1521-009X/44/1/137–150$25.00 http://dx.doi.org/10.1124/dmd.115.066308DRUG METABOLISM AND DISPOSITION Drug Metab Dispos 44:137–150, January 2016Copyright ª 2015 by The American Society for Pharmacology and Experimental Therapeutics

Threonine-408 Regulates the Stability of Human Pregnane XReceptor through Its Phosphorylation and the CHIP/

Chaperone-Autophagy Pathway s

Junko Sugatani, Yuji Noguchi, Yoshiki Hattori, Masahiko Yamaguchi, Yasuhiro Yamazaki,and Akira Ikari1

Department of Pharmaco-Biochemistry, School of Pharmaceutical Sciences, University of Shizuoka, Japan

Received July 15, 2015; accepted November 2, 2015

ABSTRACT

ThehumanpregnaneX receptor (hPXR) is a xenobiotic-sensing nuclearreceptor that transcriptionally regulates drug metabolism–relatedgenes. The aim of the present study was to elucidate the mechanismby which hPXR is regulated through threonine-408. A phosphomimeticmutationat threonine-408 (T408D) reduced the transcriptional activity ofhPXR and its protein stability in HepG2 and SW480 cells in vitro andmouse livers in vivo. Proteasome inhibitors (calpain inhibitor I andMG132) and Hsp90 inhibitor geldanamycin, but not Hsp70 inhibitorpifithrin-m, increased wild-type (WT) hPXR in the nucleus. The trans-location of the T408D mutant to the nucleus was significantly reducedeven in the presence of proteasome inhibitors, whereas the complex ofyellow fluorescent protein (YFP)-hPXR T408D mutant with heat shockcognate protein 70/heat shock protein 70 and carboxy terminusHsp70-interacting protein (CHIP; E3 ligase) was similar to that of the WT in the

cytoplasm. Treatment with pifithrin-m and transfection with anti-CHIPsmall-interfering RNA reduced the levels of CYP3A4 mRNA induced byrifampicin, suggesting the contribution of Hsp70 and CHIP to thetransactivation of hPXR. Autophagy inhibitor 3-methyladenine accu-mulated YFP-hPXR T408D mutant more efficiently than the WT in thepresence of proteasome inhibitor lactacystin, and the T408D mutantcolocalized with the autophagy markers, microtubule-associated pro-tein 1 light chain 3 and p62, which were contained in the autophagiccargo. Lysosomal inhibitor chloroquine caused the marked accumula-tion of the T408D mutant in the cytoplasm. Protein kinase C (PKC)directly phosphorylated the threonine-408 of hPXR. These resultssuggest that hPXR is regulated through its phosphorylation atthreonine-408 by PKC, CHIP/chaperone–dependent stability check,and autophagic degradation pathway.

Introduction

The pregnane X receptor (PXR) is a member of the nuclear receptorsuperfamily and is recognized as a xenobiotic sensor that regulates thetranscriptional expression of phase I, II, and III metabolic enzymes andtransporters involved in the metabolism and elimination of potentiallytoxic substances (reviewed in Timsit and Negishi, 2007; Staudingerand Lichti, 2008). PXR also plays important roles in many biologicevents such as hepatic energy metabolism, immune/inflammatoryresponses, cell proliferation, bone homeostasis, and the pathogenesis ofinflammatory bowel disease and tumor development (reviewed inKonno et al., 2008; Ihunnah et al., 2011; Zhuo et al., 2014; Rathod et al.,2014; Ma et al., 2015). A previous study demonstrated that PXR was

sequestered in the cytoplasm through the formation of a complexwith heat shock protein (Hsp) 90 (Squires et al., 2004). Upon thebinding of PXR ligands to the ligand binding domain, it translocatesto the nucleus, in which it forms a heterodimer with retinoid Xreceptor and binds to the xenobiotic-response element of the targetgene, resulting in transcriptional activation (Kawana et al., 2003;Squires et al., 2004).Post-translational modifications to the human PXR (hPXR) by

phosphorylation and through small ubiquitin-related modifier- andubiquitin-signaling pathways have been shown to influence thefunctions of hPXR by modulating intracellular localization andnuclear activation (Lichti-Kaiser et al., 2009; Pondugula et al., 2009;Staudinger et al., 2011; Sugatani et al., 2012, 2014; Smutny et al., 2013;Elias et al., 2014; Cui et al., 2015). Cyclin-dependent kinase 2 has beenshown to phosphorylate hPXR at serine-350 to suppress the bindingwith retinoid X receptor by maintaining the acetylation of the hPXRprotein, thereby downregulating hPXR activity (Lin et al., 2008;Sugatani et al., 2012). We previously reported that phosphomimeticmutations in hPXR at threonine-290 and threonine-408 suppressed thetranslocation of hPXR protein to the nucleus even after stimulation withroscovitine, resulting in reduced hPXR activities (Sugatani et al., 2012).

This work was supported in part by Grant-in-Aid for Scientific Research [Grants21590170 and 22590068] from the Ministry of Education, Culture, Sports, Scienceand Technology, Japan.

1Current affiliation: Department of Biochemistry, Faculty of PharmaceuticalScience, Gifu Pharmaceutical University, Japan.

dx.doi.org/10.1124/dmd.115.066308.s This article has supplemental material available at molpharm.aspetjournals.

org.

ABBREVIATIONS: 3-MA, 3-methyladenine; CAR, constitutive androstane receptor; CHIP, carboxy terminus Hsp70-interacting protein; DAPI, 4,6-diamidino-2-phenylindole dihydrochloride; DYRK2, dual-specificity tyrosine-(Y)-phosphorylation–regulated kinase 2; GFP, green fluorescentprotein; GR, glucocorticoid receptor; HOP, stress-induced phosphoprotein 1; hPXR, human pregnane X receptor; Hsc70, heat shock cognateprotein 70; Hsp, heat shock protein; LC3, microtubule-associated protein 1 light chain 3; MBP, maltose binding protein; PKC, protein kinase C;RBCK1, Ring-B-box-coiled-coil protein interacting with protein kinase C-1; siRNA, small-interfering RNA; UBR5, ubiquitin protein ligase E3component n-recognin 5; UGT, UDP-glucuronosyltransferase; YFP, yellow fluorescent protein.

137

http://dmd.aspetjournals.org/content/suppl/2015/11/03/dmd.115.066308.DC1Supplemental material to this article can be found at:

at ASPE

T Journals on M

ay 28, 2020dm

d.aspetjournals.orgD

ownloaded from

Furthermore, we revealed that Ca2+/calmodulin-dependent proteinkinase II (CaMKII) directly phosphorylated hPXR at threonine-290for its retention in the cytoplasm, and protein phosphatase 1 wasrecruited to the hPXR-Hsp90 complex and dephosphorylated hPXR inligand-stimulated HepG2 cells (Sugatani et al., 2014). Since thestability of hPXR is known to be regulated through a proteasomedegradation pathway (Rana et al., 2013; Ong et al., 2014) and the hPXRT408D mutant accumulates in the cytoplasm in the presence ofproteasome inhibitor MG132 (Sugatani et al., 2012), the phosphoryla-tion of hPXR at threonine-408 appears to be involved in the regulationof its stability. However, the molecular mechanisms regulating thestability and function of the T408D mutant have not been elucidated indetail (Sugatani et al., 2012).Most cellular proteins in eukaryotic cells are degraded through two

major pathways: the ubiquitin-proteasomal pathway and an autophagicprocess delivering cytoplasmic proteins and components to thelysosome (reviewed in Abounit et al., 2012; Lilienbaum, 2013). Heatshock proteins Hsc70/Hsp70 assist in the refolding of cytoplasmicproteins with cochaperone proteins such as the carboxy terminusHsp70-interacting protein (CHIP) (Kampinga et al., 2003). Chaperone-dependent E3 ligase CHIP leads to the ubiquitination of misfoldedproteins, which is facilitated by chaperone proteins to proteolyticdegradation through an autophagic process (Kampinga et al., 2003;Morishima et al., 2008). To elucidate the role of phosphorylation atthreonine-408, we herein systematically characterized and comparedthe expression of the hPXR T408D mutant and its interactions withchaperone proteins in the cytoplasm and nucleus with the wild-type(WT) in the presence and absence of proteasome inhibitors and anHsp90 inhibitor, which induced chaperone protein Hsp70 and causedautophagy.We revealed that 1) the hPXR T408Dmutant was expressedto the same extent as the WT, but was cleared more rapidly fromcultured cells in vitro and mouse livers in vivo, and that 2) the hPXRT408D mutant formed complexes with chaperone proteins Hsp90–Hsc70/Hsp70, as well as cochaperone proteins such as CHIP, stress-induced phosphoprotein 1 (HOP), and p23 in the cytoplasm, as did theWTin HepG2 cells. Furthermore, the autophagy inhibitor, 3-methyladenine(3-MA) resulted in an increase in autophagy markers such asmicrotubule-associated protein 1 light chain 3 (LC3) and p62/SQSTM1, which were colocalized with the hPXR T408D mutant, andled to the accumulation of higher levels of the hPXR T408D mutantthan the WT in the presence of lactacystin. Thus, the results presentedhere provide novel insights into the molecular mechanisms regulatingthe stability and function of hPXR through its phosphorylation atthreonine-408 and the CHIP/chaperone–autophagy system in thecytoplasm.

Materials and Methods

Materials. MG132, calpain inhibitor I, lactacystin, and geldanamycin werepurchased from Calbiochem (Darmstadt, Germany). Rifampicin, chloroquinediphosphate, and 3-MA were obtained from Sigma-Aldrich (St. Louis, MO).Pifithrin-m was from Cayman Chemical Company (Ann Arbor, MI), adenosine-S-triphosphate (ATP) from Roche Diagnostics (Mannheim, Germany), and[g-32P]ATP from PerkinElmer (Santa Clara, CA). All other chemicals andsolvents were of analytical grade and obtained from commercial sources.

Plasmids. The hPXR expression vector was generously provided byDr. Masahiko Negishi (National Institute of Environmental Health Sciences,NC). Mutations were induced in hPXR with a QuickChange site-directedmutagenesis kit (Stratagene, La Jolla, CA) according to the manufacturer’sinstructions, and expression vectors were prepared as shown in our previousstudy (Sugatani et al., 2012, 2014).

