitors, are involved in regulating cell cycle progression. Here we found that HGF treatment increased the expression of UGT1A1 and CYP2B6 in HepG2 cells, even when CAR levels decreased both in the cell and in the nucleus. A recent study by Lin et al. (2008) showed that inhibition of CDKs by roscovitine leads to activation of PXR-mediated CYP3A4 gene expression and that CDK2 negatively regulates the activity of PXR in HepG2 cells. Thus, in the present study, we investigated the cell signaling involved in the HGF-stimulated expression of UGT1A1 and CYP2B6 in HepG2 cells to examine whether CDKs participate in the cell cycle-dependent expression of UGT1A1 and CYP2B6.
Materials and Methods
Materials.
U0126, U0124, LY294002, SB203580, SP600125, roscovitine, SU9516, and the CDK inhibitor p35 were purchased from Calbiochem (Darmstadt, Germany). EGF and platelet-derived growth factor (PDGF) were obtained from Sigma-Aldrich (St. Louis, MO). Histone H1 and ATP were from Roche Diagnostics (Mannheim, Germany). All the other chemicals and solvents were of analytical grade and obtained from commercial sources.
Cell Culture Conditions and Treatments.
HepG2 and HuH6 human hepatoblastoma cells and Caco2 human colon carcinoma cells from RIKEN BioResource Center (Ibaraki, Japan) and SW480 human colon cancer cells and MCF7 human breast adenocarcinoma cells from the American Type Culture Collection (1 × 105 cells/ml; Manassas, VA) were cultured in Dulbecco's modified Eagle's medium (DMEM; HepG2, HuH6, Caco2, and MCF7 cells) and RPMI 1640 medium (SW480 cells) with 10% fetal calf serum (FCS) and antibiotics (100 μg of streptomycin and 10 U of penicillin/ml) or DMEM-Ham's F-12 serum-free medium supplemented with 2 mM glutamine, 15 mM HEPES, 5 μg/ml insulin, 10 μg/ml bovine transferrin, 10 ng/ml Na2SeO3, 2 μg/ml aminolevulinic acid, 25 mM glucose, 0.5 mg/ml linoleic acid-albumin, 1 mM pyruvate with or without 100 ng/ml human growth hormone, and the antibiotics at 37°C in the presence of 5% CO2 unless otherwise stated.
Cell Cycle Analysis.
Cells were trypsinized and harvested at various times. They were stained with 50 μg/ml propidium iodide in flow reagent (0.1% sodium citrate, 0.2% Nonidet P40, and 0.25 mg/ml DNase-free RNase). Cell cycle distribution was monitored with a BD FACS Canto II flow cytometer (BD Biosciences, San Jose, CA). At least 10,000 cells were analyzed for each sample.
Small Interfering RNA-Mediated Protein Knockdown.
Small interfering RNA (siRNA) targeting human CDK2 [validated stealth RNAi VHS40359 (CDK2 siRNA)] and human PXR, 5′-aaaugggagaagguagugucaaagg-3′ and 5′-ccuuugacacuaccuucucccauuu-3′, were obtained from Invitrogen (Carlsbad, CA). siRNA targeting human CAR, 5′-gcaacugaguaaggagcaaTdT-3′ and 5′-uugcuccuuacucaguugcTdT-3′, and human CDK4, 5′-gguaauccggagugagcaadTdT-3′ and 5′-uugcucacuccggauuaccdTdT-3′, were from B-Bridge International (Sunnyvale, CA). Cells cultured for 24 h were transfected with siRNA duplexes using TransIT-siQUEST (Mirus Bio, Madison, WI) according to the manufacturer's instructions. At 24 h after transfection, cells were given fresh medium and further transfected with the siRNA duplexes for an additional 24 h unless stated otherwise.
Quantitative Reverse Transcription-Polymerase Chain Reaction.
