Abstract
The pregnane X receptor (PXR, NR1I2) plays a pivotal role in the disposition and detoxification of numerous foreign and endogenous chemicals by increasing transcription of numerous target genes, including phase I and II drug-metabolizing enzymes and transporters. In the present study, yeast two-hybrid screening identified an E3 ubiquitin ligase, RBCK1 (Ring-B-box-coiled-coil protein interacting with protein kinase C-1), as a human pregnane X receptor (hPXR)–interacting protein. Coimmunoprecipitation studies confirmed the interaction between RBCK1 and hPXR when both were ectopically expressed in AD-293 cells. Domain mapping studies showed that the interaction between RBCK1 and hPXR involves all RBCK1 domains. We further demonstrate that RBCK1 ubiquitinates hPXR, and this may target hPXR for degradation by the ubiquitin-proteasome pathway. Simultaneous ectopic overexpression of RBCK1 and PXR decreased PXR levels in AD-293 cells, and this decrease was inhibited by the proteasomal inhibitor MG-132 (carbobenzoxy-Leu-Leu-leucinal). Furthermore, overexpression of RBCK1 decreased endogenous levels of PXR in HepG2 cells. Of importance, ectopic overexpression and silencing of endogenous RBCK1 in primary human hepatocytes resulted in a decrease and increase, respectively, in endogenous PXR protein levels and in the induction of PXR target genes by rifampicin. These results suggest that RBCK1 is important for the ubiquitination of PXR and may play a role in its proteasomal degradation.
Introduction
Protein degradation is an essential and versatile housekeeping function in eukaryotic cells that maintains cellular homeostasis. The discovery of the ATP/ubiquitin (Ub)–dependent 26S proteasomal system (Ub/26S) has revolutionized the concept of intracellular protein degradation from a nonspecific scavenger process to a highly controlled and specific cellular process. This process is performed by a complex cascade of enzymes via a three-step mechanism involving the ubiquitin-activating enzyme E1-activating ubiquitin, followed by the ubiquitin-conjugating enzyme E2-mediated transfer of ubiquitin from E1 to a member of the ubiquitin-protein ligase family, E3. E3 enzymes catalyze covalent attachment of ubiquitin to the specific substrate. The ubiquitination of protein serves as a marker for the protease for its eventual degradation (Glickman and Ciechanover, 2002).
RBCK1, RBCC (ring-B-box-coiled-coil) protein interacting with protein kinase C-1 (PKC-1) (C20orf18 or HOIL-1, XAP3, or UIP28), is a transcription factor that consists of a ubiquitin-like sequence (Tokunaga et al., 1998), two coiled-coil regions, a novel zinc finger motif (Meyer et al., 2002), and a RING-IBR (in between RING fingers) domain (Marin and Ferrus, 2002). RBCK1 is localized in both the nucleus and the cytoplasm, possessing a classic Leu-rich nuclear export signal and a nuclear localization signal (Tatematsu et al., 2005). Studies have shown that RBCK1 facilitates transcriptional coactivation after hepatitis B virus infection (Cong et al., 1997) and interacts with various proteins, including UbcM4 E2 ubiquitin ligase (Martinez-Noel et al., 1999), protein kinase C (Tokunaga et al., 1998), cAMP response element-binding protein, and promyelocytic leukemia protein (Tatematsu et al., 2005). It acts as an E3 ligase causing ubiquitin-dependent degradation of heme-oxidized iron regulatory protein-2 in iron metabolism (Yamanaka et al., 2003). The pregnane X receptor (PXR), also known as NR1I2 (nuclear receptor subfamily 1, group I, member 2), is a nuclear receptor that acts as a xenobiotic/metabolite sensor and regulates the expression of a broad array of genes involved in biotransformation and transportation of endogenous substances, natural products, drugs, and other xenobiotic chemicals (Chang, 2009). PXR is predominantly expressed in liver tissue, although it has been detected in small intestine, colon, kidney, brain, and mammary tissues (Bertilsson et al., 1998; Blumberg et al., 1998; Kliewer et al., 1998; Dotzlaw et al., 1999; Masuyama et al., 2001; Miki et al., 2005). PXR target genes include those encoding for various cytochrome P450 enzymes (CYP2A6, CYP2B6, CYP2C8, CYP2C9, CYP3A5, CYP3A7, CYP4F12, and CYP27A1), uridine diphosphate (UDP)–glucuronosyltransferases (UGT1A1, UGT1A3, UGT1A4, UGT1A6, and UGT1A9), sulfotransferases (Sult2a1), and the drug transporter genes ABCB1 (P-glycoprotein), Abcc2, and SLC21A6 (oatp2) (Rosenfeld et al., 2003; Stanley et al., 2006; Ong et al., 2011). The ligand-activated PXR forms a heterodimer with retinoid X receptor α and binds to DNA response elements of PXR target genes, resulting in increased gene transcription (Lehmann et al., 1998; Geick et al., 2001). PXR interacts with various coactivators, such as steroid receptor coactivator-1 and peroxisome proliferator-activated receptor γcoactivator 1-α (Li and Chiang, 2006) and corepressors [e.g., nuclear receptor co-repressor 1 [Roth et al., 2008] and silencing mediator for retinoid or thyroid-hormone receptor) (Johnson et al., 2006) to regulate PXR target genes. PXR transcriptional activity is also influenced by other nuclear receptors (e.g., hepatocyte nuclear factor 4α (Li and Chiang, 2006) and glucocorticoid receptor (Pascussi et al., 2001)], which increase PXR levels. In contrast, small heterodimer partner suppresses PXR activity (Li and Chiang, 2006).
Current efforts to understand the regulation of nuclear receptors have revealed that several steroid hormone receptors, including the glucocorticoid, progesterone, androgen, and estrogen receptors, are tightly regulated by the ubiquitin-proteasome system (Kinyamu et al., 2005). Previous studies have indicated that binding of certain chemicals, some of which could be PXR ligands, increases its half-life, in part because of the disruption of its interaction with the suppressor for gal1, a component of the proteasome (Masuyama et al., 2002, 2005). A recent study demonstrated an increase in ubiquitinated PXR after inhibition of the 26S proteasome with MG-132. Proteasomal inhibition also resulted in the inhibition of PXR transactivation, suggesting interplay between PXR and the ubiquitin pathway (Staudinger et al., 2011).
In the present study, we show that the E3 ligase RBCK1 interacts and ubiquitinates hPXR, and this may in part target hPXR to proteasomal degradation. Because PXR plays an important role in the transcriptional regulation of genes involved in various metabolic pathways, its regulation by RBCK1 has important physiologic and pharmacological consequences.
Materials and Methods
Yeast Two-Hybrid Screening.
ProNet technologies automated two-hybrid screening was performed by Myriad Genetics (Salt Lake City, UT) as described previously (Garrus et al., 2001). Human liver and brain cDNA libraries were used to screen for prey proteins that interact with full-length PXR or PXR domains. Plasmids isolated were transformed into yeast, and interactions were confirmed by liquid β-galactosidase assays. The identities of the prey were determined by DNA sequencing.
Antibodies and Chemicals.
Antibodies to PXR (H160) and RBCK1 (P19) were purchased from Santa Cruz Biotechnologies (Santa Cruz, CA). Mouse monoclonal antibody to PXR was purchased from R&D Systems (Minneapolis, MN). Rat anti-HA antibody (clone 3F10) was purchased from Roche Diagnostics (Indianapolis, IN). MG-132 was obtained from Calbiochem (La Jolla, CA). Rifampicin was purchased from Sigma-Aldrich (St. Louis, MO).
Cell Culture and Transfections.
AD-293 cells (Agilent Technologies, Santa Clara, CA) were grown in Dulbecco’s modified Eagle’s medium supplemented with 10% fetal bovine serum. HepG2 cells were cultured in Eagle’s minimum essential medium supplemented with 10% fetal bovine serum, 2 mM l-glutamine, 1 mM sodium pyruvate, and penicillin-streptomycin at 37°C under 5% CO2. Primary human hepatocytes were obtained from CellzDirect/Life Technologies (Durham, NC) and maintained in William’s E medium supplemented with ITS+1 (insulin, human transferrin, sodium selenite, bovine serum albumin, and linoleic acid in Earle’s balanced salt solution; Sigma-Aldrich), HEPES, l-glutamine, and 100 nM dexamethasone. All transient transfections were performed as described in the Lipofectamine 2000 protocol (Invitrogen, Carlsbad, CA). Twenty-four hours later, medium was replaced and drugs were added at the appropriate concentrations (0.1% dimethylsulfoxide (DMSO) or 10 μM rifampicin). Cells were then treated with 10 μM MG-132 or vehicle for 6 hours after 42 hours of transfection.