Cell Culture Conditions, Transfection, and Luciferase Assays. Humanliver-derived cells (HepG2 cells, the RIKEN BioResource Center, Ibaraki,Japan) were seeded on 6-well plates at 2 � 105 cells (2 ml) in Dulbecco’smodified Eagle’s medium supplemented with 10% fetal calf serum (HyClone,GE Healthcare Life Sciences, Logan, UT) and antibiotics (100 mg ofstreptomycin and 10 IU of penicillin/ml) at 37�C in the presence of 5% CO2,unless otherwise stated. Twenty-four hours later, HepG2 cells were transfectedwith UDP-glucuronosyltransferase (UGT1)A1-luciferase reporter constructsincluding the distal (-3570/-3180) and proximal (-165/-1) regions (Sugataniet al., 2001, 2008) (0.2 mg) and expression vectors [pCR3 vector (Invitrogen/Thermo Fisher Scientific, Sunnyvale, CA) (0.2 mg), pCR3-hPXR WT (0.2 mg),pCR3-hPXR T408A (0.2 mg), pCR3-hPXR T408D (0.2 mg), and pRL-SV40(Promega, Madison, WI) (0.2 mg)] using HilyMax reagent (Dojindo Laborato-ries, Kumamoto, Japan) according to the manufacturer’s instructions. Themedium (0.5 ml per well) was replaced after 12 hours with the same medium.Cells were subsequently treated for 24 hours with rifampicin (5 � 1026 M) as400� concentrated stocks in dimethylsulfoxide; controls received an equivalentvolume of dimethylsulfoxide. The total amount of DNA transfected was heldconstant in each transfection by using the corresponding empty vector. Trans-fected cells were washed once in phosphate-buffered saline and harvested in 1�passive lysis buffer, and luciferase activity was measured simultaneously withthe Dual-Luciferase reporter assay system according to the manufacturer’sinstructions (Promega); firefly luciferase values were normalized to Renillavalues for each sample.

Real-Time Quantitative PCR. Since PXR is a key transcriptional regulatorof CYP3A (Lehmann et al., 1998) and human UGT1A1 (Sugatani et al., 2004,2005), we used CYP3A4 and UGT1A1 as biomarkers of the activation of hPXR.HepG2 cells and SW480 human colon cancer cells (American Type CultureCollection) (2 � 105 cells) seeded on 6-well plates and cultured for 24 hours inDulbecco’s modified Eagle’s medium (HepG2 cells) and RPMI 1640 medium(SW480 cells), respectively, supplemented with 10% fetal calf serum (Hyclone)and antibiotics (100 mg of streptomycin and 10 IU of penicillin/ml) weretransfected with expression vectors [pCR3 (0.2 mg), pCR3-hPXR (0.2 mg), orpCR3-hPXR mutant (0.2 mg)] using TransIT-LT1 (Mirus Bio, Madison, WI)according to the manufacturer’s instructions. Cells were given fresh medium

TABLE 1

Sequences of oligonucleotides used

Gene Accession NumberSequence

Forward Primer Reverse Primer

b-Actin NM_001101 59-TGGCACCCAGCACAATGAA-39 59-CTAAGTCATAGTCCGCCTAGAAGCA-39UGT1A1 NM_000463 59-AATAAAAAAGGACTCTGCTATGCT-39 59-ACATCAAAGCTGCTTTCTGC-39CYP3A4 NM_017460 59-GTGGGGCTTTTATGATGGTCA-39 59-GCCTCAGATTTCTCACCAACACA-39Hsp90 NM_001271969 59-GCAGCTCAAGGAATTTGATGG-39 59-GGTGAAGACACAAGTCTATTGGAGA-39Hsp70 NM_006597 59-GGAAATTGCAGAAGCCTACCTTG-39 59-CTTTGGTAGCCTGACGCTGAGA-39Hsc70 NM_005345 59-GGATAACGGCTAGCCTGAGGAG-39 59-CTGGGAAGCCTTGGGACAAC-39HOP NM_001282652 59-GTGCAGCAGATCATGAGTGACC-39 59-TCACCGAATTGCAATCAGACC-39CHIP NM_005861 59-AGGCCAAGCACGACAAGTACAT-39 59-CTGATCTTGCCACACAGGTAGT-39RBCK1 NM_006462 59-CCTTCAGCTACCATTGCAAGACC-39 59- TCCTTGCAGTTCATCTGCTCA-39UBR5 NM_015902 59-ATTTCAATGAGCAGTCAGTCTCAAG-39 59-GTGCCCAGGCTGAGTTTACA-39

138 Sugatani et al.

at ASPE

T Journals on M

ay 28, 2020dm

d.aspetjournals.orgD

ownloaded from

24 hours after being transfected and were then retransfected with the expressionvectors using the same protocol as that for the first transfection. Tenminutes afterthe second transfection, cells were cultured with rifampicin (5 � 1026 M) orvehicle in the presence or absence of various inhibitors for an additional24 hours, unless otherwise stated. Total RNAwas extracted using TRIzol reagentfrom Invitrogen/Thermo Fisher Scientific, and cDNA synthesized from 75 ng oftotal RNA by reverse transcription with a Prime Script reverse transcriptionreagent kit (TaKaRa Bio, Otsu, Japan) was subjected to a quantitative real-timePCR as described previously (Sugatani et al., 2012) using SYBR Premix Ex Taqreagent (TaKaRa Bio) according to the manufacturer’s specifications forUGT1A1 and CYP3A4 (Sugatani et al., 2010), Hsp90, Hsp70, Hsc70, HOP,RBCK1, and UBR5 (TaKaRa Bio), and CHIP (Kajiro et al., 2009). All primersused for real-time PCR were listed in Table 1

Expression of the Flag-Tagged hPXR WT and T408D Mutant Proteinsin Mouse Livers In Vivo. To investigate the stability of the hPXR WT andT408D mutant, plasmids [pCMV-DYKDDDDK-hPXR WT (1 mg) or thepCMV-DYKDDDDK-hPXR T408D mutant (1 mg)] were injected into the tailveins of male imprinting-control-region (ICR) mice (4 weeks old) using theTransIT-QR Hydrodynamic Delivery Starter Kit (Mirus Bio) according to themanufacturer’s instructions. Twelve and 24 hours later, livers were rapidlyexcised and weighed. Whole livers were lysed with a freshly prepared lysisbuffer containing 50 mM Tris-HCl (pH 7.4), 1% NP-40, 0.25% sodiumdeoxycholate, 150 mM NaCl, 1 mM EDTA, 1 mM phenylmethylsulfonylfluoride, 1 mM Na3VO4, 1 mM NaF, 1 mM benzamidine, 1 mg/ml aprotinin,and 1 mg/ml leupeptin at 4�C. The lysates were centrifuged at 14,000g for10 minutes. Protein concentrations were determined using a DC protein assay(Bio-Rad, Hercules, CA).

Western Blot Analysis. HepG2 cells (2� 106 cells) seeded on 75-cm2 flasksand cultured for 24 hours were transfected with expression vectors [pCR3-hPXRWT (10 mg), the pCR3-hPXR T408D mutant (10 mg), pCMV-DYKDDDDK-hPXR WT (10 mg), the pCMV-DYKDDDDK-hPXR T408D mutant (10 mg),pEYFP-hPXR WT (10 mg), or the pEYFP-hPXR T408D mutant (10 mg)] usingTransIT-LT1 (Mirus Bio) according to the manufacturer’s instructions. Cellswere given fresh medium 24 hours after transfection, were further transfectedwith the expression vectors, and then cultured with rifampicin (5 � 1026 M) orvehicle for an additional 24 hours unless stated otherwise. Treated and untreatedcells were washed three times with ice-cold phosphate-buffered saline and lysedwith a freshly prepared lysis buffer containing 50 mM Tris-HCl (pH 7.4), 1%NP-40, 0.25% sodium deoxycholate, 150 mM NaCl, 1 mM EDTA, 1mMphenylmethylsulfonyl fluoride, 1 mMNa3VO4, 1 mMNaF, 1 mM benzamidine,1 mg/ml aprotinin, and 1 mg/ml leupeptin at 4�C. The lysates were centrifugedat 14,000g for 10 minutes. Nuclear extracts and cytoplasmic fractions wereprepared using a nuclear extract kit (Active Motif, Carlsbad, CA). Proteinconcentrations were determined using the DC protein assay (Bio-Rad).Cytoplasmic fractions, nuclear extracts, cell lysates, or liver lysates (20 mg) wereresolved by electrophoresis on an SDS 5–20% polyacrylamide gel (ATTOCorporation, Tokyo, Japan), and electroblotted onto a polyvinylidene difluoridemembrane (MerckMillipore, Darmstadt, Germany).Membranes were blocked at4�C overnight in Blocking One (Nacalai Tesque, Kyoto, Japan) and probed for1 hour with primary antibodies, including the anti-DDDDK (Flag) tag (M185-3L), anti-green fluorescent protein (GFP) (598), anti-LC3 (M152-3), and anti-p62/SQSTM1 (M162-3) from Medical Biologic Laboratories (Nagoya, Japan),anti-Hsp90 (#4874) from Cell Signaling Technology (Danvers, MA), anti-Histone H1 (ab125027), anti-Hsp70 (ab2787), anti-HOP (ab56873), anti-CHIP(ab109103), and anti-p23 (ab2814) from Abcam (Cambridge, England), anti-Hsc70 (SMC-151A/B) from StressMarq Bioscience (Victoria, BC, Canada),anti-ubiquitin (MAB701) from R&D Systems (Minneapolis, MI), anti-PXR(AB10302) fromMerck Millipore, and anti-a-tubulin (CP06) from Calbiochem/Merck Millipore. Antigen-antibody complexes were detected using an appro-priate secondary antibody conjugated to horseradish peroxidase [horseradishperoxidase-conjugated anti-rabbit, anti-goat, or anti-mouse immunoglobulin(Jackson Immuno Research)] and visualized with an enhanced chemilumines-cence system (GE Healthcare, Little Chalfont, England).