Total RNA was extracted using TRIzol reagent from Invitrogen, and cDNAs were synthesized with a PrimeScript RT reagent kit (Takara Bio, Otsu, Japan) according to the directions. cDNA synthesized from 100 ng of total RNA was subjected to quantitative real-time polymerase chain reaction (PCR) with Premix Ex Taq or SYBR Premix ExTaq (Takara Bio) for UGT1A1 (NM_000463), CYP1A1 (NM_000499), CYP2B6 (NM_000767), CYP3A4 (NM_017460), CAR (NM_005122), PXR (NM_022002), retinoid X receptor (RXR; NM_002957), AhR (NM_001621), p16 (NM_000077), p21 (NM_000389), p27 (NM_004064), CDK2 (NM_001798), and CDK4 (NM_000075), as described previously (Osabe et al., 2009).
Western Blot Analysis.
Treated and untreated cells were washed three times with ice-cold phosphate-buffered saline and lysed with a freshly prepared lysis buffer containing 50 mM Tris-HCl, pH 7.4, 1% Nonidet P40, 0.25% sodium deoxycholate, 150 mM NaCl, 1 mM EDTA, 1 mM phenylmethylsulfonyl fluoride, 1 mM Na3VO4, 1 mM NaF, 1 mM benzamidine, 1 μg/ml aprotinin, and 1 μg/ml leupeptin at 4°C. The lysates were centrifuged at 110,000g for 10 min. Nuclear extracts were prepared using a nuclear extract kit (Active Motif, Carlsbad, CA). Microsomal proteins were prepared as described previously (Sugatani et al., 2004). The protein concentrations were determined using the Bradford protein assay kit (Bio-Rad Laboratories, Hercules, CA). Western blotting was performed as described (Osabe et al., 2008). In brief, nuclear extracts, microsomal proteins, or cell lysates (50 μg) were resolved on 12.5% SDS-polyacrylamide gels, and the proteins were transferred electrophoretically to polyvinylidene difluoride membranes (Millipore Corporation, Billerica, MA). Membranes were blocked at 4°C overnight in Blocking One or Blocking One-P (Nacalai Tesque, Kyoto, Japan) and probed for 1 h with primary antibodies including anti-UGT1A1, anti-CYP2B6, and anti-CYP3A4 from BD Gentest (Woburn, MA); anti-CYP1A1 and anti-NADPH-P450 reductase (CPR) from Daiichi Pure Chemicals Co. (Tokyo, Japan); anti-CAR (sc-13065), anti-PXR (sc-7737), anti-RXR (sc-553), anti-histone H1 (sc-8030), anti-cyclin B1 (sc-7393), anti-cyclin D1 (sc-718), and anti-CDK4 (sc-601) from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA); anti-phospho-ERK, anti-ERK, anti-p16, anti-CDK1(CDC2), anti-phospho-CDK1, anti-CDK2, anti-phospho-CDK2, and anti-β-actin from Cell Signaling Technology Inc. (Danvers, MA); anti-p21, anti-p27, and anti-cyclin A from Millipore Corporation; anti-cyclin E; and anti-α-tubulin (Calbiochem). Antigen-antibody complexes were detected using the appropriate secondary antibody conjugated to horseradish peroxidase (horseradish peroxidase-conjugated anti-rabbit, anti-goat, or anti-mouse Ig; Jackson ImmunoResearch Laboratories Inc., West Grove, PA) and visualized with an enhanced chemiluminescence system (GE Healthcare, Little Chalfont, Buckinghamshire, UK).
In Vitro Kinase Assay.