Plasmids and siRNAs.
Full-length RBCK1 was cloned from MGC clone EX-U1469-M02 into pcDNA3.1 vector. Full-length RBCK1 (1-470 aa) and the deletion constructs RBCK1-I (1-260 aa), RBCK1-II (172-350 aa), and RBCK1-III (254-470 aa) were then cloned in pGEX-4T1. All constructs were verified by sequencing. PXR was cloned in pCR3 by PCR amplification as described earlier (Chen et al., 2005). pMT123-HA-Ub was a kind gift from Dr. Andrew Wallace (NC State University, Raleigh, NC). For silencing RBCK1 expression, RBCK1 Stealth Select RNAi siRNA siRBCK1-I and Stealth RNAi siRNA Negative Control Hi GC (12935-400) as control siRNA (siC) were purchased from Invitrogen. siRNA transfections were performed using a final concentration of 50–100 nM oligo (at 40–60% confluence) using Lipofectamine 2000 transfection reagent (Invitrogen) according to the manufacturer’s recommendation.
GST Pulldown Assays.
GST fusion proteins of full length and RBCK1 deletion constructs were expressed in Escherichia coli BL-21 and were purified on GSH-Sepharose beads (Amersham, Piscataway, NJ). The PXR protein was translated in vitro in rabbit reticulocyte lysate (Promega, Madison, WI), using a TNT Quick coupled transcription and translation system with radiolabeled [35S]methionine (MP Biomedicals, Solon, OH). Expression and purification of recombinant GST fusion proteins and interaction assays were as described previously (Surapureddi et al., 2008).
Immunoblotting.
Forty-eight hours after transfection, cells were lysed using M-PER mammalian protein extraction reagent (Thermo Fisher Scientific, Waltham, MA) supplemented with complete protease inhibitor mixture (Roche, Nutley, NJ). After a Bradford protein assay (Bio-Rad Laboratories, Hercules, CA) to determine protein concentration, equal amounts (30 μg) of whole cell extracts were resolved by Nu-PAGE gel electrophoresis, transferred to a nitrocellulose membrane (Invitrogen), and analyzed using the appropriate antibodies, including actin as a loading control. Proteins were detected using horseradish peroxidase-conjugated secondary antibodies and a SuperSignal West Femto kit (Thermo Fisher Scientific). Membranes were stripped between antibody probes using Restore Western Blot Stripping Buffer (Thermo Fisher Scientific).
Immunoprecipitations and Ubiquitination Assays.
Cells were cotransfected with PXR and RBCK1. After 48 hours, whole-cell lysates were prepared using M-PER mammalian protein extraction reagent (Thermo Fisher Scientific) supplemented with complete protease inhibitor mixture (Roche). The lysates were immunoprecipitated using agarose beads coupled with rabbit anti-PXR or goat anti-RBCK1 antibodies using the manufacturer’s protocol. The eluted proteins were run on a 10% NuPAGE gel (Invitrogen) and transferred to a nitrocellulose membrane. The membrane was blocked with 5% nonfat milk and probed with mouse anti-PXR or goat anti-RBCK1 IgG (1:1000) for 2 hours at room temperature. After washing, the blot was incubated for 1 hour with appropriate horseradish peroxidase-conjugated secondary antibody (Promega). Detection was achieved using a SuperSignal West Femto kit. Similarly, endogenous PXR protein from HepG2 cells was immunoprecipitated using rabbit anti-PXR antibody, and immunoprecipitates were probed with mouse anti-PXR or goat-RBCK1 antibody. For ubiquitination assays, cells were transfected with PXR, RBCK1, and HA-tagged ubiquitin. Cell lysates were prepared as described previously and immunoprecipitated with rabbit-PXR antibody, and eluted proteins were probed with anti-HA-Ub antibody.
Quantitative Polymerase Chain Reaction.