Coimmunoprecipitation. HepG2 cells (2 � 106 cells) seeded on 75-cm2

flasks and cultured for 24 hours were transfected with expression vectors[pEYFP-hPXR WT (10 mg) or the pEYFP-hPXR T408D mutant (10 mg)] usingTransIT-LT1 according to the manufacturer’s instructions. Cells were givenfresh medium 24 hours after transfection, further transfected with the expression

vectors, and cultured with rifampicin (5 � 1026 M) or vehicle for an additional24 hours. The cytoplasmic (530 mg) and nuclear (200 mg) proteins, unlessotherwise stated, were precleared with protein G agarose beads and rabbit IgG(Santa Cruz Biotechnology, Dallas, TX) for 30 minutes at 4�C. Yellowfluorescent protein (YFP)-tagged proteins were immunoprecipitated using ananti-GFP antibody combined with protein G agarose beads. The agarose beads

Fig. 1. Transactivation capacity of the hPXR T408D mutant in HepG2 and SW480cells (A–C), and its protein stability in vitro in HepG2 cells (D) and in vivo in mouselivers (E). (A–C) HepG2 and SW480 cells cultured for 24 hours were transfectedwith expression vectors encoding the hPXR WT or T408D mutant. The medium wasreplaced 24 hours after transfection, and cells were retransfected and then culturedwith rifampicin (5 � 1026 M) or vehicle (DMSO) for an additional 24 hours.UGT1A1 reporter activity assessed by the expression of a reporter gene encodingfirefly luciferase (A), and CYP3A4 mRNA levels in HepG2 (B) and SW480 (C)cells treated with rifampicin or vehicle were expressed by taking the control valueobtained from hPXR WT-expressing and vehicle-treated cells as one. Data presentedare the average 6 S.E. of three experiments. **P , 0.01; ***P , 0.001 versus thehPXR WT in cells treated without rifampicin; #P , 0.05; ##P , 0.01; ###P , 0.001versus the hPXR WT in cells stimulated with rifampicin. (D, E) A Western blotanalysis of the Flag-tagged hPXR WT and T408D mutant proteins in HepG2 cellsafter the in vitro transfection (D) and in the mouse livers 12 hours and 24 hours afterthe in situ DNA injection (E). HepG2 cells cultured for 33 hours were transfectedwith expression plasmids for hPXR. The medium was replaced 24 hours aftertransfection, and cells were cultured with MG132 (1 � 1026 M) or vehicle (DMSO)for an additional 15 hours. Protein levels in cells before the treatment with MG132were taken as the level at time 0. Flag-hPXR proteins in HepG2 cell lysates (D) or inliver lysates (E) were determined by Western blotting with the indicated antibodies[anti-Flag and anti-a-tubulin (loading control) antibodies]. (E) Band intensities weremeasured with ImageJ software, and each quantified value of the Flag-hPXR bandswas corrected for that of a-tubulin. Relative levels are expressed by taking theprotein levels obtained from mouse livers expressing Flag-hPXR WT 12 hours afterthe in situ DNA injection as one. Data presented are the average 6 S.E. of three tofive experiments. *P , 0.05 versus mouse livers expressing Flag-hPXR WT24 hours after the in situ DNA injection. Western blottings were performed twicewith similar results, and representative data are shown. WT, wild-type hPXR;T408A, T408A-mutated hPXR; T408D, T408D-mutated hPXR.

Stability Control of PXR via Chaperone-Autophagy Pathway 139

at ASPE

T Journals on M

ay 28, 2020dm

d.aspetjournals.orgD

ownloaded from

were washed three times with Dulbecco’s phosphate buffer and subjected toSDS-PAGE on a 5–20% gradient gel (ATTO Corporation). The proteins weretransferred to an Immobilon-P transfer membrane (Merck Millipore), and theblots were probed with antibodies as indicated.

Small-Interfering RNA–Mediated Protein Knockdown. Small-InterferingRNAs (siRNAs) targeting human CHIP (NM_005861_stealth_725; sense 59-CAGCUGGAGAUGGAGAGCUAUGAUG-39 and antisense 59-CAU-CAUAGCUCUCCAUCUCCAGCUG-39) and the human CHIP control(NM_005861_stealth_control_725; sense 59-CAGAGGUAGAGGGAGAUC-GUCUAUG-39 and antisense 59-CAUAGACGAUCUCCCUCUACCUCUG-39) were obtained from Invitrogen/Thermo Fisher Scientific. HepG2 cellscultured for 24 hours were transfected with siRNA duplexes (100 nM) usingTransIT-siQUEST (Mirus Bio) according to the manufacturer’s instructions.Cells were further transfected with expression vectors [the pCR3-hPXR WT,pCR3-hPXR T408D mutant, pCMV-DYKDDDDK-hPXR WT, or pCMV-DYKDDDDK-hPXR T408D mutant] 12 hours after siRNA transfection, and12 hours after the second transfection were cultured with rifampicin (5� 1026 M)or vehicle for an additional 24 hours, unless stated otherwise. Treated and untreatedcells were washed three times with ice-cold phosphate-buffered saline and lysed witha freshly prepared lysis buffer containing 50 mM Tris-HCl (pH 7.4), 1% NP-40,0.25% sodium deoxycholate, 150 mM NaCl, 1 mM EDTA, 1 mM phenyl-methylsulfonyl fluoride, 1 mM Na3VO4, 1 mM NaF, 1 mM benzamidine, 1 mg/mlaprotinin, and 1 mg/ml leupeptin at 4�C. The lysates were centrifuged at 14,000g for10 minutes. Total RNA was extracted using TRIZOL reagent from Invitrogen/Thermo Fisher Scientific, and cDNA synthesized from 75 ng of total RNA wassubjected to real-time quantitative PCRas described previously (Sugatani et al., 2012).

In Vitro Ubiquitination Assays. The maltose-binding protein (MBP)-hPXRWT and T408D mutant fusion proteins were expressed in NEB ExpressCompetent E. coli (High Efficiency) and purified using amylose resin accordingto the manufacturer’s protocols (New England BioLabs, Boston, MA).Recombinant hPXR prepared by the cleavage of MBP-hPXR with Factor Xa,the recombinant His-tagged hPXR ligand-binding domain (138-434) (Sigma-Aldrich), or recombinant Hsc70 (StressMarq Biosciences) was incubated withUbcH5c (UBE2D3, E2 ubiquitin-conjugating enzyme; Abcam) and CHIP (E3enzyme; R&DSystems) at 37�C for 30 and 60minutes using an in vitro ubiquitinkit (Abcam) according to the manufacturer’s instructions. The amounts of eachsubstrate used were 21–108 ng in 50 ml of the reaction solution. Ubiquitinatedproteins were determined by Western blotting using an anti-ubiquitin antibody.

Colocalization of the Fluorescent Protein–Tagged hPXR WT or T408DMutant and Autophagy Markers P62 and LC3 in HepG2 Cells. To detectthe colocalization of the hPXRWT or T408Dmutant and the autophagy markersp62 and LC3, HepG2 cells (3.5 � 104 /well) cultured on four-well glass slides(Laboratory-TekII Chamber Slide, cat. no. 154526; Thermo Fisher Scientific)for 24 hours were transfected with expression vectors [the pEYFP-hPXR WT(0.5 mg) or pEYFP-hPXR T408D mutant (0.5 mg)] using TransIT-LT1according to the manufacturer’s instructions. Twelve hours after transfection,cells were incubated with 3-methyladenine (5 mM) for 12 hours, and then furthercultured with rifampicin (5 � 1026 M) or vehicle in the presence or absence oflactacystin (5 mM) for an additional 24 hours. Cells were blocked with methanoland incubated with the anti-LC3 (1:250) or anti-p62 (1:250) antibody at roomtemperature for 30 minutes. These cells were counterstained at room temperaturefor 30 minutes with an Alexa Fluor 555-conjugated anti-mouse IgG (H + L)antibody (1:1000) (Thermo Fisher Scientific). Coverslips were mounted withVectashield antifade mounting medium (Vector Laboratories Inc., Burlingame,CA) with 4,6-diamidino-2-phenylindole dihydrochloride (DAPI; Vector Labo-ratories) to visualize nuclei, and images were acquired with a Zeiss LSM510confocal microscope (Carl Zeiss, Oberkochen, Germany).

In Vitro Kinase Assays. The MBP-hPXR WT and T408A mutant fusionproteins were expressed in NEB Express Competent E. coli (High Efficiency)

Fig. 2. Effects of proteasome inhibitors on the transactivation capacity of the hPXRWT and T408D mutant in HepG2 cells. HepG2 cells cultured for 24 hours weretransfected with expression vectors encoding the hPXR WT or T408D mutant. Themedium was replaced 24 hours after transfection, and cells were retransfected andthen cultured with rifampicin (5 � 1026 M) or vehicle (DMSO) in the presence orabsence of calpain inhibitor I (1 � 1025 M) (A), MG132 (1 � 1026 M) (B),lactacystin (5 � 1026 M) (C), or vehicle (DMSO, control) for an additional24 hours. CYP3A4 mRNA levels were expressed by taking the value obtained fromhPXR WT-expressing and vehicle-treated control cells as one. Data presented are

the average 6 S.E. of three experiments. *P , 0.05; **P , 0.01; ***P , 0.001versus the hPXR WT in vehicle-treated control cells; ##P , 0.01; ###P , 0.001versus the hPXR WT in rifampicin-stimulated control cells; ^P , 0.05 versus thehPXR T408D mutant in vehicle-treated control cells; +++P , 0.001 versus thehPXR T408D mutant in rifampicin-stimulated control cells. WT, wild-type hPXR;T408D, T408D-mutated hPXR.

140 Sugatani et al.

at ASPE

T Journals on M

ay 28, 2020dm

d.aspetjournals.orgD

ownloaded from

and purified using amylose resin according to the manufacturer’s protocol (NewEngland Biolabs, Boston, MA). Kinase assays for protein kinase C (PKC;Promega) and DYRK2 (SignalChem Lifesciences, Richmond, BC, Canada)were performed according to the manufacturer’s specifications. PKC consistedprimarily of the a, b, and g isoforms with lesser amounts of the d and z isoforms(Walton et al., 1987). Briefly, kinase reactions were performed in 30 ml of thekinase buffer [for PKC, 20 mM HEPES (pH 7.4), 3.4 mM CaCl2, and 10 mMMgCl2; for DYRK2, 25 mM MOPS (pH 7.2), 12.5 mM b-glycerol-phosphate,25 mM MgCl2, 5 mM EGTA, 2 mM EDTA, and 0.25 mM dithiothreitol]containing 0.75–4.5 mg of the substrate protein, 100 mMATP, and 0.12 MBq of[g-32P]ATP at 30�C for 10 and 20 minutes, unless otherwise stated. Reactionswere stopped by adding the same volume of 2� Laemmli sample buffer [125 mMTris-HCl (pH 6.8), 4% SDS, 20% glycerol, 0.002% Bromophenol blue, and 10%2-mecaptoethanol] and samples were resolved by SDS-PAGE on a 5–20%gradient gel (ATTO Corporation) after heat denaturation. Phosphorylation of thesubstrate was visualized by autoradiography.

Statistical Analysis. Values are expressed as the mean 6 standard error.All data were analyzed using a one-way analysis of variance (ANOVA).The significance of differences between groups was analyzed using an

ANOVA or unpaired t test. The level set for significant differences wasP , 0.05.