CDK2 kinase assays were performed as described previously (Walter et al., 2002). Cell lysates were cleared by centrifugation, and equal amounts of protein in the cell extracts (200 μg) were immunoprecipitated with 2 μg of anti-CDK2 antibody for 3 h at 4°C. Precipitated immune complexes were washed three times with the cell lysis buffer and twice with a kinase buffer (50 mM Tris-HCl, pH 7.5, containing 10 mM MgCl2, 1 mM dithiothreitol, 20 mM EGTA, 10 mM β-glycerophosphate, 0.1 mM sodium orthovanadate, and 1 mM NaF). Kinase reactions were performed in 10 μl of kinase buffer containing 25 μg of histone H1, 100 μM ATP, and 0.37 MBq of [γ-32P]ATP at 30°C for 30 min. Reactions were stopped by addition of 10 μl of 2× Laemmli sample buffer (125 mM Tris-HCl, pH 6.8, 4% SDS, 20% glycerol, 0.002% bromophenol blue, and 10% 2-mecaptoethanol), and samples were resolved by SDS-polyacrylamide gel electrophoresis on a 12.5% gel after heat denaturation. Phosphorylation of the substrate was visualized by autoradiography.
Statistics.
Values are expressed as the mean ± S.E. All data were analyzed using a one-way analysis of variance. The statistical significance of differences between groups was analyzed using the analysis of variance or an unpaired t test. The level for statistically significant differences was set at p < 0.05.
Results
Expression of UGT1A1, CYP2B6, and Cyclin Kinase Inhibitors p16, p21, and p27 during HGF-Induced HepG2 Growth Inhibition.
Whereas HGF is a potent mitogen for many types of cells including human hepatocytes and the tumor cell lines HuH6, Caco2, and MCF7, it is also an antimitogenic factor for several types of tumor cells such as HepG2 cells. Treatment of HepG2 cells with HGF (100 ng/ml)) decreased the cell number compared with control cells (Fig. 1A). Whereas HGF down-regulated the expression of P450 isozymes in human hepatocytes (Donato et al., 1998), its function in tumor cells remains to be resolved. To assess the effects of growth factor on the expression of drug-metabolizing enzymes in tumor cells, the response to HGF was compared between two opposite responding types of human hepatoblastoma cells, HepG2 and HuH6 (Shima et al., 1998) (Fig. 1B). The mRNA expression of CDK inhibitors (p16, p21, and p27) was induced 24 h after treatment with HGF in HepG2 cells, but the levels did not change in HuH6 cells. Whereas the mRNA level of CAR decreased in HepG2 cells, the mRNA levels of UGT1A1 and CYP2B6 increased (Fig. 1B). In contrast, in Caco2 and MCF7 cells after 24 h of culture with HGF, neither the mRNA levels of UGT1A1 and CYP2B6 nor the mRNA levels of p16, p21, and p27 were increased (data not shown).
To investigate the cell signaling specifically required for stimulation by growth factors, effects of growth factors on the expression of drug-metabolizing enzymes and nuclear receptors in serum-free medium were determined (Fig. 2). The mRNA level of UGT1A1 was increased after treatment with HGF and EGF but not PDGF, whereas the mRNA levels of CAR and PXR but not RXR and AhR were decreased (Fig. 2). The mRNA level of CYP2B6 was increased by HGF. The protein levels of UGT1A1 and CYP2B6 but not CPR were increased, whereas that of CAR in the nucleus was decreased and those of PXR and RXR remained largely unchanged after HGF treatment (Fig. 2, II and III). In serum-free medium, the production of p16, p21, and p27 proteins was induced 1 h after treatment with HGF, and the expression of p21 and p27 was sustained for at least 24 h, consistent with the phosphorylation of ERK1/2 (Fig. 3A). Next, to assess the signaling pathway of UGT1A1 and CYP2B6 expression after HGF treatment, we examined the effects of U0126 (ERK inhibitor), LY294002 [phosphatidylinositol 3 (PI3) kinase inhibitor], SB203580 (p38 mitogen-activated protein kinase inhibitor), and SP600125 [c-Jun NH2-terminal kinase (JNK) inhibitor] on the expression of UGT1A1 and CYP2B6 mRNAs in cells cultured in serum-free medium. U0126 but not U0124 (negative control) decreased the mRNA levels of UGT1A1 and CYP2B6 in HGF-treated cells (Fig. 3B), whereas the mRNA levels of CAR but not PXR, RXR, or AhR increased in the cells; the mRNA levels of CAR increased to 2.0 ± 0.3- and 1.5 ± 0.3-fold of the control by U0126 in the absence and presence of HGF, respectively (data not shown). Likewise, U0126 decreased the mRNA levels of UGT1A1 in EGF-treated cells cultured in serum-free medium (data not shown). The HGF-induced expression of endogenous cyclin kinase inhibitors (p16, p21, and p27) was also suppressed by U0126 (Fig. 3A). These results suggest the ERK pathway to be involved in the effect of growth factors such as HGF and EGF on UGT1A1 expression, as well as the induction of p16, p21, and p27 expression. On the other hand, LY294002 and SP600125 increased the mRNA levels of UGT1A1 and CYP2B6 in HGF-treated cells compared with those in the control cells, suggesting PI3 kinase and JNK to be involved in the HGF-induced expression of UGT1A1 and CYP2B6 in HepG2 cells.
