Characterization of Activating Signal Cointegrator-2 as a Novel ...

0888-8809/05/$15.00/0 Printed in U.S.A.

Molecular Endocrinology 19(7):1711–1719 Copyright © 2005 by The Endocrine Society doi: 10.1210/me.2005-0066

Characterization of Activating Signal Cointegrator-2 as a Novel Transcriptional Coactivator of the Xenobiotic Nuclear Receptor Constitutive Androstane Receptor Eunho Choi,* Seunghee Lee,* Seon-Yong Yeom, Geun Hyang Kim, Jae Woon Lee, and Seung-Whan Kim Department of Life Science (E.C.), Pohang University of Science and Technology, Pohang 790-784, Korea; Department of Molecular and Cellular Biology (S.L.), Division Diabetes, Endocrinology & Metabolism (J.W.L.), Department of Medicine, Baylor College of Medicine, Houston, Texas 77030; and Asan Institute for Life Sciences (S.-Y.Y., G.H.K., S.-W.K.), University of Ulsan College of Medicine, Seoul 138-736, Korea Activating signal cointegrator-2 (ASC-2) is a recently isolated transcriptional coactivator protein for a variety of different transcription factors, including many members of the nuclear receptor superfamily. In this report, we demonstrate that ASC-2 also serves as a coactivator of the xenobiotic nuclear receptor constitutive androstane receptor (CAR). First, transcriptional activation by CAR was enhanced by cotransfected ASC-2 in CV-1 and HeLa cells. In contrast, CAR transactivation was significantly impaired in HepG2 cells stably expressing specific small interfering RNA directed against ASC-2. Consistent with these results, chromatin immunoprecipitation experiments revealed that ASC-2 is recruited to the known CAR target genes in a ligand-dependent manner. Secondly, CAR specifically interacted with the first LXXLL motif of ASC-2, and these interactions were stimulated by CAR agonist 1,4-bis[2(3,5-dichloropyridyloxy)]benzene and repressed by

CAR inverse agonist androstanol, suggesting that this motif may mediate the interactions of ASC-2 and CAR in vivo. In support of this idea, DN1, a fragment of ASC-2 encompassing the first LXXLL motif, suppressed CAR transactivation, and coexpressed ASC-2 but not other LXXLL-type coactivators such as thyroid hormone receptor-associated protein 220 reversed this repression. Finally, CAR was recently found to play a pivotal role in effecting the severe acetaminophen-induced liver damage. Interestingly, transgenic mice expressing DN1 were resistant to the acetaminophen-induced hepatotoxicity and expression of a series of the known CAR target genes was specifically repressed in these transgenic mice. Taken together, these results strongly suggest that ASC-2 is a bona fide coactivator of the xenobiotic nuclear receptor CAR and mediate the specific xenobiotic response by CAR in vivo. (Molecular Endocrinology 19: 1711–1719, 2005)

T

out specific DNA-binding activity are essential for transcriptional activation, which ultimately led to the identification of many proteins interacting with the C-terminal ligand-dependent transactivation domain of nuclear receptors (for reviews, see Refs. 2 and 3). These coactivators, including the p160 family, cAMP response element binding protein-binding protein/ p300, p300/CBP-associated factor, thyroid hormone receptor-associated protein (TRAP)/vitamin D receptor-interacting protein, and many others, bridge transcription factors and the basal transcription apparatus and/or remodel the chromatin structures. Activating signal cointegrator-2 (ASC-2), also named AIB3, TRBP, RAP250, NRC, and PRIP, is a recently isolated transcriptional coactivator molecule, which is gene-amplified and overexpressed in certain human cancers and stimulates transactivation by many members of the nuclear receptor superfamily, activator protein-1, nuclear factor-␬B, serum response factor, and numerous other transcription factors (4–12). In