Results

Phosphomimetic Mutation of hPXR at T408 Reduced theInduction of CYP3A4 mRNA by Rifampicin. We previouslyreported that the phosphomimetic mutation of hPXR at T408 attenuatedthe induction of UGT1A1 mRNA expression by roscovitine andrifampicin in HepG2 cells (Sugatani et al., 2012). The present studyalso revealed that the phosphomimetic T408D mutation attenuatedUGT1A1 reporter activity by rifampicin, whereas the phosphodeficientT408A mutation exhibited similar activity to that of the WT (Fig. 1A).Furthermore, the T408D mutation reduced the induction of CYP3A4mRNA by rifampicin not only in HepG2 but also in SW480 cells (Fig.1, B and C).We then determined whether the phosphomimetic mutationof hPXR at T408 altered the expression of hPXR in HepG2 cells in vitro

Fig. 3. Immunoprecipitation and Western blot analyses of the nucleocytoplasmic distribution of the hPXR WT and T408D mutant and their interactions with chaperone andcochaperone proteins in proteasome inhibitor–treated HepG2 cells. HepG2 cells cultured for 24 hours were transfected with expression vectors encoding the YFP-hPXR WTor T408D mutant. The medium was replaced 24 hours after transfection, and cells were retransfected and then cultured with rifampicin (5 � 1026 M) or vehicle (DMSO) inthe presence or absence of calpain inhibitor I (1 � 1025 M), MG132 (1 � 1026 M), or vehicle (DMSO, control) for an additional 24 hours. Cytoplasmic, 530 mg (A) andnuclear, 200 mg (B) proteins prepared from these cells and a blank sample (vehicle) were precipitated with anti-GFP antibody. The resulting bound proteins were subjected toa Western blot analysis with the indicated antibodies. Cytoplasmic and nuclear YFP-tagged hPXR and chaperone and cochaperone proteins were determined by Westernblotting using the indicated antibodies: anti-a-tubulin (loading control) and anti-histone H1 (loading control) antibodies. Experiments were performed three times withsimilar results, and representative data are shown. WT, wild-type hPXR; T408D, T408D-mutated hPXR.

Stability Control of PXR via Chaperone-Autophagy Pathway 141

at ASPE

T Journals on M

ay 28, 2020dm

d.aspetjournals.orgD

ownloaded from

and in mouse livers in vivo or reduced the stability of hPXR, resulting inimpaired transcriptional activity. No significant change was observed inthe protein levels of the hPXRWT or T408Dmutant expressed 24 hoursafter transfection with expression plasmids for hPXR, and 39 hoursafter transfection the level of the T408D mutant protein was reduced inthe absence of MG132 but increased to a level similar to that of the WTprotein in the presence of MG132 (Fig. 1D). We also investigated theexpression of the hPXRWT and T408Dmutant proteins in mouse liversafter the in situ WT- or T408D mutant-expressing plasmid DNA

injection. Whereas no significant difference was observed in the levelsof the hPXRWT and T408D mutant proteins expressed in mouse livers12 hours after the in situ DNA injection, the levels of the T408Dmutantprotein were significantly lower than those of the WT protein 24 hoursafter the in situ DNA injection (Fig. 1E). These results suggested thatthe impaired function of the phosphomimetic mutant was not caused byits reduced expression, and also that phosphorylation at T408 reducedthe protein stability of hPXR, thereby impairing its function.Effects of Proteasome Inhibitors on the Nucleocytoplasmic

Distribution of the hPXR WT and T408D Mutant and TheirInteractions with Chaperone and Cochaperone Proteins. SincehPXR stability was previously shown to be regulated viaphosphorylation-facilitated ubiquitination through dual-specificitytyrosine-(Y)-phosphorylation–regulated kinase 2 (DYRK2) and

Fig. 4. Effects of proteasome inhibitors on the nucleocytoplasmic distribution of thehPXR WT and T408D mutant (A, B) and their interactions with chaperone andcochaperone proteins (C–F). The band intensities shown in Fig. 3 were measuredwith ImageJ software. (A, B) Each quantified value of the YFP-hPXR bands wascorrected for that of a-tubulin (A) or histone H1 (B), and relative levels areexpressed by taking the values obtained from cells expressing the YFP-hPXR WTand treated without rifampicin as one. (C–F) Cytosolic YFP-PXR was immunopre-cipitated using an anti-GFP antibody and other proteins bound to PXR were detectedby Western blotting. Each quantified value of the cytosolic YFP-hPXR, Hsp90,Hsp70, Hsc70, and CHIP bands was corrected for that of IgG heavy chain (HC).Relative levels of the Hsp90, Hsp70, Hsc70, and CHIP to the hPXR were expressedby taking the values obtained from cells expressing the YFP-hPXR WT and treatedwithout rifampicin as one. Data presented are the average 6 S.E. of threeexperiments. a, P , 0.05 versus the hPXR WT in vehicle-treated control cells; b,P , 0.05; bb, , 0.01 versus the hPXR WT in rifampicin-stimulated control cells;cc, P , 0.01 versus the hPXR T408D mutant in vehicle-treated control cells; d,P , 0.05 versus the hPXR T408D mutant in rifampicin-stimulated control cells; e,P , 0.05 versus the hPXR WT in vehicle-treated cells with calpain inhibitor I;f, P , 0.05 versus the hPXR WT in rifampicin-stimulated cells with calpaininhibitor I; g, P, 0.05 versus the hPXR WT in vehicle-treated cells with MG132; h,P , 0.05 versus the hPXR WT in rifampicin-stimulated cells with MG132. WT,wild-type hPXR; T408D, T408D-mutated hPXR.

Fig. 5. Effects of geldanamycin (A–C) and pifithrin-m (D, E) on the transactivationcapacity of the hPXR WT and T408D mutant (A, D), expression of chaperone andcochaperone mRNAs in cells expressing the hPXR WT (B, C), and nucleocyto-plasmic distribution of the hPXRWT and T408D mutant (E) in HepG2 cells. HepG2cells cultured for 24 hours were transfected with expression plasmids for hPXR. Themedium was replaced 24 hours after transfection, and cells were retransfected andthen cultured with rifampicin (5 � 1026 M) or vehicle (DMSO) in the presence ofpifithrin-m (1 � 1025 M), geldanamycin (1 � 1026 M), or vehicle (DMSO, control)for an additional 24 hours. (A–D) Messenger RNA levels in control cells treatedwithout rifampicin were taken as one. Data presented are the average 6 S.E. of threeexperiments. *P , 0.05; **P , 0.01; ***P , 0.001 versus the hPXR WT invehicle-treated control cells; ##P , 0.01; ###P , 0.001 versus the hPXR WT inrifampicin-stimulated control cells; +++p,0.001 versus the hPXR T408D mutantin rifampicin-stimulated control cells. (E) Cytoplasmic and nuclear YFP-taggedhPXR from cells treated with pifithrin-m (1 � 1025 M) or vehicle were determinedby Western blotting using anti-GFP, anti-a-tubulin (loading control), and anti-histone H1 (loading control) antibodies. Experiments were performed twice withsimilar results, and representative data are shown. WT, wild-type hPXR; T408D,T408D-mutated hPXR.

142 Sugatani et al.

at ASPE

T Journals on M

ay 28, 2020dm

d.aspetjournals.orgD

ownloaded from

ubiquitin protein ligase E3 component n-recognition 5 (UBR5) (Onget al., 2014) as well as via ubiquitination by the Ring-B-box-coiled-coilprotein interacting with PKC-1 (RBCK1) and its proteasomal degra-dation (Rana et al., 2013), we herein examined the effects of proteasomeinhibitors on the stability of the T408D mutant. Calpain inhibitor I at10 mM and lactacystin at 5 mM, which did not significantly suppresscell growth, enhanced the basic levels of CYP3A4 mRNA and also theCYP3A4 mRNA levels induced by rifampicin, respectively, in HepG2cells exogenously expressing the hPXR WT (Supplemental Fig. 1 andFig. 2). MG132 at 1 mM led to the accumulation of the hPXRWT, andT408D mutant proteins in whole cell lysates appeared to slightlyincrease the basic levels of CYP3A4 mRNA of the hPXR WT, andincreased those of the T408D mutant, whereas it significantly sup-pressed cell growth (Supplemental Figs. 1 and 2). On the other hand,even in the presence of calpain inhibitor I at 10 mM and MG132 at1mM, the protein levels of the T408Dmutant were lower than or similarto those of the WT in the cytoplasm and were significantly lower in thenucleus (Figs. 2–4). Immunoprecipitation and Western blot analysesrevealed that the complexes of the hPXR T408D mutant with Hsp90,Hsc70/Hsp70, CHIP, HOP, and p23 formed as well as that with theWTin the cytoplasm, whereas interactions between hPXR and Hsp90,CHIP, and p23 were not observed in the nucleus (Figs. 3 and 4).These results suggested that the T408D mutant was degraded througha proteasome-independent degradation pathway in addition to aubiquitin/proteasome pathway.Effects of Hsp90 and Hsp70 Inhibitors on Transcriptional

Activities and Nucleocytoplasmic Distribution of the T408D Mutant,and Its Complex Formation with Chaperone and CochaperoneProteins in HepG2 Cells. We investigated the mechanisms by which theinhibition of Hsp90 and Hsp70 influenced the transcriptional activity andcellular distribution of hPXR using the Hsp90 inhibitor geldanamycin andHsp70 inhibitor pifithrin-m. Geldanamycin at 1 mM increased the basiclevels of CYP3A4 mRNA in nonstimulated cells, and pifithrin-m at10 mM reduced CYP3A4 mRNA levels in the hPXR WT and T408Dmutant-expressing rifampicin-stimulated cells, whereas cell growth wassuppressed in cells treated with geldanamycin at 1 mM but not in thosetreated with pifithrin-m at 10 mM (Fig. 5 and Supplemental Fig. 1).