Cyclin Kinase Inhibitors Stimulate the Expression of UGT1A1, CYP1A1, CYP2B6, and CYP3A4 in HepG2 Cells.
To explore whether cyclin kinases are involved in regulating the expression of drug-metabolizing enzymes in human cells, we examined the effects of the CDK inhibitors roscovitine (CDK1, CDK2, and CDK5 inhibitor) (Meijer et al., 1997), SU9516 (CDK1, CDK2, and CDK4 inhibitor) (Yu et al., 2002), and CDK inhibitor p35 (CDK1 and CDK2 inhibitor) (Vermeulen et al., 2002) on the expression of UGT1A1, CYP1A1, CYP2B6, and CYP3A4 mRNAs in HepG2 cells. Expression of UGT1A1 mRNA was enhanced by roscovitine, SU9516, and CDK inhibitor p35 in a dose-dependent manner (Fig. 4A). Roscovitine markedly stimulated the expression of UGT1A1 mRNA and protein, whereas it slightly stimulated the expression of CYP1A1, CYP2B6, and CYP3A4 mRNAs and proteins (Fig. 4). Overall, the findings indicated that CDKs were clearly involved in regulating the expression of UGT1A1 and CYP2B6 in HepG2 cells.
Effects of Anti-CDK4 siRNA and Anti-CDK2 siRNA on Induction of UGT1A1 and CYP2B6 Expression.
Next, to determine whether CDK4 and CDK2, the activities of which were inhibited by p16, p21, and p27, were required for the down-regulation of UGT1A1 and CYP2B6 expression, anti-CDK4 siRNA or anti-CDK2 siRNA was introduced into HepG2 and SW480 cells. As shown in Fig. 5, transfection with the anti-CDK4 siRNA dramatically reduced the levels of CDK4 mRNA and protein in HepG2 cells but was not followed by an increase in mRNA and/or protein levels of UGT1A1, CYP1A1, CYP2B6, and CYP3A4. In contrast, transfection with the anti-CDK2 siRNA in HepG2 and SW480 cells was followed by an increase in mRNA levels of UGT1A1, CYP2B6, and CYP3A4 (Fig. 6). Furthermore, transfection with the anti-CDK2 siRNA led to stimulated expression of UGT1A1, CYP2B6, and CYP3A4 proteins but not CYP1A1 protein in HepG2 and SW480 cells, as well as a decrease in CDK2 protein levels, with the extent of the stimulation by anti-CDK2 siRNA greater in SW480 cells than in HepG2 cells (Fig. 7).
To determine whether the CDK2 activity was reduced in HepG2 cells after treatment with HGF, we analyzed the extent of phosphorylation of histone H1 after incubation with the cell lysate. As shown in Fig. 8, the extent of phosphorylation was decreased in HGF-treated cell lysate compared with control lysate, indicating that the CDK2 activity was decreased in the HGF-treated HepG2 cells, and that CDK2 activity was required for down-regulation of UGT1A1, CYP2B6, and CYP3A4 expression.