HE NUCLEAR RECEPTOR superfamily is a group of proteins that regulate, in a ligand-dependent manner, transcriptional initiation of target genes by binding to specific DNA sequences named hormone response elements (for a review, see Ref. 1). Genetic studies indicated that transcription coactivators withFirst Published Online March 10, 2005 * E.C. and S.L. contributed equally to this work. Abbreviations: ALT, Alanine aminotransferase; APAP, Nacetyl-p-aminophenol; ASC-2, activating signal cointegrator-2; CAR, constitutive androstane receptor; ChIP, chromatin immunoprecipitation; GST, glutathione-S-transferase; LXR, liver X receptor; mCAR, mouse CAR; NAPQI, N-acetyl-p-benzoquinone imine; Q-PCR, quantitative PCR; RAR, retinoic acid receptor; RARE, RAR element; siRNA, small interfering RNA; SRC, steroid receptor coactivator; TCPOBOP, 1,4-bis[2-(3,5-dichloropyridyloxy)]benzene; TRAP, thyroid hormone receptor-associated protein. Molecular Endocrinology is published monthly by The Endocrine Society (http://www.endo-society.org), the foremost professional society serving the endocrine community.

1711

1712 Mol Endocrinol, July 2005, 19(7):1711–1719

Choi et al. • Essential Role of ASC-2 in CAR Transactivation

particular, the single cell microinjection results with ASC-2 antibody demonstrated that endogenous ASC-2 is required for transactivation by many nuclear receptors and activator protein-1 (4, 5, 12). More recently, we found that ASC-2 exists in a steady-state complex of approximately 2 MDa that contains histone H3-lysine 4-specific methyltransferase enzymes, and this complex named ASCOM (for ASC-2 complex) specifically associates with the retinoid-responsive ␤-RARE [retinoic acid receptor (RAR) element] and p21WAF1 promoter regions in a ligand-dependent manner (13). Interestingly, ASC-2 contains two nuclear receptor-interaction domains (11), both of which are dependent on the integrity of their core LXXLL sequences (14, 15). The C-terminal LXXLL motif specifically interacts with oxysterol receptors liver X receptor (LXR) ␣ and LXR␤, whereas the N-terminal motif binds a broad range of nuclear receptors (11). The xenobiotic nuclear receptor constitutive androstane receptor (CAR; NR1I3) is known to heterodimerize with retinoid X receptor and function as a cellular sensor that is capable of responding to a variety of xenobiotic exposure (reviewed in Refs. 16 and 17). CAR/retinoid X receptor heterodimers bind many distinct hormone response elements and constitutively activate transcription in the absence of added ligands. The constitutive activity of mouse CAR can be inhibited by the inverse agonist ligand androstanol (18). In contrast, the constitutive activity of both mouse and human CARs are increased by the agonist 1,4-bis[2(3,5-dichloropyridyloxy)]benzene (TCPOBOP) (19). In particular, one of the best-characterized CAR-binding sites are found within a phenobarbital-enhancer element on the CYP2B gene, and the predominant phenobarbital-inducible CYP2B genes in mouse and rat are CYP2B10 and CYP2B1/2, respectively (16, 17). Overdoses of acetaminophen (APAP; also known as 4⬘-hydroxyacetanilide, N-acetyl-p-aminophenol, and paracetamol) are the leading cause of hospital admission for acute liver failure in the United States (20). Ingestion of amounts of APAP only two to three times the maximum daily recommended dose can cause hepatotoxicity, and higher doses result in centrilobular necrosis that is potentially fatal (21, 22). The basis for this toxicity has been well studied. Particularly at high doses, cytochrome P-450 enzymes CYP1A2, CYP2E1, and isoforms of CYP3A convert APAP to a reactive quinone form, N-acetyl-p-benzoquinone imine (NAPQI) (23–26), that covalently binds to cellular macromolecules and also causes production of reactive oxygen species (27, 28). At subtoxic doses, NAPQI is inactivated by glutathione-S-transferases (GSTs) via conjugation with reduced glutathione, but NAPQI accumulates when glutathione levels are depleted. Among the numerous GST enzymes, the GSTPi isoforms are particularly effective at inactivating NAPQI (29). Their importance in APAP toxicity was confirmed by the unexpected demonstration that knockout mice lacking both GSTPi isoforms are relatively resistant to APAP hepatotoxicity because of a