Geldanamycin induced the expression of Hsp90 and Hsp70 mRNAs, didnot alter Hsc70 or HOP mRNA levels, and reduced CHIP mRNA levels(Fig. 5 and Supplemental Fig. 2). Furthermore, it markedly increasedHsp70, hPXR WT, and T408D mutant protein levels in the cytoplasm,which then appeared to increase WT and T408D mutant protein levels inthe nucleus, as well as CYP3A4 mRNA levels (Figs. 5 and 6). Theimmunoprecipitation assay revealed that geldanamycin increased complexformation by the hPXRWT and T408Dmutant with Hsp70, Hsc70, HOP,and CHIP, but not p23 in the cytoplasm (Fig. 6). The binding of the hPXRWT and T408D mutant with Hsp90, CHIP, and p23 was not observed inthe nucleus of geldanamycin-treated cells, the same as in the control andproteasome inhibitor-treated cells (Figs. 3 and 6). In contrast, pifithrin-m at10 mM accumulated the hPXR WT and T408D mutant proteins in thecytoplasm but not in the nucleus (Fig. 5). These results suggested thatHsp70 contributed to the folding of client proteins such as hPXR in thecytoplasm.CHIP Deletion Attenuated hPXR Transcriptional Activity in

HepG2 Cells Treated with and without Lactacystin. Since theknockdown of E3 ligases RBCK1 and UBR5 was previously reportedto result in the accumulation of cellular hPXR and concomitantincreases in the induction of hPXR target genes by rifampicin (Ranaet al., 2013; Ong et al., 2014), we evaluated the effects of calpaininhibitor 1 and geldanamycin on the expression of CHIP, RBCK1, andUBR5 mRNAs. The inhibition of proteasome and Hsp90 did notmarkedly affect the expression of CHIP, RBCK1, or UBR5 mRNAs;the calpain inhibitor at 10 mM slightly increased CHIP and UBR5mRNA levels but did not alter those of RBCK1 mRNA (SupplementalFig. 2). Furthermore, geldanamycin at 1 mM reduced CHIP andRBCK1 mRNA levels but did not alter those of UBR5 mRNA(Supplemental Fig. 2). To examine the effects of CHIP knockdown,anti-CHIP siRNA was introduced into HepG2 cells and the hPXR WTand T408D mutant were then exogenously expressed in these cells.Transfection with anti-CHIP siRNA reduced CHIP mRNA levels to0.29–0.33 fold of the control and CHIP protein levels in whole cells to0.30–0.34 fold of the control, and also significantly attenuated thelevels of CYP3A4 mRNA induced by rifampicin, demonstrating thatCHIP may contribute to the transactivation of hPXR (Fig. 7). The

Fig. 6. Effects of geldanamycin on the nucleocytoplasmicdistribution of the hPXR WT and T408D mutant and theirinteractions with chaperone and cochaperone proteins. HepG2cells cultured for 24 hours were transfected with expressionvectors encoding the YFP-hPXR WT and T408D mutant. Themedium was replaced 24 hours after transfection, and cells wereretransfected and then cultured with rifampicin (5 � 1026 M) orvehicle (DMSO) in the presence or absence of geldanamycin(1 � 1026 M) for an additional 24 hours. Cytoplasmic, 500 mg(A) and nuclear, 200 mg (B) proteins prepared from these cellsand a blank sample (vehicle) were precipitated with an anti-GFPantibody. The resulting bound proteins were subjected to aWestern blot analysis with the indicated antibodies. Cytoplas-mic and nuclear YFP-tagged hPXR and chaperone andcochaperone proteins were determined by Western blottingusing the indicated antibodies: anti-a-tubulin (loading control)and anti-histone H1 (loading control) antibodies. Experimentswere performed twice with similar results, and representativedata are shown. IP, Immunoprecipitate analysis; WB, Westernblot analysis; WT, wild-type hPXR; T408D, T408D-mutatedhPXR.

Stability Control of PXR via Chaperone-Autophagy Pathway 143

at ASPE

T Journals on M

ay 28, 2020dm

d.aspetjournals.orgD

ownloaded from

knockdown of CHIP led to the significant accumulation of the hPXRWT in cells treated without rifampicin in the presence of lactacystin(Fig. 7B). These results revealed that CHIP may contribute to thedegradation of hPXR as well as its transactivation.The Lysosome Inhibitor Chloroquine Increased the Stability of

the hPXR T408D Mutant in the Cytoplasm. In an attempt to explorethe hypothesis that the hPXR T408D mutant is degraded in lysosomesthat fuse with an autophagic vacuole, we investigated the effects ofchloroquine on the transcriptional activity and nucleocytoplasmicdistribution of the T408D mutant. Chloroquine at 20 mM, which didnot affect cell growth, increased the induction of CYP3A4 mRNA byrifampicin and a concomitant increase was observed in hPXR WTprotein levels in the nucleus (Supplemental Fig. 1 and Fig. 8). In

contrast, the T408D mutant protein accumulated more efficiently thanthe WT in the cytoplasm but not in the nucleus in the presence ofchloroquine, resulting in reduced transcriptional activities (Fig. 8).These results indicated that the T408D mutant was degraded in thelysosome.The Complex of hPXR with Chaperone and Cochaperone Proteins

Was Ubiquitinated in Cells Treated with or without ProteasomeInhibitors. To determine whether hPXR was directly ubiquitinated byE3 ubiquitin ligase CHIP, we assayed the in vitro ubiquitination of thehPXR WT by CHIP with the E2 ubiquitin-conjugating enzyme UbcH5cand compared with Hsc70, because Hsc70 is a major target of CHIP (Jianget al., 2001). As shown in Fig. 9A, CHIP directly ubiquitinated Hsc70 butnot the hPXRWT. In contrast, when immunoprecipitated proteins with the

Fig. 7. Effects of anti-CHIP siRNA transfection on the expression of CYP3A4 mRNA and hPXR, CHIP, and a-tubulin proteins. HepG2 cells cultured for 24 hours weretransfected with anti-CHIP siRNA (1 � 1027 M) or mock-transfected. Cells were retransfected with expression vectors encoding the Flag-tagged hPXR WT or T408Dmutant 12 hours after the first transfection. Twelve hours after the second transfection, cells were cultured with rifampicin (5 � 1026 M) or vehicle (DMSO) in the presenceor absence of lactacystin (5 � 1026 M) for an additional 24 hours. In cells expressing the hPXR WT and T408D mutant (A) and those expressing the hPXR WT and treatedwith or without lactacystin (B), CHIP and CYP3A4 mRNA levels in mock-transfected and vehicle-treated control cells without lactacystin were taken as one. Data presentedare the average 6 S.E. of three experiments. The resulting proteins were determined by Western blotting using anti-Flag, anti-CHIP, and anti-a-tubulin (loading control)antibodies. Band intensities were measured with ImageJ software, and each quantified value of the Flag-hPXR and CHIP bands was corrected for that of a-tubulin. Relativelevels are expressed by taking the values obtained from the mock-treated and vehicle-treated cells expressing the Flag-hPXR WT without lactacystin as one. Westernblottings were performed three times with similar results, and representative data are shown. Data presented are the average 6 S.E. of three experiments. *P,0.05; ***P,0.001 versus the hPXR WT in mock-transfected and vehicle-treated cells without lactacystin; ##P,0.01: ###P , 0.001 versus the hPXR WT in mock-transfected andrifampicin-stimulated cells without lactacystin; ^^^P , 0.001 versus the hPXR T408D mutant in mock-transfected and vehicle-treated cells without lactacystin (A) or thehPXR WT in mock-transfected and lactacystin- and vehicle-treated cells (B); +++P , 0.001 versus the hPXR T408D mutant in mock-transfected and rifampicin-stimulatedcells without lactacystin (A) or mock-transfected and lactacystin- and rifampicin-treated cells (B). WT, wild-type hPXR; T408D, T408D-mutated hPXR.

144 Sugatani et al.

at ASPE

T Journals on M

ay 28, 2020dm

d.aspetjournals.orgD

ownloaded from

anti-GFP antibody for the cytoplasm of vehicle- and calpain inhibitor I-treated cells were immunoblotted for ubiquitin, a protein band near 73 kDa[band marked with two asterisks (**) in Fig. 9B] in the hPXR/chaperonecomplex and smeared bands were detected; however, the complex proteins

weremarkedly smeared and polyubiquitinated for the cytoplasmofMG132-treated cells (Fig. 9B).Autophagic Cargo Contained hPXR. Wu et al. (2010) reported

that although 3-MA promoted an autophagic flux (resulting in an

Fig. 8. Effects of chloroquine on transactivation capacities of the hPXR WT and T408D mutant and their nucleocytoplasmic distribution in HepG2 cells. HepG2 cells cultured for24 hours were transfected with expression vectors encoding the Flag-tagged hPXR WT or T408D mutant. The medium was replaced 24 hours after transfection, and cells wereretransfected and then cultured with rifampicin (5 � 1026 M) or vehicle (DMSO) in the presence or absence of chloroquine (2 � 1025 M) for an additional 24 hours. (A) CYP3A4mRNA levels in cells treated with rifampicin or vehicle in the presence or absence of chloroquine (2� 1025 M) were expressed by taking the control value obtained from hPXRWT-expressing and vehicle-treated cells without chloroquine as one. Data presented are the average6 S.E. of three experiments. **P, 0.01 versus the hPXRWT in vehicle-treated cellswithout chloroquine; ###P , 0.001 versus the hPXR WT in rifampicin-stimulated cells without chloroquine; +++P , 0.001 versus the hPXR T408D mutant in chloroquine-treatedand rifampicin-stimulated cells. Cytoplasmic (B) and nuclear (C) proteins were determined by Western blotting using anti-Flag, anti-a-tubulin (loading control), and anti-histone H1(loading control) antibodies. Experiments were performed twice with similar results, and representative data are shown. WT, wild-type hPXR; T408D, T408D-mutated hPXR.

Stability Control of PXR via Chaperone-Autophagy Pathway 145

at ASPE

T Journals on M

ay 28, 2020dm

d.aspetjournals.orgD

ownloaded from

increase in autophagic markers) when treated under nutrient-richconditions for a prolonged period of time, it was still capable ofsuppressing starvation-induced autophagy. Since inhibition of theproteasome has been shown to induce macroautophagy (Ding et al.,2007; Tang et al., 2014) and 3-MA blocks the maturation of theautophagophore into a closed autophagosome in the macroautophagicprocess (Abounit et al., 2012), we evaluated the effects of lactacystinand 3-MA on the cellular stability of hPXR and intracellular traffickingof the hPXR T408D mutant to the autophagic cargo to further exploreour hypothesis that the hPXR T408D mutant is degraded throughautophagy. A Western blot analysis revealed that the treatment ofthe hPXR WT- or T408D mutant-expressing cells with lactacystin or3-MA alone accumulated not only cellular hPXR WT but also theT408D mutant (Fig. 10). The hPXR T408D mutant accumulated moreefficiently than the WT in cells treated with lactacystin in the presenceof 3-MA. The autophagic markers LC3 and p62 markedly increased incells treated with 3-MA and/or lactacystin in full medium. To determine

whether the autophagic cargo contained hPXR, we examined thecolocalization of the hPXR WT and T408D mutant with LC3 using ananti-LC3 antibody to identify the autophagic structure. Many LC3puncta contained the hPXR WT and the T408D mutant (Fig. 11).Similar results were obtained for the ubiquitinated substrate adaptor p62in cells treated with lactacystin and/or 3-MA (Fig. 11 and SupplementalFig. 3). These results suggested that LC3- and p62-positive autophagiccargo functioned to deliver hPXR to the lysosome.hPXR Was Phosphorylated at Threonine-408 by PKC In Vitro.