CDK2 Is Involved in the Cell Cycle-Dependent Down-Regulation of UGT1A1, CYP2B6, and CYP3A4.
To investigate whether the expression of UGT1A1, CYP2B6, and CYP3A4 changes during the cell cycle, we undertook a detailed analysis of protein levels in synchronously dividing SW480 cells. SW480 cells arrested in S phase by a double-thymidine block were released into a synchronous cell cycle and sampled every 2 h. Cells completed S phase in 6 h of their release from the thymidine block and mitosis in 10 h (Fig. 9). Protein levels of cyclin kinases, cyclins, and drug-metabolizing enzymes in the cell lysate were determined by Western blotting (Fig. 9). As described in the previous study (Osabe et al., 2008), cyclin A and cyclin B1 protein levels decreased at 12 to 16 h after release from the thymidine block as cells exited G1. Cyclin D1 was present at very low levels at 0 h after the release and started accumulating. The protein levels of cyclin E1 remained unchanged. CDK4 was present at 0 to 6 h after release from the thymidine block and started accumulating at 10 h. We found that whereas the total protein levels of CDK1 and CDK2 remained largely unchanged in the cells after release from the thymidine block for 16 h, the active form of CDK2 (phospho-CDK2) was increased and peaked at 2 h, followed by the active form of CDK1 (phospho-CDK1), which peaked at 4 h. The time-dependent changes in the phospho-CDK2 level matched the time-dependent accumulation of S phase cells. Whereas the active form of CDK2 (phospho-CDK2) was present at high levels at 2 to 6 h after release from the thymidine block and these levels decreased at 8 h, UGT1A1 and CYP2B6 were present at very low levels at 0 to 6 h and started accumulating at 8 h. CYP3A4 was present at a very low level at 0 to 4 h after release from the thymidine block and started accumulating at 6 h, whereas the protein levels of CPR remained largely unchanged. Moreover, similar cell cycle-dependent expression of UGT1A1, CYP2B6, and CYP3A4 in HepG2 cells was found as in SW480 cells (data not shown).
Regulatory Mechanism of HGF- and Roscovitine-Stimulated UGT1A1 and CYP2B6 Gene Expression Was Different from That of the CYP3A4 Gene Expression.
Lin et al. (2008) recently reported that CDK2 directly phosphorylates PXR and negatively regulates human PXR-mediated CYP3A4 gene expression in HepG2 cells. To determine whether the HGF- and roscovitine-stimulated UGT1A1 and CYP2B6 gene expression is regulated through a pathway mediated by CAR or PXR, as well as CYP3A4 gene expression, anti-CAR siRNA or anti-PXR siRNA was introduced into HepG2 cells. As shown in Fig. 10, A and B, transfection with anti-CAR siRNA and anti-PXR siRNA reduced the levels of CAR and PXR mRNAs in HepG2 cells, respectively, but did not change the mRNA expression of UGT1A1 stimulated by HGF. Transfection with anti-CAR siRNA did not change the mRNA levels of UGT1A1 and CYP3A4 stimulated by roscovitine (Supplemental Fig. 2). Furthermore, whereas transfection with anti-PXR siRNA reduced the mRNA levels of CYP3A4 stimulated by roscovitine, mRNA levels of UGT1A1 and CYP2B6 after treatment with roscovitine remained similar to those in control cells (Fig. 10C). These findings suggest that the mechanism to down-regulate the expression of UGT1A1 and CYP2B6 gene by HGF and roscovitine may differ from that of CYP3A4 gene.