decreased rate of glutathione depletion (30). Interestingly, CAR was recently identified as a key regulator of this APAP metabolism and hepatotoxicity (31). CAR activators as well as high doses of APAP-induced expression of CYP1A2, CYP3A11, and GSTPi in wildtype but not in CAR null mice, and the CAR null mice were resistant to APAP toxicity. In addition, inhibition of CAR activity by administration of the inverse agonist ligand androstanol 1 h after APAP treatment blocked hepatotoxicity in wild-type but not in CAR null mice (31). Gene targeting approaches to elucidate the role of many coactivators in mice have often been hampered by early embryonic lethality or functional redundancy. In particular, deletion of the ASC-2 gene also resulted in early embryonic lethality (32–35). As an alternative approach, we have expressed a dominant-negative fragment of ASC-2 encompassing the N-terminal LXXLL motif (named DN1) in mice, which specifically inhibited recruitment of the endogenous ASC-2 to nuclear receptors (36). These DN1-TG mice exhibited a plethora of developmental and phenotypic abnormalities in mice, including problems with eye, heart, motor activities, and fat metabolism in the liver, and these mice were significantly compromised for their ability to respond to exogenous ligands, including retinoids and others (36). More recently, we have reported a similar approach in establishing transgenic mice expressing DN2, a dominant-negative fragment of ASC-2 that encompasses the LXR-specific second LXXLL motif and potently represses transactivation by LXRs in cotransfections (37). Accordingly, these DN2-TG mice exhibited phenotypes that are highly homologous to those previously observed with LXR␣ null mice (38), including a rapid accumulation of large amounts of cholesterol and down-regulation of the known lipid metabolizing target genes of LXR␣ in the liver upon feeding high-cholesterol diet (37). Together with the DN1-TG mice results (36), these results strongly suggested that ASC-2 is a physiologically pivotal transcriptional coactivator protein of LXRs and other nuclear receptors in vivo. It is important to note that both DN1 and DN2 appear to be specific to ASC-2; i.e. DN1/2-mediated suppression of nuclear receptor transactivation was specifically restored by coexpressed ASC-2 but not by other LXXLL-type coactivators steroid receptor coactivator (SRC)-1 and TRAP220, as demonstrated with chromatin immunoprecipitations (ChIPs) and cotransfections (36, 37). The basis for this specificity is not entirely clear. However, Kraus et al. (39) have recently provided an important clue to this specificity. They showed that although SRCs and TRAP220 share overlapping binding sites on nuclear receptors, information contained in the receptor-coactivator interface allows the receptor to distinguish between them. In support of this conclusion, they have identified an estrogen receptor ␣ AF-2 point mutant (L540Q) that selectively binds and recruits TRAP220 but not SRCs (39). Collectively, their results clearly demonstrated that facilitated recruit-


Mol Endocrinol, July 2005, 19(7):1711–1719 1713

ment, rather than competition, drives the sequential recruitment of different LXXLL-type coactiavtor complexes to nuclear receptors. To more definitely clarify this specificity issue, however, we are currently mutating each of these LXXLL motifs in the endogenous ASC-2 gene using a specific knock-in strategy in mice. In this report, we demonstrate that ASC-2 is a bona fide coactivator of the xenobiotic nuclear receptor CAR and plays an essential role in the specific xenobiotic response mediated by this receptor in vivo. Overall, our results suggest that ASC-2, like CAR itself, may serve as a novel therapeutic target for treating the adverse effects of APAP and potentially other hepatotoxic agents.

port of an idea that ASC-2 is indeed a transcriptional coactivator of CAR, an increasing amount of ASC-2expression vector significantly stimulated both the basal and TCPOBOP-induced levels of transactivation mediated by CAR in cotransfected CV-1 cells (Fig. 1A). Similar results were also obtained with HeLa, HepG2, and HEK293 cells (data not shown). To further confirm these results, we have attempted to express small interfering RNA (siRNA) directed against ASC-2 in different cell lines to specifically knock down the expression level of ASC-2. Thus far, we have succeeded with three different human cell lines: HepG2, HeLa, and HEK293 (Fig. 1B and data not shown). Among four different regions of human ASC-2 we have tested, siRNA against the C-terminal region (nucleotides 6153–6173; AACCAGTGCGGTGCAATCCAA) was found to efficiently knock down expression of ASC-2. In Western analyses, three independent clones of HepG2 cells expressing this siRNA exhibited less than 50% of ASC-2 from cells expressing control siRNA (Fig. 1B). Confirming the specificity of this siRNA against ASC-2, these cells didn’t affect the expression level of ␣/␤-tubulins. We have cotransfected clone no. 1 HepG2 cells with the CAR-responsive RARE␤2-TKLUC reporter construct and an expression vector for