In our previous study (Sugatani et al., 2014), we demonstrated that themRNA levels of UGT1A1 induced by rifampicin or roscovitine wereslightly higher in HepG2 cells simultaneously treated with Go6983(PKC inhibitor) but were lower in HepG2 cells simultaneously treatedwith Rp-8-Br-cAMPS (PKA inhibitor). Since hPXR contained theminimal recognition motif S/T-X-R at amino acids 408–410, whichmay potentially be phosphorylated by PKC (Kang et al., 2012), andDYRK2 phosphorylates hPXR to facilitate its subsequent ubiquitina-tion by UBR5 (Ong et al., 2014), we performed assays with PKC orDYRK2 as protein serine/threonine kinase, [g-32P]ATP and engineeredMBP fusion proteins containing hPXR. In vitro kinase assays revealedthat hPXR was phosphorylated by PKC and DYRK2, as assessed byradiography (Fig. 12A). To identify the phosphorylated site(s), weinvestigated whether the hPXR T408A mutant protein was phosphor-ylated by these kinases and found that mutating the threonine-408 ofhPXR to alanine led to significantly lower phosphorylation levels byPKC than by the hPXR WT (0.73 6 0.08 of the WT; P , 0.05 versusthe WT), whereas DYRK2 phosphorylated the hPXR T408A mutantmore than the hPXR WT (1.266 0.10 of the WT; P, 0.05 versus theWT) (Fig. 12). Taken together, these results indicated that thethreonine-408 of hPXR in the ligand-binding domain was one ofthe PKC-specific phosphorylation sites.

Discussion

The glucocorticoid receptor (GR), which is a ligand-dependentnuclear receptor, is known to be phosphorylated through cell- andtissue-specific signaling systems, leading to conformational changes,which in turn modulate its transcriptional responses and stability(reviewed in Galliher-Beckley and Cidlowski, 2009). Previous studiesdemonstrated that phosphomimetic mutations at the threonine-57,threonine-290, serine-350, and threonine-408 of hPXR suppressedhPXR activities (Lichti-Kaiser et al., 2009; Pondugula et al., 2009;Sugatani et al., 2012, 2014; Wang et al., 2015). Therefore, hPXRfunctions are also considered to be modulated through the phosphor-ylation of hPXR at specific amino acid residues as well as nuclearreceptors such as constitutive active/androstane receptor (CAR) and GR(Galliher-Beckley and Cidlowski, 2009; Mutoh et al., 2013). We alsopreviously reported that the phosphorylation of hPXR at threonine-408may be associated with its degradation through proteasomal machinerybecause the hPXR T408D mutant markedly accumulated in thecytoplasm of cells treated with MG-132 at an extremely high 5 mMconcentration (Sugatani et al., 2012). However, the results of thepresent study suggest the presence of a proteasome-independentdegradation pathway in addition to the proteasome-dependent degra-dation pathway because the protein level of the T408D mutant was lessthan or similar to that of the WT in the cytoplasm and significantlylower than the WT in the nucleus, even in the presence of proteasomeinhibitors (Figs. 1–4). Thus, we herein examined the molecularmechanisms regulating hPXR functions and stability, especially thoseunderlying autophagy for its degradation.The chaperone proteins Hsc70/Hsp70 assist in the correct protein

folding of nascent or denatured client proteins for intracellular functions

Fig. 9. Ubiquitination of hPXR by an in vitro assay (A) and the chaperone/hPXRcomplex in HepG2 cells (B). (A) Recombinant hPXR, recombinant hPXR ligandbinding domain, or recombinant Hsc70 was incubated with UbcH5c and CHIP at37�C for ;30–60 minutes using an in vitro ubiquitin kit (Abcam), as described inMaterials and Methods. Ubiquitinated proteins were determined by Western blottingusing an anti-ubiquitin antibody. (B) HepG2 cells cultured for 24 hours weretransfected with expression vectors encoding the YFP-hPXR WT or T408D mutant.The medium was replaced 24 hours after transfection, and cells were retransfectedand then cultured with rifampicin (5 � 1026 M) or vehicle (DMSO) in the presenceor absence of calpain inhibitor I (1 � 1025 M) or MG132 (1 � 1026 M) for anadditional 24 hours. Cytoplasmic proteins (530 mg) prepared from these cells and ablank sample (vehicle) were precipitated with an anti-GFP antibody. The resultingbound proteins were subjected to a Western blot analysis with an anti-ubiquitin oranti-PXR antibody. Experiments were performed three times with similar results,and representative data are shown. WT, wild-type hPXR; T408D, T408D-mutatedhPXR. *Nonspecific bands; **see text.

146 Sugatani et al.

at ASPE

T Journals on M

ay 28, 2020dm

d.aspetjournals.orgD

ownloaded from

and, in contrast, work with a ubiquitin pathway for the proteolyticdegradation of unstable or misfolded client proteins (Hayer-Hartl et al.,1996; Wickner et al., 1999; Meacham et al., 2001; Kampinga et al.,2003). CHIP interacts with Hsc70/Hsp70 and Hsp90, which results inthe refolding of chaperone complexes with client proteins, and is alsoinvolved in the direct ubiquitination of client proteins such as GR as anE3 ligase for delivery to the proteasome (reviewed in McDonough andPatterson, 2003). HOP has been shown to assist in protein transfer in theHsp70- and Hsp90-dependent protein-folding pathways (Yamamotoet al., 2014). Hsp90, which regulates transcription by maintainingtranscription factors in the cytoplasm, is known to interact with p23 andCHIP through different sites, and CHIP interacts with the carboxyl-terminal domain of Hsc70 (reviewed in McDonough and Patterson,2003). In addition to GR and CAR, hPXR has been identified as anHsp90 client substrate (Sanchez et al., 1985; Honkakoski et al., 1998;Squires et al., 2004). The present study revealed that hPXR wasinvolved in the complex of chaperones Hsp90 and Hsc70/Hsp70 withsuch cochaperones as CHIP, HOP, and p23 in the cytoplasm (Figs. 3, 4,and 6). Proteasome inhibitors (calpain inhibitor I and MG132) andthe Hsp90 inhibitor geldanamycin induced Hsp70, which may haveresulted in the more effective formation of the chaperone/hPXRcomplex (Figs. 3, 4, and 6). This study revealed that the Hsp70inhibitor pifithrin-m and CHIP knockdown reduced the transcriptionalactivity of hPXR (Figs. 5 and 7). CHIP has been shown to enhance thefolding capacity of Hsp70 in mammalian cells (Kampinga et al., 2003);therefore, the refolding of hPXR by CHIP/Hsp70 may enhance thetranslocation of hPXR from the cytoplasm to the nucleus, resulting in itselevated transactivation. In contrast, since the interaction of the hPXR

T408D mutant with Hsc70/Hsp70 and CHIP was similar to that of theWT in the cytoplasm of cells treated with and without the proteasomeinhibitors, and because the hPXR T408D mutant did not easilytranslocate to the nucleus and did not accumulate in the cytoplasm, itmay have been misfolded by the CHIP/Hsp70 quality check and thendegraded in the cytoplasm through autophagy (Figs. 3– 6, 7, 10, and11). Furthermore, the hPXR WT accumulated in CHIP-deleted controlcells in the presence of lactacystin (Fig. 7), suggesting that CHIP mayplay a key role in the proteasome-independent degradation pathway aswell as chaperone-dependent quality control (Figs. 3, 6, and 7).RBCK1 and UBR5 ubiquitinate target proteins for proteasomal

degradation in innumerable physiologic reactions; for example,RBCK1 negatively regulates tumor necrosis factor a– and interleukin-1–triggered nuclear factor kB activation by targeting transforminggrowth factor b–activated kinase 1 binding proteins 2/3 for degradation(Tian et al., 2007), and UBR5 is involved in regulating the stability ofthe gluconeogenic rate-limiting enzyme phosphoenolpyruvate car-boxykinase (Jiang et al., 2011). RBCK1 and UBR5 were previouslyshown to be responsible for the ubiquitination of hPXR to regulatehPXR degradation through proteasomal machinery (Rana et al., 2013;Ong et al., 2014), whereas the ubiquitin-proteasomal regulation ofcytoplasmic CAR retention protein and Hsp70, but not CAR itself,may be an important contributor to the regulation of cytoplasmicCAR levels (Timsit and Negishi, 2014). Unlike RBCK1 and UBR5,CHIP functions in chaperone-dependent ubiquitination for substratedelivery to the proteasome (reviewed in McDonough and Patterson,2003). Although CHIP ubiquitinated Hsc70 and Hsp70 (Jiang et al.,2001; Soss et al., 2015), but not hPXR, in vitro (Fig. 9), it currently

Fig. 10. Autophagy inhibitor 3-methyladenine (3-MA) led tothe greater accumulation of the hPXR T408D mutant than theWT in the presence of proteasome inhibitor lactacystin. HepG2cells cultured for 24 hours were transfected with expressionvectors encoding the Flag-tagged hPXR WT or T408D mutant.Cells were incubated with 5 mM 3-MA or vehicle 12 hours aftertransfection. Twelve hours later, cells were cultured withlactacystin (5 � 1026 M) or vehicle for an additional 24 hours.(A) Flag-hPXR, LC3, and p62 proteins in whole cell lysateswere determined by Western blotting with the indicatedantibodies and anti-a-tubulin antibody (loading control). (B)Band intensities were measured with ImageJ software, and eachquantified value of the Flag-hPXR bands was corrected for thatof a-tubulin. Relative levels were expressed by taking thevalues obtained from vehicle-treated cells expressing the Flag-hPXR WT as one. Western blottings were performed threetimes with similar results, and representative data are shown.Data presented are the average 6 S.E. of three experiments.*P , 0.05; **P , 0.01; ***P , 0.001 versus the hPXR WT invehicle-treated cells; ##P, 0.01; ###P, 0.001 versus the hPXRT408D mutant in vehicle-treated cells; +P , 0.05 versus thehPXR WT in lactacystin- and 3-MA-treated cells. WT, wild-type hPXR; T408D, T408D-mutated hPXR.

Stability Control of PXR via Chaperone-Autophagy Pathway 147

at ASPE

T Journals on M

ay 28, 2020dm

d.aspetjournals.orgD

ownloaded from

remains unclear whether CHIP participates in the delivery of theubiquitinated chaperone complex to the proteasome or autophagiccargo containing p62 or if ubiquitination of the chaperone complex byCHIP induces substrate recognition and recruitment to the proteasomeor autophagy. Since Hsc70 ubiquitinated by CHIP is stable (Jiang et al.,2001; Soss et al., 2015), it is conceivable that the ubiquitinated proteinmay have been detected in the chaperone complex with hPXR (Fig. 9).