Discussion
Reduced bilirubin glucuronosyltransferase (UGT1A1) activity is associated with the development of unconjugated hyperbilirubinemia (Crigler-Najjar syndrome and Gilbert's syndrome) (Mackenzie et al., 1997) and increased side effects of drug treatment such as the predisposition of patients to toxicity initiated by SN-38, an active metabolite of the anticancer drug irinotecan (Gagné et al., 2002; Tukey et al., 2002). Understanding the molecular mechanisms of the induction of human UGT1A1 may provide information for the prevention and treatment of unconjugated hyperbilirubinemia and the side effects of drugs. Whereas the proximal 165-base pair (bp) promoter motif is regulated by hepatic nuclear factor 1α (Bernard et al., 1999), the distal 290-bp enhancer module (gtPBREM) of human UGT1A1 is regulated by CAR as a transcription factor in response to PB treatment (Sugatani et al., 2001), by PXR as a nuclear receptor responsible for rifampicin-induced activation of gtPBREM (Sugatani et al., 2004), by GR as a nuclear hormone receptor capable of activating gtPBREM by dexamethasone (Sugatani et al., 2005a), by the receptor-type transcription factor AhR responsible for p-naphthoflavone-induced activation (Yueh et al., 2003), and by peroxisome proliferator-activated receptor α responsible for WY-14643 (Senekeo-Effenberger et al., 2007). The 290-bp gtPBREM is characterized as a composite regulatory element containing the multiple binding sites DR4, gtNR1, DR3, glucocorticoid-response elements 1 and 2 for the nuclear receptors CAR, PXR, and GR (Sugatani et al., 2001, 2004, 2005a), as well as the AhR response element (Yueh et al., 2003) and the peroxisome proliferator-activated receptor α response element (Senekeo-Effenberger et al., 2007). It is noteworthy that the nuclear receptor CAR is essential for regulating UGT1A1 and the CYP2B subfamily by PB and PB-like inducers (Sugatani et al., 2005b; Timsit and Negishi, 2007).
HGF is known not only as a potent mitogen for hepatocytes and several types of tumor cells, including HuH6, Caco2, and MCF7 cells, but also as an antimitogenic factor for some types of tumor cells such as HepG2 cells. In regenerating rat liver induced by extended hepatectomy, the expression of liver-specific genes such as the serum albumin and CYP2B genes has been shown to be suppressed (Kakizaki et al., 2007). Moreover, Thasler et al. (2006) showed that an augmenter of liver regeneration, HGF, reduced CAR but not PXR or AhR expression and down-regulated basal and induced P450 expression in human hepatocytes in vitro, indicating that growth signals influence hepatic drug metabolism. Down-regulation of P450 expression by related factors such as EGF and transforming growth factor α has been also observed in humans (Greuet et al., 1997). In contrast, this study showed that HGF stimulated the expression of UGT1A1 and CYP2B6 in HepG2 cells whose growth was suppressed by it, whereas HGF did not stimulate the expression of these enzymes in HuH6 cells (Figs. 1–3). Whereas EGF also stimulated the expression of UGT1A1 in HepG2 cells, it has been shown that the receptor of HGF, c-Met, associates with EGF receptor in human hepatoma cell lines, including HepG2, and this association facilitates the phosphorylation of c-Met in the absence of HGF (Jo et al., 2000). Shima et al. (1998) found that 1) protein levels of p21 markedly decreased in HGF-treated HuH6 cells, but high levels of p21 were sustained for 24 h after HGF treatment in HepG2 cells; and 2) significant levels of p16 and p27 were observed after 24 h only in HepG2 cells. The induction of p16, p21, and p27 expression by HGF in HepG2 cells is considered to lead to growth inhibition by inhibiting CDKs (CDK2 and CDK4) as reported previously (Shima et al., 1998; Han et al., 2005). Here we found that 1) inhibition by U0126 of HGF downstream of mitogen-associated protein kinase kinase led to the repression of expression of both CDK inhibitors (p16, p21, and p27) and the drug-metabolizing enzymes UGT1A1 and CYP2B6 despite up-regulating CAR expression in HepG2 cells (Figs. 1–3); 2) HGF did not induce the expression of UGT1A1 and CYP2B6 in HuH6 cells, which did not follow the induction of endogenous CDK inhibitors (Fig. 1); and 3) the CDK inhibitor roscovitine stimulated the expression of UGT1A1 and CYP2B6 in HepG2 cells (Fig. 4). The findings suggested the endogenous CDK inhibitors to be involved in the HGF-induced expression of UGT1A1 and CYP2B6 in HepG2 cells (Fig. 11).