RESULTS AND DISCUSSION ASC-2 as a Transcriptional Coactivator of CAR ASC-2 has been demonstrated to function as a transcriptional coactivator of many members of the nuclear receptor superfamily (4–12). These led us to test whether ASC-2 is also involved with transactivation by the xenobiotic nuclear receptor CAR (16, 17). In sup-

Fig. 1. ASC-2, as a Transcriptional Coactivator of CAR A, CAR-responsive RARE␤2-TK-LUC reporter construct was cotransfected into CV-1 cells along with LacZ expression vector (100 ng) and expression vectors for CAR and ASC-2 either in the absence or presence of 0.25 ␮M of TCPOBOP, as indicated. Normalized luciferase expressions from triplicate samples were calculated relative to the LacZ expressions. B, Western analyses of HepG2 cells stably expressing either control siRNA or specific siRNA against ASC-2 (three independent clones are as shown). The expression level of ASC-2 in all three clones was significantly lower than that from cells expressing control siRNA. In contrast, the expression level of ␣/␤-tubulins was not affected. C, The above clone no. 1 and control HepG2 cells were cotransfected with reporter constructs RARE␤2-TK-LUC (left panel), Smad3-responsive p(CAGA)9-MLP-LUC (right panel), LacZ expression vector (100 ng), and expression vectors for CAR and ASC-2 either in the absence or presence of 0.25 ␮M of TCPOBOP (left panel) or Smad3 (right panel), as indicated. Normalized luciferase expressions from triplicate samples were calculated relative to the LacZ expressions.


mouse CAR (mCAR) either in the absence or presence of 0.25 ␮M of the CAR agonist TCPOBOP because these cells express only a negligible amount of CAR (16). Interestingly, CAR transactivation in these ASC-2 knockdown cells was dramatically impaired when compared with that in cells expressing control siRNA (Fig. 1C, left panel). In contrast, transactivation by Smad3 was still intact in these ASC-2 knockdown cells when measured with Smad3-responsive reporter construct p(CAGA)9-MLP-LUC (40) (Fig. 1C, right panel). Similarly, transactivation directed by Gal4 fusion to VP16 was also intact in these ASC-2 knockdown cells (data not shown). Thus, the diminished expression of ASC-2 specifically affects transactivation by CAR but has no effect with Smad3 and VP16. Overall, these two different lines of cotransfection data strongly suggest that ASC-2 is likely an essential transcriptional coactivator protein of CAR. ASC-2 Is Recruited to CAR Target Genes in ChIP We have previously shown that ASC-2 is directly recruited to RAR and LXR target genes in a liganddependent manner, as demonstrated by ChIP assays (13, 37). The above cotransfection results suggest that ASC-2 will also be recruited to CAR target genes. To further confirm this prediction, we used HEK293 cells, mainly due to their high transfection efficiency, to determine whether ASC-2 is indeed recruited to two well-known CAR target genes CYP3A4 (41) and CYP2B6 (42). In ChIP assays, when cells were not cotransfected with an expression vector for mCAR, anti-ASC-2 antibody failed to detect any recruitment of ASC-2 to CYP3A4 even in the presence of increasing concentration of TCPOBOP (Fig. 2A). However, when mCAR was cotransfected, 0.25 ␮M of TCPOBOP was observed to stimulate the recruitment of ASC-2 to both of these two target genes (Fig. 2, B and C). It was also evident that 4 ␮M of the CAR inverse agonist androstanol suppressed the recruitment of ASC-2 to the CYP2B6 gene, although this effect wasn’t clearly observed with CYP3A4. In addition, the ASC-2 recruitment was time dependent with the maximum loading of ASC-2 in three min after adding TCPOBOP (Fig. 2, B and C). Collectively, these results clearly demonstrate that ASC-2 is a bona fide transcriptional coactivator of CAR, and ASC-2 is directly recruited to CAR target genes in vivo in a CAR/ligand-dependent manner. Specific Interactions of CAR with the First LXXLL Motif of ASC-2 ASC-2 contains two distinct LXXLL motifs, and we have previously demonstrated that the second LXXLL motif is highly specific to LXRs, whereas the first motif recognizes a variety of different nuclear receptors (11). In addition, we and others have shown that these two motifs likely mediate the interactions between ASC-2/ ASCOM and its target nuclear receptors (4–12, 36, 37).