Total RNA was extracted using RNeasy mini prep system (Qiagen, Valencia, CA) according to the manufacturer's procedure. Reverse-transcription PCR analysis was performed as described earlier (Rana et al., 2010). In brief, cDNAs were prepared using Superscript II (Invitrogen) reverse transcriptase. Quantitative polymerase chain reaction (qPCR) analyses were then performed on an ABI 7900 using TaqMan Universal PCR Master Mix (Applied Biosystems, Bedford, MA) TaqMan gene expression assays for CYP2C9, CYP3A4, UGT1A9, ABCB1, GSTA2, ABCC2, with TATA-binding protein (TBP) as the internal control. The relative quantities were calculated as 2−∆∆Ct.
Statistical Analysis.
For qPCR results and the effect of ectopic RBCK1 expression in AD-293 cells and HepG2 cells) pairwise comparisons were made using the Holm-Sidak method subsequent to two-way analysis of variance in Sigma Plot, version 11. Values for protein expression of Ubiquitin, PXR, and RBCK1 and the effects of siRBCK1 were compared using a paired, two-tailed t test in GraphPad Prism (Graphpad, La Jolla, CA).
Results
RBCK1 Is a Novel Interactor of PXR.
For the first time, we identified RBCK1 as a PXR-interacting protein in a yeast two-hybrid screening. Fourteen different full-length or partial domains of PXR were used as baits. A human liver cDNA library was used as the prey to identify PXR-interacting proteins. Five of 14 baits showed interaction with various proteins, including some known PXR-interacting proteins, such as retinoid X receptor α, SRC1, and GRIP1. In addition, the PXR domain containing amino acids 107-437 from a brain library interacted with RBCK1 (also known as C20orf18 or HOIL-1, XAP3, and UIP2). RBCK1 is an E3 ubiquitin ligase and is known to target certain proteins for degradation in a proteasomal-dependent manner. To confirm that RBCK1 is a PXR-binding protein, we first ectopically transfected AD-293 cells (which do not normally express PXR) with plasmid expression constructs for PXR and RBCK1. Cells were lysed and subjected to immunoprecipitation with nonspecific IgG or antibody to PXR or RBCK1 and subjected to Western blot analysis. As shown in Fig. 1A, RBCK1 was immunoprecipitated with PXR antibody, and conversely, PXR was immunoprecipitated with RBCK1 antibody. The interaction of RBCK1 with PXR was further examined by GST pull-down assays. We generated a series of RBCK1 deletion constructs (Fig. 1B), including RBCK1-I [1-260 aa, containing NZF (novel zinc finger) and CC (coiled-coil)], RBCK1-II (172-350 aa, containing NZF, CC and RING1), and RBCK1-III (254-470 aa, containing –RING and IBR). GST pull-down assays indicated that all three constructs interacted with PXR (Fig. 1C). These results suggest that both the RING1 and the coiled coil domains contribute to the interaction of RBCK1 interaction with PXR.
RBCK1 is identified as a novel PXR-interacting protein. (A) RBCK1 interaction with PXR. AD-293 cells were transfected with ectopic PXR and RBCK1 constructs as indicated. The cell extracts were immunoprecipitated using PXR antibody and probed with RBCK1 (left panel), and reciprocally, the extracts were immunoprecipitated with RBCK1 antibody and probed with PXR antibody (right panel). (B) Schematic representation of RBCK1 along with generated fragments for domain mapping with PXR. The numbers indicate the amino acid (aa) residues and the length of the recombinant protein fragments. RBCK1 consists of the N-terminal ubiquitin-like (UBL) domain, two coiled-coil (CC) domains, a novel zinc finger (NZF) domain, two RING fingers, and an in-between-RING fingers (IBR) domain. (C) Full-length RBCK1 (FL) and the three fragments were expressed as GST fusion proteins and incubated with full-length in vitro-translated 35S-labeled PXR.
RBCK1 Downregulates PXR Protein Levels in a Proteasome-Dependent Manner.