In the regulatory network of progression through the mammalian cell cycle, cyclin D-CDK4 is activated at G1, cyclin E-CDK2 is necessary for G1/S transition, and the activation of CDK4 and CDK2 leads to the nuclear accumulation of cyclin-CDK complexes during G1 (Jaumot et al., 1999). As shown in Fig. 11, cyclin D1/CDK4/CDK6 and cyclin E/CDK2 complexes cooperate in retinoblastoma protein phosphorylation, leading to its inactivation, dissociation of the transcription factor E2F, and cell cycle progression (Lundberg and Weinberg, 1999). Thus, to investigate whether CDK plays a key role in the induction of UGT1A1 and CYP2B6 expression, we examined the effect of CDK inhibitors in HepG2 cells. We showed that the expression of UGT1A1 mRNA was induced by roscovitine, SU9516, and CDK inhibitor p35, which all inhibit CDK1 and CDK2 activities (Fig. 4). Because p16, p21, and p27 repress CDK2 and CDK4 activities, the induction of UGT1A1 expression by roscovitine may be caused by the inhibition of CDK2 activity. To confirm whether CDK2 contributes to the induction of UGT1A1 and CYP2B6 expression through cell cycle progression, we deleted CDK2 using siRNA in HepG2 cells expressing the HGF receptor and SW480 cells expressing no HGF receptor (Zeng et al., 2004). CDK2 deletion resulted in the expression of UGT1A1 and CYP2B6 in both cells, whereas CYP1A1 protein levels were not largely changed in CDK2-deleted cells (Figs. 6 and 7). In fact, the level of CDK2 activity in HGF-treated cells was decreased (Fig. 8). CDK2, a key regulator of G1-S cell cycle progression, is activated by Akt-mediated phosphorylation (Maddika et al., 2008). The inhibition by LY294002 of HGF downstream of the PI3 kinase-Akt signaling pathway led to an increase in the expression of UGT1A1 and CYP2B6 (Fig. 3), also suggesting the involvement of CDK2 in the HGF-induced expression of UGT1A1 and CYP2B6 in HepG2 cells (Fig. 11). In contrast, CDK4 deletion using siRNA in HepG2 cells did not lead to an increase in the expression of UGT1A1 and CYP2B6 (Fig. 5). Furthermore, there was a clear dissociation among the expression of activated CDK2 and that of UGT1A1, CYP2B6, and CYP3A4 in cells arrested in S phase by a double-thymidine method and then released into a synchronous cell cycle (Fig. 9). The findings suggested activated CDK2 to be essential for down-regulating the expression of UGT1A1 and CYP2B6. In addition, the inhibition by SP600125 of HGF downstream of the JNK signaling pathway also increased the expression of UGT1A1 and CYP2B6 (Fig. 3). Lu et al. (2009) showed that docosahexaenoic acid down-regulates PB-induced CYP2B1 gene expression in rat primary hepatocytes via the JNK pathway; thus, JNK may be involved in regulating the expression of UGT1A1 and CYP2B6 by HGF (Fig. 11).
Does CAR or PXR play an essential role in UGT1A1 and CYP2B6 expression in HGF-stimulated HepG2 cells? The CAR protein level not only in the cell lysate (Osabe et al., 2008) but also in the nucleus (Fig. 2) was reduced after HGF treatment. In addition, the suppression of CAR expression using anti-CAR siRNA did not repress the HGF- and roscovitine-induced expression of UGT1A1 (Fig. 10). These observations indicate that changes in the expression of CAR did not affect UGT1A1 levels in HGF- and roscovitine-treated HepG2 cells. A recent study by Lin et al. (2008) showed that CDK2 directly phosphorylates PXR and negatively regulates the activity of PXR, leading to the repression of PXR-mediated CYP3A4 expression by roscovitine. However, in the present study, although PXR deletion using siRNA repressed the roscovitine-stimulated CYP3A4 expression in HepG2 cells, it did not affect the HGF- or roscovitine-stimulated expression of UGT1A1 and CYP2B6 (Fig. 10). These observations suggest that CDK2 negatively regulates the expression of UGT1A1 and CYP2B6, as well as CYP3A4, but the factor(s) phosphorylated by CDK2, which is involved in the UGT1A1 and CYP2B6 gene expression, may be different from PXR.