Fig. 2. Recruitment of ASC-2 to CAR Target Genes A, ChIP experiments were executed in HEK293 cells. ASC-2 was not recruited to CYP3A4 in the absence of cotransfected mCAR even in the increasing concentration of TCPOBOP. B and C, HEK293 cells transfected with expression vector for mCAR were used in ChIP assays. The endogenous ASC-2 was recruited to the promoter regions of two CAR target genes CYP3A4 (B) and CYP2B6 (C) in a liganddependent manner. A total of 0.25 ␮M of TCPOBOP and 4 ␮M of androstanol were used in these experiments, respectively. The ASC-2 recruitment to these genes were not observed with IgG (data not shown).

Because the results shown in Figs. 1 and 2 clearly demonstrated that ASC-2 functions as a coactivator of CAR and ASC-2 is likely recruited to CAR target genes via its interactions with CAR, we examined whether these motifs are indeed involved with recruiting ASC2/ASCOM to CAR. We have employed the yeast twohybrid assays in which we used two sets of ASC-2 fragments containing the first and second LXXLL motifs, as depicted in Fig. 3A. In yeast, coexpression of LexA fusion to DN1, but not DN1/m in which the LXXLL motif was mutated to LXXAA, and B42 fusion to mCAR resulted in significant activation of LexA-driven LacZ reporter construct, whereas LexA-DN1 or B42-mCAR alone was inert (Fig. 3B). These results suggest that CAR interacts with the first LXXLL motif of ASC-2. In further support of the CAR-specificity of these interactions, this LacZ activation was disrupted by 4 ␮M of androstanol but significantly stimulated by 0.25 ␮M of TCPOBOP (Fig. 3B). As expected from the remarkable specificity of the second LXXLL motif of ASC-2 toward LXRs (11), DN2 containing the second LXXLL motif didn’t show any detectable interactions with CAR in the yeast two-hybrid assays, whereas DN2, but not DN2/m in which the LXXLL motif was mutated to LXXAA, strongly interacted with LXR (Fig. 3B). Consistent with these results, GST fusion to DN1 interacted with 35S-labeled, in vitro-translated mCAR protein, and these interactions were also suppressed by 4 ␮M of androstanol but stimulated by 0.25 ␮M of TCPOBOP (Fig. 3C). Taken together, these results strongly suggest that the first LXXLL motif of ASC-2 specifically


Fig. 3. Interactions of ASC-2 and CAR A, Schematic representation of ASC-2, DN1, DN1/m, DN2, and DN2/m. DN1/m and DN2/m have mutations in their LXXLL motifs, and thus they are unable to interact with nuclear receptors. B, The indicated B42- and LexA-plasmids were transformed into a yeast strain containing an appropriate LacZ reporter gene, as described (4). Open, closed, and gray boxes indicate the presence of vehicle alone, 4 ␮M of androstanol, and 0.25 ␮M of TCPOBOP, respectively. C, The full-length mCAR was labeled with [35S]methionine by in vitro translation and incubated with glutathione beads containing GST alone and GST-DN1, as indicated. Beads were washed, and specifically bound material was eluted with reduced glutathione and resolved by SDS-PAGE. A total of 4 ␮M of androstanol and 0.25 ␮M of TCPOBOP were used in these experiments, respectively. Approximately 20% of the total reaction mixture was loaded as input.