It has been shown previously that RBCK1 acts as an E3 ubiquitin ligase and regulates the expression of various proteins involved in iron metabolism and inflammatory signaling pathways in a proteasome-dependent manner (Tian et al., 2007; Zhang et al., 2008). Here, we examined the possible role of RBCK1 in regulating PXR protein expression. We transfected AD-293 cells with expression constructs for RBCK1, PXR, and HA-tagged ubiquitin. Western blot analysis indicated that ectopic overexpression of RBCK1 specifically down-regulates ectopic expression of PXR protein (Fig. 2A). To indicate whether RBCK1 regulates PXR in a proteasome-dependent manner, we transfected AD-293 cells with ubiquitin, PXR plus ubiquitin, or PXR, ubiquitin, and RBCK1 for 48 hours. During the last 6 hours, we treated the transfected cells with the proteasomal inhibitor, MG-132 (10 μM), or vehicle control. Treatment with MG-132 increased PXR protein levels in the PXR-transfected cells and partially reversed the effect of RBCK1 on PXR levels in cells transfected with RBCK1, PXR, and ubiquitin (Fig. 2B). These results suggest that RBCK1 down-regulates PXR through the proteasomal-dependent pathway. Because RBCK1 is a known E3 ubiquinating ligase, we examined whether RBCK1 ubiquitinates PXR using AD-293 cells transfected with expression plasmids for RBCK1, PXR, and HA-ubiquitin. Cell lysates were immunoprecipitated with anti-PXR antibody, and the inputs and immunoprecipitates were analyzed using Western blot with anti-HA antibody to detect HA-tagged ubiquitin (Fig. 2C). These results indicate that RBCK1 targets PXR protein for the proteasomal pathway by ubiquitination.
RBCK1 regulates PXR protein expression in a proteasome-dependent manner, and degradation is not affected by treatment with the PXR ligand rifampicin (Rif). (A) RBCK1 down-regulates ectopically expressed PXR protein in AD-293 cells. Cells were transfected with PXR, HA-ubiquitin (HA-Ub), and RBCK1 as indicated. After 48 hours, cell lysates were analyzed by Western blotting with anti-PXR, anti-RBCK1, or β-actin antibodies. (B) The decrease of PXR expression in cells cotransfected with RBCK1 was partially blocked by pretreatment with MG-132, a proteasomal inhibitor. AD-293 cells were transfected with the indicated plasmids, and after 40 hours, cells were treated with 10 μM MG-132 for 6 hours. Expression levels of PXR protein were compared using Western blot analysis. (C) RBCK1 ubiquitinates PXR. AD-293 cells were transfected with PXR, PXR and HA-ubiquitin, or PXR, HA-ubiquitin, and RBCK1 as indicated. After 40 hours, cells were treated with MG-132 (10 μM) for 6 hours. Cell lysates were immunoprecipitated with PXR antibody (IP-PXR ab), subjected to gel electrophoresis, and probed with anti-HA antibody to detect ubiquitination of PXR. The ubiquinated ladder was more pronounced when the cells were transfected with PXR, ubiquitin, and RBCK1, compared with those transfected with PXR and ubiquitin alone. (D) The PXR ligand rifampicin had no effect on RBCK1-mediated ubiquitination of PXR protein. AD-293 cells were transfected with PXR, HA-ubiquitin, and RBCK1 plasmids as indicated. After 24 hours, cells were treated with 10 μM rifampicin or DMSO. Cell lysates were immunoprecipitated using PXR antibody and analyzed for polyubiquitinated PXR protein using anti-HA antibody. (E) Western blot analysis showed that RBCK1 decreased the expression of PXR, and this decrease was independent of rifampicin treatment. AD-293 cells were transfected with the indicated plasmids. After 24 hours, cells were treated with 10 μM rifampicin (+) or DMSO controls (−). Cell lysates were subjected to Western blot analysis using antibodies to PXR or β-actin. (F) Densitometric analysis of the protein bands on Western blots relative to actin loading controls indicated that the relative amounts of ectopically expressed PXR were significantly decreased (**P < 0.01; ***P < 0.001) when PXR was coexpressed with RBCK1. Data represent the means ± S.E. (n = 3). IP, Immunoprecipitation.
To investigate whether PXR ligands affect RBCK1 regulation of PXR, we coexpressed PXR, RBCK1, and HA-ubiquitin in AD-293 cells in the presence or absence of a prototypical human PXR ligand (10 μM rifampicin). RBCK1 increased ubiquitination of PXR, as demonstrated by Western blot analysis, but rifampicin had no effect on the ubiquitination of PXR by RBCK1 (Fig. 2D). In addition, RBCK1 transfection decreased PXR protein levels, as demonstrated by Western blotting and subsequent densitometric analysis (Fig. 2, E and F), but rifampicin had no significant effect on PXR degradation.
Overexpression of RBCK1 Down-Regulates PXR Genes in HepG2 Cells and Primary Human Hepatocytes.