Does the cell cycle-dependent expression of drug-metabolizing enzymes in human tissues have any implications? When treating patients with drugs modulating cell cycle progression, such as the anticancer agent E7070, which arrests cells at the G1/S boundary (van den Bongard et al., 2004), the pharmacokinetics of other therapeutic drugs may be influenced, resulting in side effects. One has to pay attention to drug-drug interactions under such conditions. Because in HepG2 and SW480 cells CDK2 deletion using siRNA led to an increase in the gene expression of UGT1A1 and CYP2B6 (Fig. 6), the CDK2 signaling pathway may play an essential role in the induction of UGT1A1 and CYP2B6 expression in human cells, including hepatocytes. It remains to be clarified how the expression of UGT1A1 and CYP2B6 is modulated by CDK2 and JNK through the cell cycle. We are now studying the molecular mechanism by which UGT1A1 and CYP2B6 expression is negatively regulated by CDK2 and JNK in human tumor cells and normal cells.
Footnotes
This work was supported in part by the 21st Century Center of Excellence (COE21) Program, Global COE, a grant-in-aid for Scientific Research [Grants 19590070, 19590151, and 21590170]; and Cooperation of Innovative Technology from the Ministry of Education, Culture, Sports, Science and Technology, Japan.
Article, publication date, and citation information can be found at http://dmd.aspetjournals.org.
doi:10.1124/dmd.109.029785
↵ The online version of this article (available at http://dmd.aspetjournals.org) contains supplemental material.
-
- CAR
- constitutive androstane receptor
- PXR
- pregnane X receptor
- PB
- phenobarbital
- GR
- glucocorticoid receptor
- AhR
- aryl hydrocarbon receptor
- P450
- cytochrome P450
- EGF
- epidermal growth factor
- TCPOBOP
- 1,4-bis[2-(3,5-dichloropyridyloxy)]benzene
- ERK
- extracellular signal-regulated kinase
- HGF
- hepatocyte growth factor
- U0126
- 1,4-diamino-2,3-dicyano-1,4-bis(phenylthio)butadiene
- CDK
- cyclin-dependent kinase
- U0124
- 1,4-diamino-2,3-dicyano-1,4-bis(aminomethylthio)butadiene
- LY294002
- 2-(4-morpholinyl)-8-phenyl-1(4H)-benzopyran-4-one hydrochloride
- SB203580
- 4-(4-fluorophenyl)-2-(4-methylsulfinylphenyl)-5-(4-pyridyl)1H-imidazole
- SP600125
- anthra[1,9-cd]pyrazole-6 (2H)-one
- SU9516
- 3-[1-(3H-imidazol-4-yl)-meth-(Z)-ylidene]-5-methoxy-1,3-dihydro-indol-2-one
- PDGF
- platelet-derived growth factor
- DMEM
- Dulbecco's modified Eagle's medium
- FCS
- fetal calf serum
- siRNA
- small interfering RNA
- PCR
- polymerase chain reaction
- RXR
- retinoid X receptor
- CPR
- NADPH-cytochrome P450 reductase
- PI3
- phosphatidylinositol 3
- JNK
- c-Jun NH2-terminal kinase
- bp
- base pair
- WY-14643
- 4-chloro-6-(2,3-xylidino)-2-pyrimidinylthioacetic acid
- BSA
- bovine serum albumin.
- Received August 3, 2009.
- Accepted September 28, 2009.
- Copyright © 2010 by The American Society for Pharmacology and Experimental Therapeutics