interacts with CAR, which may play a pivotal role to recruit ASC-2/ASCOM to CAR on its target genes. ASC-2 as an Important Player in CAR/APAPMediated Hepatotoxicity We have previously reported the characterization of transgenic mice expressing DN1, a dominant-negative fragment of ASC-2 that contains the first LXXLL motif and specifically competes with the endogenous ASC-2 to bind retinoic acid and other nuclear recep-


tors in vivo (36). Similarly, DN1 specifically blocked ASC-2 from mediating CAR transactivation in CV-1 and other cells: i.e. expression of DN1 suppressed CAR transactivation, which was significantly restored by coexpressed ASC-2 but not by TRAP220 (Fig. 4A). SRC-1 was also unable to rescue this repression (data not shown). Interestingly, DN1/m resulted in a significant enhancement with CAR transactivation (Fig. 4A). The molecular mechanisms for these intriguing results are not currently clear. Although resolution of this issue warrants additional studies, it is noted that the structure of the DN1 region of ASC-2 is considered highly complex with the presence of multiple other protein-protein interaction interfaces (4, 5, 8, 11). Nonetheless, the apparent specificity of DN1 against ASC-2 but not other LXXLL-containing coactivators TRAP220 and SRC-1 in cotransfections strongly suggests that CAR transactivation should be significantly impaired in DN1-TG mice. CAR was recently shown to be important for the APAP toxicity (31). Thus, to directly test this idea, we have examined whether DN1-TG mice (no. 104, as described in Ref. 36) are resistant to APAP toxicity. We pretreated both wildtype and DN1-TG mice with either TCPOBOP (0.1 mg/ kg of body weight) or the vehicle control, followed by APAP (250 mg/kg of body weight). Of the various clinical laboratory markers for hepatic injury, serum transaminases, especially alanine aminotransferase (ALT), are the most universally important indicators for studies ranging from early preclinical animal testing to postmarketing patient monitoring. Neither TCPOBOP nor this dose of APAP alone induced hepatotoxicity (31), as indicated by the serum levels of the liver enzyme ALT (Fig. 4B and data not shown). Histological examination also indicated that normal intact structures around their centrilobular veins (indicated as arrows in Fig. 4C) were observed with the livers from either wild-type mice treated with vehicle or APAP alone or all the DN1-TG mice. However, animals treated with TCPOBOP plus APAP showed significantly enhanced ALT levels at 24 h (Fig. 4B) as well as severe hepatic centrilobular necrosis, which reveals highly destructured, necrotic lesions around their hepatic centrilobular veins (indicated as arrows in Fig. 4C). Another independent DN1-TG line no. 71 that we have previously reported (36) also resulted in similar resistance to the APAP hepatotoxicity, whereas transgenic mice expressing DN1/m (36) exhibited no such resistance at all (data not shown). Thus, DN1-TG mice appear to be more resistant to the APAP-mediated hepatotoxicity. These results are highly homologous to those reported with CAR knockout mice (31, 43). Taken together, these results strongly suggest that ASC-2 is indeed a physiologically important coactivator of CAR in vivo. Among genes associated with APAP toxicity, TCPOBOP treatment is known to induce CYP1A2, CYP3A11, and GSTPi mRNAs in wild-type animals but not in CAR null mice (31). Similarly, we have observed that these three genes as well as another well-char-



Fig. 4. Resistance to the APAP-Induced Hepatotoxicity in DN1-TG A, CAR-responsive RARE␤2-TK-LUC reporter construct was cotransfected into CV-1 cells, along with LacZ expression vector (100 ng) and expression vectors for CAR, DN1, DN1/m, ASC-2 and TRAP220, as indicated. Similar results were also obtained with HeLa cells (data not shown). Normalized luciferase expressions from triplicate samples were calculated relative to the LacZ expressions. B, Blood samples were collected from the treated animals in (B) 24 h later, and serum ALT levels were measured (n ⫽ 5–7). C, Wild-type (WT) and DN1-TG animals pretreated with TCPOBOP or vehicle alone were administered a 250-mg/kg dose of APAP by ip injection (n ⫽ 5–7 per treatment group). Liver sections from different treatments were examined by histological staining. Liver samples from all treated animals were analyzed, but only representative histology is presented. TCPOBOPpretreated livers from wild-type animals showed extensive hepatic centrilobular necrosis with APAP administration. In contrast, other treatments with wild-type mice as well as all the treatment with DN1-TG mice showed normal morphology. Centrilobular veins are indicated as arrows. D, Wild-type and DN1-TG animals were treated with TCPOBOP or vehicle alone for 3 d. Total liver RNA was prepared and subjected to RT-PCR analysis with the indicated probes. More quantitative Q-PCR experiments were also executed for CYP2B10. Open and closed boxes indicate samples from wild-type and DSN1-TG mice, respectively. GAPDH, Glyceraldehyde 3-phosphate dehydrogenase.