We then investigated the effect of RBCK1 on endogenous PXR protein levels in HepG2 cells. A statistically significant decrease in endogenous levels of PXR protein was observed by Western blot analysis followed by densitometric analysis when RBCK1 was ectopically over-expressed in HepG2 cells (P < 0.001), but rifampicin had no further effect on this decrease (Fig. 3, A and B). HepG2 cells were also lysed and subjected to immunoprecipitation with nonspecific IgG or antibody to PXR and analyzed by Western blot analysis using antibody to RBCK1. Western blot analysis of the immunoprecipitates showed that endogenous RBCK1 was immunoprecipitated with antibody to PXR, showing interaction of the endogenous levels of PXR and RBCK1 in HepG2 cells (unpublished data).
RBCK1 down-regulates expression of PXR protein in HepG2 cells and human hepatocytes and PXR target mRNAs in primary human hepatocytes. (A) HepG2 cells were transfected with RBCK1 plasmid or the empty vector pcDNA3.1 and treated with DMSO (−) or rifampicin (10 μM) (+) for 24 hours. Cell lysates were analyzed by Western blot analysis using antibodies for PXR or β-actin as a loading control. Ectopic expression of RBCK1 resulted in decreased endogenous levels of PXR protein and there was no further effect of rifampicin treatment. (B) Densitometric scanning of the blots indicated a significant decrease in PXR levels after expression of RBCK1 relative to actin controls (P < 0.001). (C) Ectopic over-expression of RBCK1 in primary human hepatocytes decreases endogenous PXR levels. Primary hepatocytes were transfected with RBCK1 and PXR and RBCK1 protein expression relative to actin were analyzed by Western blot 48 hours later of a pooled sample from three wells of a 24-well plate. Results indicate a single analysis. (D--I) Overexpression of RBCK1 also decreases the induction of mRNA of PXR target genes in primary human hepatocytes as measured by qPCR. Values are expressed relative to TBP internal control. The relative quantities were calculated by the 2-∆∆Ct method with rifampicin values expressed relative to vehicle controls set to one. Black bars represent cells treated with rifampicin, and white bars cells treated with the vehicle DMSO. Data represent means ± S.E. (n = 3). Rifampicin-treated cells had significantly higher amounts of the indicated mRNAs than did DMSO controls (**P < 0.01, ***P < 0.001). Rifampicin induction in RBCK1-transfected cells was significantly lower than in control cells (†P < 0.05, ††P < 0.01, †††P < 0.001).
Of importance, in primary human hepatocytes, endogenous RBCK1 and PXR were clearly detected by Western blot analysis of three pooled wells from a 24-well plate, and ectopic overexpression of RBCK1 decreased endogenous PXR protein expression (Fig. 3C). Using qPCR as described in methods, we examined the effect of RBCK1 on the expression of mRNAs for PXR-target genes that are involved in metabolizing xenobiotic compounds, including phase I and II enzymes and transporter genes. Overexpression of RBCK1 significantly reduced rifampicin-mediated upregulation of mRNA for various PXR-regulated genes, such as CYP2C9 (47% decrease), CYP3A4 (44% decrease), UGT1A9 (38% decrease), GSTA2 (30% decrease), and two of the multidrug resistance reporters, ABCB1 (P-glycoprotein; 33% decrease) and ABCC2 (32% decrease).
Silencing RBCK1 Up-Regulates Expression of PXR Target Genes in Primary Human Hepatocytes.
We then silenced endogenous RBCK1 expression in primary human hepatocytes. We used two siRBCK1 oligos along with si-control (si-C) at two different concentrations (50 and 100 ng). As indicated by Western blot analysis in Fig. 4, A and B, 100 ng of siRBCK1-1 effectively down-regulated RBCK1 expression significantly (P < 0.05). Endogenous PXR levels were increased 2–3-fold when RBCK1 expression was silenced with 50 or 100 ng of siRBCK1, compared with untransfected cells or those transfected with si-C (Fig. 4, A and B). In addition, silencing RBCK1 significantly enhanced rifampicin induction of the PXR target genes CYP2C9 and CYP3A4 (phase I enzymes), UGT1A9 and GSTA (phase II enzymes), and ABCB1 and ABCC2 (transporters) in primary human hepatocytes (Fig. 4, C–H).