Immunocytochemistry revealed that YFP-fused hPXR and p62, whichacts as a cargo receptor of ubiquitinated substrates, accumulated andcolocalized in cells treated with 3-MA and lactacystin (Figs. 10 and 11).Thus, p62 may play a role in the transport of the ubiquitinatedchaperone complex as an acceptor of phagophores in the autophagy-lysosome pathway. Autophagy is a process by which cellular compo-nents are delivered to the lysosome for degradation and involves threemain pathways: macroautophagy, chaperone-mediated autophagy, andmicroautophagy (reviewed in Feng et al., 2014). Inhibitions of theproteasome by its inhibitors and Hsp90 by geldanamycin have beenshown to induce macroautophagy and chaperone-mediated autophagy,respectively, and ultimately client degradation in the lysosome (Finnet al., 2005; Qing et al., 2006; Ge et al., 2009). We herein demonstratedthat inhibition of the lysosome by chloroquine resulted in the accumu-lation of the hPXR WT and T408D mutant in the cytoplasm (Fig. 8),whereas proteasome inhibition was linked to the enhanced degradation ofthe T408D mutant because 3-MA suppressed the degradation of hPXR,particularly that of the T408Dmutant, in the presence of lactacystin (Figs.3 and 10). The inhibition of Hsp90 led to the accumulation of the hPXRWT and T408D mutant in the cytoplasm and nucleus (Fig. 6), indicatingthat the balance of the hPXR quality check by Hsp90 and Hsp70/Hsc70affected its stability, and also that the T408D mutant was targeted fordegradation in a manner that was largely mediated by macroautophagybut not chaperone-mediated autophagy.PKC cell signaling is known to repress the transcriptional activity of

PXR by increasing the strength of the interaction between PXR and thenuclear receptor corepressor protein NCoR (Ding and Staudinger,2005). We herein demonstrated that PKC directly phosphorylated thethreonine-408 of hPXR in vitro (Fig. 12). Although the cellularsignaling pathways by which PKC is activated to phosphorylate hPXRin vivo remain unclear, the T408D mutant was degraded faster than theWT in the cytoplasm (Figs. 1, 3, and 6). Thus, PKCmay play a key rolein regulating the cellular stability of hPXR through phosphorylation atthreonine-408. The T408D mutant reduced its stability not only incultured cancer cells but also in normal cells in mouse livers (Fig. 1E).Therefore, emerging evidence prompted us to propose a schematicmodel of the regulation of hPXR transactivation and its degradationthrough the CHIP–Hsc70/Hsp70 protein quality check system (Fig.

Fig. 12. Threonine-408 of hPXR was phosphorylated by PKC in vitro. (A) MBP-hPXR WT and T408A mutant fusion proteins (0.25–1.5 mg) were incubated withPKC (0.63–5.0 ng) and DYRK2 (0.75 to 6.0 ng) in the presence of [32P]-labeledATP at 30�C for 10 minutes, then electrophoresed, and exposed to an autoradiographicfilm. (B) The MBP-hPXR WT and T408A mutant fusion proteins (20 mg each) werestained with Coomassie Brilliant Blue. Band intensities were measured with ImageJsoftware. Relative levels were expressed by taking the value obtained from the MBP-hPXRWT fusion protein as one. Experiments were performed three times with similarresults, and representative data are shown. WT, wild-type hPXR; T408A, T408A-mutated hPXR.

Fig. 11. The hPXR T408D mutant was colocalized with LC3 and p62 in HepG2cells treated with 3-MA and/or lactacystin. HepG2 cells cultured for 24 hours weretransfected with expression vectors encoding the YFP-tagged hPXR WT or T408Dmutant. Cells were incubated with 5 mM 3-MA or vehicle 12 hours aftertransfection. Twelve hours later, cells were cultured with lactacystin (5 � 1026 M)or vehicle for an additional 24 hours. The subcellular localization of YFP-hPXR wasexamined by confocal microscopy. Blue, DAPI-labeled nuclei; red, Alexa Fluor555-labeled LC3 or p62; green, YFP-labeled hPXR. Merged images are shown inthe rightmost column and the yellow-stained puncta indicated LC3 or p62,colocalized with hPXR. Experiments were performed twice with similar results, andrepresentative data are shown. Scale bar, 50 mm. WT, wild-type hPXR; T408D,T408D-mutated hPXR.

Fig. 13. Schematic model for the regulation of hPXR stability through thephosphorylation of hPXR at threonine-408 and the CHIP/chaperone–autophagypathway. After the treatment with an activator, the CHIP/Hsp70/Hsc70 complexmay check and enable the correct folding of the hPXR WT but not the T408Dmutant for transfer into the nucleus through nuclear pores. The chaperone–hPXRT408D mutant complex ubiquitinated by CHIP binds to p62 in the autophagophorefor delivery to the lysosome for its degradation.

148 Sugatani et al.

at ASPE

T Journals on M

ay 28, 2020dm

d.aspetjournals.orgD

ownloaded from

13); we suggest that 1) hPXR receives a refolded response to a Hsp90/Hsp70–based protein quality check and is delivered through nuclearpores to the nucleus, 2) when hPXR is phosphorylated at T408 by PKC,it is misfolded and retained in the cytoplasm, and 3) p62, which isinvolved in the autophagophore, recruits the hPXR/chaperone complexand binds it to deliver hPXR to the lysosome. Hsp90 inhibitorganetespib is expected to have therapeutic effects in the treatment ofadenocarcinoma lung tumors (Gomez-Casal et al., 2015) and systemiclupus erythematosus (Liu et al., 2015), and the proteasome inhibitorbortezomib has been used as an antineoplastic drug to treat multiplemyeloma (Torimoto et al., 2015). Thus, these inhibitors may suppresshPXR degradation, resulting in the enhanced induction of various drugmetabolism–related genes by hPXR (Pondugula and Mani, 2013). Themolecular mechanisms by which the hPXR-CHIP/chaperone complexenables the refolding of hPXR for its transfer to the nucleus and thephosphorylation of hPXR at threonine-408 for its delivery to theautophagophore for degradation in vivo currently remain unclear butmay provide a novel insight into the regulation of hPXR functions andits degradation.

Authorship ContributionsParticipated in research design: Sugatani.Conducted experiments: Sugatani, Noguchi, Hattori, Yamaguchi, Yamazaki.Contributed new reagents or analytic tools: Sugatani, Noguchi, Hattori.Performed data analysis: Sugatani, Noguchi, Hattori, Yamaguchi, Yama-

zaki, Ikari.Wrote or contributed to the writing of the manuscript: Sugatani.

References

Abounit K, Scarabelli TM, and McCauley RB (2012) Autophagy in mammalian cells. World JBiol Chem 3:1–6.

Cui W, Sun M, Galeva N, Williams TD, Azuma Y, and Staudinger JL (2015) SUMOylation andubiquitylation circuitry controls pregnane X receptor biology in hepatocytes. Drug MetabDispos 43:1316–1325.

Ding X and Staudinger JL (2005) Repression of PXR-mediated induction of hepatic CYP3A geneexpression by protein kinase C. Biochem Pharmacol 69:867–873.

Ding WX, Ni HM, Gao W, Yoshimori T, Stolz DB, Ron D, and Yin XM (2007) Linking ofautophagy to ubiquitin-proteasome system is important for the regulation of endoplasmic re-ticulum stress and cell viability. Am J Pathol 171:513–524.

Elias A, High AA, Mishra A, Ong SS, Wu J, Peng J, and Chen T (2014) Identification andcharacterization of phosphorylation sites within the pregnane X receptor protein. BiochemPharmacol 87:360–370.

Feng Y, He D, Yao Z, and Klionsky DJ (2014) The machinery of macroautophagy. Cell Res 24:24–41.

Finn PF, Mesires NT, Vine M, and Dice JF (2005) Effects of small molecules on chaperone-mediated autophagy. Autophagy 1:141–145.

Galliher-Beckley AJ and Cidlowski JA (2009) Emerging roles of glucocorticoid receptor phos-phorylation in modulating glucocorticoid hormone action in health and disease. IUBMB Life61:979–986.

Ge PF, Zhang JZ, Wang XF, Meng FK, Li WC, Luan YX, Ling F, and Luo YN (2009) Inhibitionof autophagy induced by proteasome inhibition increases cell death in human SHG-44 gliomacells. Acta Pharmacol Sin 30:1046–1052.

Gomez-Casal R, Bhattacharya C, Epperly MW, Basse PH, Wang H, Wang X, Proia DA,Greenberger JS, Socinski MA, and Levina V (2015) The Hsp90 inhibitor ganetespib radio-sensitizes human lung adenocartinoma cells. Cancers (Basel) 7:876–907.

Hayer-Hartl MK, Weber F, and Hartl FU (1996) Mechanism of chaperonin action: GroES bindingand release can drive GroEL-mediated protein folding in the absence of ATP hydrolysis.EMBO J 15:6111–6121.

Honkakoski P, Zelko I, Sueyoshi T, and Negishi M (1998) The nuclear orphan receptor CAR-retinoid X receptor heterodimer activates the phenobarbital-responsive enhancer module of theCYP2B gene. Mol Cell Biol 18:5652–5658.

Ihunnah CA, Jiang M, and Xie W (2011) Nuclear receptor PXR, transcriptional circuits andmetabolic relevance. Biochim Biophys Acta 1812:956–963.

Jiang J, Ballinger CA, Wu Y, Dai Q, Cyr DM, Höhfeld J, and Patterson C (2001) CHIP is aU-box-dependent E3 ubiquitin ligase: identification of Hsc70 as a target for ubiquitylation.J Biol Chem 276:42938–42944.

Jiang W, Wang S, Xiao M, Lin Y, Zhou L, Lei Q, Xiong Y, Guan KL, and Zhao S (2011)Acetylation regulates gluconeogenesis by promoting PEPCK1 degradation via recruiting theUBR5 ubiquitin ligase. Mol Cell 43:33–44.

Kajiro M, Hirota R, Nakajima Y, Kawanowa K, So-ma K, Ito I, Yamaguchi Y, Ohie SH,Kobayashi Y, and Seino Y, et al. (2009) The ubiquitin ligase CHIP acts as an upstreamregulator of oncogenic pathways. Nat Cell Biol 11:312–319.

Kampinga HH, Kanon B, Salomons FA, Kabakov AE, and Patterson C (2003) Overexpression ofthe cochaperone CHIP enhances Hsp70-dependent folding activity in mammalian cells. MolCell Biol 23:4948–4958.

Kang J-H, Toita R, Kim CW, and Katayama Y (2012) Protein kinase C (PKC) isozyme-specificsubstrates and their design. Biotechnol Adv 30:1662–1672.

Kawana K, Ikuta T, Kobayashi Y, Gotoh O, Takeda K, and Kawajiri K (2003) Molecularmechanism of nuclear translocation of an orphan nuclear receptor, SXR. Mol Pharmacol 63:524–531.