acterized CAR target gene CYP2B10 are induced by TCPOBOP (0.1 mg/kg of body weight) in wild-type animals (Fig. 4D, left panel). However, both the basal and induced levels of their expressions were significantly impaired in DN1-TG mice no. 104. The results with CYP2B10 were independently confirmed in more quantitative real-time quantitative PCR (Q-PCR) analyses (Fig. 4D, right panel). We have also obtained similar results with DN1-TG no. 71 (data not shown). It is important to note that DN1 is a competitive inhibitor of the endogenous ASC-2 and doesn’t completely abolish the basal and TCPOBOP-induced levels of transactivation by CAR. Thus, loss of CAR function in DN1-TG animals appears to result in resistance to APAP toxicity that is associated with the impaired induction of APAP-metabolizing enzymes. Taken together, these results suggest that ASC-2 participate in APAP-mediated hepatotoxicity as a transcriptional coactivator of the xenobiotic nuclear receptor CAR in mice. In summary, we have shown that ASC-2 is a functional coactivator of the xenobiotic nuclear receptor CAR and likely mediates its specific xenobiotic response in vivo. Thus, CAR joins and extends the list of

members of the nuclear receptor superfamily that require ASC-2 as a transcriptional coactivator. We have found that the transcriptional coregulatory activities of ASC-2 can be modulated by calmodulin-dependent kinase IV (44) as well as a variety of other signal transduction pathways (our unpublished results), and thus it is an exciting possibility that the identification of ASC-2 antagonists along with specific inverse agonists of CAR (such as androstanol) may provide a clinically useful means to treat toxicity of APAP and potentially other hepatotoxic agents.

MATERIALS AND METHODS Plasmids The mammalian expression vectors for CAR, ASC-2, DN1, DN1/m, Smad3, and TRAP220; the GST pull-down vectors encoding GST and GST-DN1; and the yeast two-hybrid vectors encoding LexA, LexA-DN1, LexA-DN1/m, LexA-DN2, LexA-DN2/m, B42, LexA-B42, B42-mCAR, and B42-LXR␤, along with the transfection indicator construct pRSV-␤-gal and the reporter constructs RARE␤2-LUC, p(CAGA)9-MLP-



LUC and LexA-LacZ, were as previously described (4, 5, 11–13, 19, 40).

vitrogen Corp., Carlsbad, CA), and RT-PCRs were performed as described previously (36, 37). For the SYBR Green Q-PCR, 250 ng of cDNA was used per reaction. Each 25-␮l SYBR Green reaction consisted of 5 ␮l of cDNA (50 ng/␮l), 12.5 ␮l of 2⫻ Universal SYBR Green PCR Master Mix (PE Biosystems, Foster City, CA), and 3.75 ␮l of 50 nM forward and reverse primers. Optimization was performed for each genespecific primer before the experiment to confirm that 50 nM primer concentrations did not produce nonspecific primerdimer amplification signal in no-template control tubes. QPCR was performed on ABI 5700 PCR Instrument (Applied Biosystems, Foster City, CA) by using three-stage program parameters provided by the manufacturer as follows: 2 min at 50 C, 10 min at 95 C, and then 40 cycles of 15 sec at 95 C, and 1 min at 60 C. Specificity of the produced amplification product was confirmed by examination of dissociation reaction plots. A distinct single peak indicated that single DNA sequence was amplified during PCR. In addition, end reaction products were visualized on ethidium bromide-stained 1.4% agarose gels, and appearance of a single band of the correct molecular size confirmed specificity of the PCR. Each sample was tested in triplicate with Q-PCR. Cyclophilin A mRNA levels were used as a reference standard.