Silencing endogenous RBCK1 protein in primary human hepatocytes increases PXR expression and rifampicin induction of PXR target genes. (A) Western blot analysis of RBCK1, PXR, and actin in primary human hepatocytes transfected with 50 or 100 ng of RBCK1 siRNA oligos (siRBCK1-I), or control oligos (si-C). (B) Quantitation was performed by densitometry of Western blots from two different experiments and normalized to actin (N = 2). The effects of siRBCK1 on RBCK1 levels were compared using a paired, two-tailed t test in GraphPad Prism. *P < 0.05. (C–H) Silencing RBCK1 also enhanced rifampicin induction of PXR target genes. Primary human hepatocytes were transfected with 100 ng of siRBCK1-I or si-C and treated with 10 μM rifampicin or DMSO for 24 hours. mRNA for PXR target genes was analyzed by qPCR as above. Black bars indicate rifampicin-treated cells, and white bars represent DMSO controls. Values are expressed relative to TBP and then normalized to untreated controls. Rifampicin increased the indicated mRNAs over the levels in DMSO controls [means ± S.E. (n = 3); ***P < 0.001], whereas treatment with siRBCK1 significantly increased rifampicin induction over si-C-treated controls. ††P < 0.01, †††P < 0.001.
Discussion
In the present study, RBCK1, an E3 ubiquitin ligase, was shown to interact with PXR in yeast two hybrid screens, GST pull-down assays, and immunoprecipitation studies. Of importance, we describe the involvement of RBCK1 in the ubiquination of PXR presumably targeting it for degradation by the ubiquitin-proteasome system. PXR, a xenobiotic-sensing receptor, up-regulates a large number of drug-metabolizing enzymes and transporters (Tolson and Wang, 2010; Wang et al., 2011). PXR has also been reported to be involved in lipid homeostasis and atherosclerosis (Sui et al., 2011). A recent study demonstrated that PXR activation induces tumor aggressiveness in humans and mice through increased cellular proliferation and drug resistance (Wang et al., 2011). Although PXR regulates many target genes, its regulation is not fully understood. Here, we identified RBCK1 as a novel interacting protein of PXR, which is involved in the regulation of hPXR protein expression. Domain mapping studies indicate that PXR exhibits strong interaction with full-length RBCK1 and the domain containing the NZF, CC, and RING-IBR region. Previously, studies have shown that the N terminus of RBCK1, containing the NZF and CC domains, is essential for its interaction with various proteins, such as TAB2/3 (Tian et al., 2007) and IRF3 (Zhang et al., 2008), but show some interaction with other domains. Our data show that the N-terminal region of RBCK1 containing a ubiquitin-like domain, NZF, and CC interacts with PXR and the CC and RING-IBR regions. The RING-IBR region was previously shown to be required for ubiquitin ligase activity (Tian et al., 2007). Similar observations were also reported in a study involving another E3 ubiquitin ligase protein, Parkin, which interacts with its target protein, Septin 5 (SEPT5_v2), through either the N-terminal ubiquitin-like domain or the RING-IBR region (Choi et al., 2003). Future studies would be required to determine whether the binding of PXR to this region is also important for the ability of RBCK1 to ubiquitinate PXR.
Our results show that RBCK1 down-regulates PXR protein expression. RBCK1 acts as an E3 ligase, facilitating ubiquitination, and, thus, presumably targets PXR for eventual degradation. The proteasomal inhibitor MG-132 caused a marked increase in PXR protein levels when RBCK1, PXR, and ubiquitin were ectopically expressed in AD-293 cells, suggesting a role for RBCK1 in proteasomal degradation of PXR. However, we failed to completely restore PXR protein expression with MG-132. Similar results were also observed for the effect of RBCK1 on the degradation of IRF3 under experimental conditions similar to those in our study (Zhang et al., 2008). Our results suggest that RBCK1 down-regulates PXR expression by promoting proteasomal-dependent degradation, but we cannot exclude degradation by other mechanisms or effects on synthesis. Pulse-labeling experiments will be required to determine whether RBCK1 increases degradation of PXR. Our results are consistent with previous studies demonstrating that RBCK1 acts as an E3 ubiquitin ligase that regulates the degradation of heme-oxidized iron regulatory protein-2 (Yamanaka et al., 2003). Other studies have shown that RBCK1 is involved in negative regulation of TNF and IL-1–induced inflammatory signaling by targeting TGF-β-activated kinase 1-binding protein 2/3 (TAB2/3) for degradation (Tian et al., 2007).