Konno Y, Negishi M, and Kodama S (2008) The roles of nuclear receptors CAR and PXR inhepatic energy metabolism. Drug Metab Pharmacokinet 23:8–13.

Lehmann JM, McKee DD, Watson MA, Willson TM, Moore JT, and Kliewer SA (1998) Thehuman orphan nuclear receptor PXR is activated by compounds that regulate CYP3A4 geneexpression and cause drug interactions. J Clin Invest 102:1016–1023.

Lichti-Kaiser K, Brobst D, Xu C, and Staudinger JL (2009) A systematic analysis of predictedphosphorylation sites within the human pregnane X receptor protein. J Pharmacol Exp Ther331:65–76.

Lilienbaum A (2013) Relationship between the proteasomal system and autophagy. Int J BiochemMol Biol 4:1–26.

Lin W, Wu J, Dong H, Bouck D, Zeng F-Y, and Chen T (2008) Cyclin-dependent kinase 2negatively regulates human pregnane X receptor-mediated CYP3A4 gene expression in HepG2liver carcinoma cells. J Biol Chem 283:30650–30657.

Liu Y, Ye J, Shin Ogawa L, Inoue T, Huang Q, Chu J, Bates RC, Ying W, Sonderfan AJ, and RaoPE, et al. (2015) The HSP90 inhibitor ganetespib alleviates disease progression and augmentsintermittent cyclophosphamide therapy in the MRL/lpr mouse model of systemic lupuserythematosus. PLoS One 10:e0127361.

Ma X, Chen J, and Tian Y (2015) Pregnane X receptor as the “sensor and effector” in regulatingepigenome. J Cell Physiol 230:752–757.

McDonough H and Patterson C (2003) CHIP: a link between the chaperone and proteasomesystems. Cell Stress Chaperones 8:303–308.

Meacham GC, Patterson C, Zhang W, Younger JM, and Cyr DM (2001) The Hsc70 co-chaperoneCHIP targets immature CFTR for proteasomal degradation. Nat Cell Biol 3:100–105.

Morishima Y, Wang AM, Yu Z, Pratt WB, Osawa Y, and Lieberman AP (2008) CHIP deletionreveals functional redundancy of E3 ligases in promoting degradation of both signaling pro-teins and expanded glutamine proteins. Hum Mol Genet 17:3942–3952.

Mutoh S, Sobhany M, Moore R, Perera L, Pedersen L, Sueyoshi T, and Negishi M (2013)Phenobarbital indirectly activates the constitutive active androstane receptor (CAR) by in-hibition of epidermal growth factor receptor signaling. Sci Signal 6:ra31.

Ong SS, Goktug AN, Elias A, Wu J, Saunders D, and Chen T (2014) Stability of the humanpregnane X receptor is regulated by E3 ligase UBR5 and serine/threonine kinase DYRK2.Biochem J 459:193–203.

Pondugula SR, Brimer-Cline C, Wu J, Schuetz EG, Tyagi RK, and Chen T (2009) A phos-phomimetic mutation at threonine-57 abolishes transactivation activity and alters nuclear lo-calization pattern of human pregnane x receptor. Drug Metab Dispos 37:719–730.

Pondugula SR and Mani S (2013) Pregnane xenobiotic receptor in cancer pathogenesis andtherapeutic response. Cancer Lett 328:1–9.

Qing G, Yan P, and Xiao G (2006) Hsp90 inhibition results in autophagy-mediated proteasome-independent degradation of IkappaB kinase (IKK). Cell Res 16:895–901.

Rana R, Coulter S, Kinyamu H, and Goldstein JA (2013) RBCK1, an E3 ubiquitin ligase,interacts with and ubiquinates the human pregnane X receptor. Drug Metab Dispos 41:398–405.

Rathod V, Jain S, Nandekar P, and Sangamwar AT (2014) Human pregnane X receptor: a noveltarget for anticancer drug development. Drug Discov Today 19:63–70.

Sanchez ER, Toft DO, Schlesinger MJ, and Pratt WB (1985) Evidence that the 90-kDa phos-phoprotein associated with the untransformed L-cell glucocorticoid receptor is a murine heatshock protein. J Biol Chem 260:12398–12401.

Smutny T, Mani S, and Pavek P (2013) Post-translational and post-transcriptional modificationsof pregnane X receptor (PXR) in regulation of the cytochrome P450 superfamily. Curr DrugMetab 14:1059–1069.

Soss SE, Rose KL, Hill S, Jouan S, and Chazin WJ (2015) Biochemical and proteomic analysis ofubiquitination of Hsc70 and Hsp70 by the E3 ligase CHIP. PLoS One 10:e0128240.

Squires EJ, Sueyoshi T, and Negishi M (2004) Cytoplasmic localization of pregnane X receptorand ligand-dependent nuclear translocation in mouse liver. J Biol Chem 279:49307–49314.

Staudinger JL and Lichti K (2008) Cell signaling and nuclear receptors: new opportunities formolecular pharmaceuticals in liver disease. Mol Pharm 5:17–34.

Staudinger JL, Xu C, Biswas A, and Mani S (2011) Post-translational modification of pregnane xreceptor. Pharmacol Res 64:4–10.

Sugatani J, Kojima H, Ueda A, Kakizaki S, Yoshinari K, Gong QH, Owens IS, Negishi M,and Sueyoshi T (2001) The phenobarbital response enhancer module in the human bilirubinUDP-glucuronosyltransferase UGT1A1 gene and regulation by the nuclear receptor CAR.Hepatology 33:1232–1238.

Sugatani J, Yamakawa K, Tonda E, Nishitani S, Yoshinari K, Degawa M, Abe I, Noguchi H,and Miwa M (2004) The induction of human UDP-glucuronosyltransferase 1A1 mediatedthrough a distal enhancer module by flavonoids and xenobiotics. Biochem Pharmacol 67:989–1000.

Sugatani J, Sueyoshi T, Negishi M, and Miwa M (2005) Regulation of the human UGT1A1 geneby nuclear receptors CAR, PXR and GR, in Conjugation Enzymes. Methods in Enzymology(Sies H and Packer L eds) pp 92–104, Elsevier, Philadelphia.

Sugatani J, Mizushima K, Osabe M, Yamakawa K, Kakizaki S, Takagi H, Mori M, Ikari A,and Miwa M (2008) Transcriptional regulation of human UGT1A1 gene expression throughdistal and proximal promoter motifs: implication of defects in the UGT1A1 gene promoter.Naunyn Schmiedebergs Arch Pharmacol 377:597–605.

Sugatani J, Osabe M, Kurosawa M, Kitamura N, Ikari A, and Miwa M (2010) Induction ofUGT1A1 and CYP2B6 by an antimitogenic factor in HepG2 cells is mediated through sup-pression of cyclin-dependent kinase 2 activity: cell cycle-dependent expression. Drug MetabDispos 38:177–186.

Sugatani J, Uchida T, Kurosawa M, Yamaguchi M, Yamazaki Y, Ikari A, and Miwa M (2012)Regulation of pregnane X receptor (PXR) function and UGT1A1 gene expression by post-translational modification of PXR protein. Drug Metab Dispos 40:2031–2040.

Sugatani J, Hattori Y, Noguchi Y, Yamaguchi M, Yamazaki Y, and Ikari A (2014) Threonine-290regulates nuclear translocation of the human pregnane X receptor through its phosphorylation/dephosphorylation by Ca2+/calmodulin-dependent protein kinase II and protein phosphatase 1.Drug Metab Dispos 42:1708–1718.

Tang B, Cai J, Sun L, Li Y, Qu J, Snider BJ, and Wu S (2014) Proteasome inhibitors activateautophagy involving inhibition of PI3K-Akt-mTOR pathway as an anti-oxidation defense inhuman RPE cells. PLoS One 9:e103364.

Stability Control of PXR via Chaperone-Autophagy Pathway 149

at ASPE

T Journals on M

ay 28, 2020dm

d.aspetjournals.orgD

ownloaded from

Tian Y, Zhang Y, Zhong B, Wang YY, Diao FC, Wang RP, Zhang M, Chen DY, Zhai ZH, and ShuHB (2007) RBCK1 negatively regulates tumor necrosis factor- and interleukin-1-triggered NF-kappaB activation by targeting TAB2/3 for degradation. J Biol Chem 282:16776–16782.

Timsit YE and Negishi M (2007) CAR and PXR: the xenobiotic-sensing receptors. Steroids 72:231–246.

Timsit YE and Negishi M (2014) Coordinated regulation of nuclear receptor CAR by CCRP/DNAJC7, HSP70 and the ubiquitin-proteasome system. PLoS One 9:e96092.

Torimoto Y, Shindo M, Ikuta K, and Kohgo Y (2015) Current therapeutic strategies for multiplemyeloma. Int J Clin Oncol 20:423–430.

Walton GM, Bertics PJ, Hudson LG, Vedvick TS, and Gill GN (1987) A three-step purificationprocedure for protein kinase C: characterization of the purified enzyme. Anal Biochem 161:425–437.

Wang YM, Chai SC, Lin W, Chai X, Elias A, Wu J, Ong SS, Pondugula SR, Beard JA,and Schuetz EG, et al. (2015) Serine 350 of human pregnane X receptor is crucial for itsheterodimerization with retinoid X receptor alpha and transactivation of target genes in vitroand in vivo. Biochem Pharmacol 96:357–368.

Wickner S, Maurizi MR, and Gottesman S (1999) Posttranslational quality control: folding,refolding, and degrading proteins. Science 286:1888–1893.

Wu YT, Tan HL, Shui G, Bauvy C, Huang Q, Wenk MR, Ong CN, Codogno P, and Shen HM(2010) Dual role of 3-methyladenine in modulation of autophagy via different temporal pat-terns of inhibition on class I and III phosphoinositide 3-kinase. J Biol Chem 285:10850–10861.

Yamamoto S, Subedi GP, Hanashima S, Satoh T, Otaka M, Wakui H, Sawada K, Yokota S,Yamaguchi Y, and Kubota H, et al. (2014) ATPase activity and ATP-dependent conformationalchange in the co-chaperone HSP70/HSP90-organizing protein (HOP). J Biol Chem 289:9880–9886.

Zhuo W, Hu L, Lv J, Wang H, Zhou H, and Fan L (2014) Role of pregnane X receptor inchemotherapeutic treatment. Cancer Chemother Pharmacol 74:217–227.

Address correspondence to: Dr. Junko Sugatani, Department of Pharmaco-Biochemistry, School of Pharmaceutical Sciences, University of Shizuoka, 52-1Yada, Suruga-ku, Shizuoka 422-8526, Japan. E-mail: sugatani@u-shizuoka-ken.ac.jp

150 Sugatani et al.

at ASPE

T Journals on M

ay 28, 2020dm

d.aspetjournals.orgD

ownloaded from