One study (Masuyama et al., 2002) reported that some steroids (progesterone and dexamethasone) modify PXR protein levels through the proteasomal pathway in the mouse. Although the authors suggest that these compounds might be PXR ligands, they are not prototypical PXR ligands. We used the prototype human PXR agonist rifampicin to analyze its potential to affect PXR degradation in AD-293 cells or HepG2 cells. We found that ubiquitination of PXR in AD-293 cells was unaffected by the PXR ligand rifampicin after ectopic expression of PXR and RBCK1 in this cell line. In addition, over-expression of RBCK1 in HepG2 cells caused downregulation of endogenous levels of PXR independent of rifampicin treatment. Our findings are consistent with those of Sheflin et al. (2000) who found that the androgen receptor is degraded via the proteasome in the absence of ligand.
Of importance, RBCK1 decreases endogenous levels of PXR in primary human hepatocytes. Because human PXR is an essential regulator of a large number of drug disposition genes (Staudinger et al., 2001a,b; Maglich et al., 2002; Rosenfeld et al., 2003; Hartley et al., 2006), we examined the impact of overexpression and silencing of RBCK1 on induction of various PXR-regulated genes in primary human hepatocytes by rifampicin, including phase I enzymes, such as P450 enzymes; phase II enzymes, such as UGTs, glutathione S-transferases, sulfotransferases; and two members of the ATP-binding cassette (ABC) multidrug transistance transporters, ABCC2 and ABBC1 (P-glycoprotein). Western blot analysis indicated that overexpression of RBCK1 down-regulated endogenous PXR levels in primary human hepatocytes, and silencing the expression of RBCK1 by RNAi resulted in increased PXR levels. RBCK1-mediated decreases in PXR protein levels down-regulated the rifampicin-mediated PXR induction of CYP2C9, CYP3A4, UGT1A9, SULT2A1, GSTA2, ABCB1, and ABCC2 mRNAs in primary human hepatocytes. Conversely, silencing endogenous RBCK1 expression in primary human hepatocytes resulted in increased induction of PXR target genes by rifampicin.
In conclusion, we identified RBCK1 as a new PXR-interacting protein that regulates PXR protein expression via increased ubiquitination and presumably targets it for proteasomal degradation in human primary hepatocytes and other cell lines. Ectopic expression of RBCK1 increased PXR degradation and, thus, decreased induction of various genes involved in drug metabolism. Silencing RBCK1 in primary hepatocytes increased rifampicin induction of these genes. Our study provides an important insight in the role of RBCK1 in the ubiquitination and, presumably, the targeting of PXR for degradation by the proteasomal system and, thereby, the regulation of PXR and various PXR target genes.
Authorship Contributions
Participated in research design: Rana, Kinyamu, Goldstein.
Conducted experiments: Rana, Coulter.
Performed data analysis: Rana, Coulter.
Wrote or contributed to the writing of the manuscript: Rana, Goldstein, Kinyamu, Coulter.
Footnotes
This research was supported by the Intramural Research Program of the National Institutes of Health National Institute of Environmental Health Sciences [Grants Z01ES02124 and Z01ES071006-12].
Abbreviations
- aa
- amino acids
- ABC
- ATP-binding cassette
- CC
- coiled-coil
- E3
- E3 ubiquitin ligase
- DMSO
- dimethylsulfoxide
- GRIP-1
- glucocorticoid receptor interacting protein/steroid receptor coactivator-2
- GST
- glutathione S-transferase
- HA
- hemagglutinin
- hPXR
- human pregnane X receptor
- IBR
- in between RING fingers
- MG-132
- carbobenzoxy-Leu-Leu-leucinal
- NZF
- novel zinc finger
- P450
- cytochrome P450
- PKC-1
- protein kinase C-1
- PXR
- pregnane X receptor
- RBCC
- ring-B-box-coiled-coil
- RBCK1
- RBCC protein interacting with protein kinase C1
- TBP
- TATA-binding protein
- Received August 30, 2012.
- Accepted October 13, 2012.
- Copyright © 2013 by The American Society for Pharmacology and Experimental Therapeutics
References
In this issue
An E3 Ligase RBCK1 Ubiquinates hPXR