Transfections CV-1, HeLa, or HepG2 cells were grown in 24-well plates with medium supplemented with 10% fetal calf serum for 24 h and transfected with 100 ng of LacZ expression vector pRSV-␤gal and 100 ng of RARE␤2-LUC or Smad3-TK-LUC, along with indicated amount of various mammalian expression vectors. Total amounts of expression vectors were kept constant by adding pcDNA3. Transfections and luciferase assays were done as described (4), and the results were normalized to the LacZ expression. Similar results were obtained in more than two similar experiments. RNA Interference The siRNA target finder algorithm in the Ambion (Austin, TX) web site (www.ambion.com) was used to select 21-nucleotide oligomers in human ASC-2 to be tested for RNA interference. A control 19-nucleotide sequence was generated that had the same G-C content as the selected oligomers but did not display sequence identity with ASC-2. Basic local alignment and search tool analysis ensured that sequence identity between oligonucleotides and mammalian cDNAs in expressed sequence tag databases was 15 nucleotides or less. Double-stranded DNA fragments, which contained the selected RNA interference sequences positioned downstream of the human U6 RNA polymerase III promoter, were generated by PCR using the Ambion Silencer Express kit according to the manufacturer’s instructions. The DNA fragments were cloned into the pSec vector (Ambion) and sequenced. HepG2 cells were transfected with pSec derivatives. Two days later, ASC-2 expression levels were determined by immunoblots using the ASC-2 monoclonal antibody (13). Yeast Two-Hybrid and GST Pull-Down Assays The yeast two-hybrid assay was done as described (4). For each experiment, at least three independently derived colonies expressing chimeric proteins were tested. The GST fusions or GST alone was expressed in Escherichia coli, bound to glutathione-Sepahrose-4B beads (Pharmacia, Piscataway, NJ) in binding buffer [25 mM HEPES (pH 7.8), 0.2 mM EDTA, 20% glycerol, 100 mM KCl, and 0.1% Nonidet P-40], and incubated with labeled proteins expressed by in vitro translation by using the TNT-coupled transcription-translation system, with conditions as described by the manufacturer (Promega, Madison, WI). Specifically bound proteins were eluted from beads with 40 mM reduced glutathione in 50 mM Tris (pH 8.0) and analyzed by SDS-PAGE and autoradiography as described (4). ChIPs 293T cells were cotransfected with an expression vector for mouse CAR and treated with either 4 ␮M of androstanol or 0.25 ␮M of TCPOBOP for 0–10 min. Soluble chromatin from these cells was prepared and immunoprecipitated with antiASC-2 antibody, as recently described (45). The final DNA extractions were amplified using pairs of primers that encompass the CAR-responsive element of the CYP3A4 and CYP2B6 promoter regions (41, 42). RT-PCR and Q-PCR Total RNA was isolated from mouse hepatocytes after lysis in TRIzol reagent according to the manufacturer’s protocol (In-

Immunohistochemistry and ALT Assays The livers were excised, fixed with 10% formaldehyde, embedded in paraffin, sectioned in 4-␮m slices and stained with hematoxylin and eosin as described previously (36). The ALT assays were performed as described (31).

Acknowledgments Received January 27, 2005. Accepted March 4, 2005. Address all correspondence and requests for reprints to: Jae Woon Lee, Ph.D., Division Diabetes, Endocrinology & Metabolism, Department of Medicine, Baylor College of Medicine, Houston, Texas 77030. E-mail: [email protected]; or Seung-Whan Kim, Ph.D., Asan Institute for Life Sciences, University of Ulsan College of Medicine, Seoul 138-736, Korea. E-mail: [email protected]. This work was supported by 21C Frontier Functional Proteomics Project from the Ministry of Science & Technology (to J.W.L. and Y.K.P.), National Institutes of Health Grant DK064678 (to J.W.L.), and the Korea Research Foundation Grant KRF-2003-041-C00251 and KRF-2004-015-E00059 (to S.-W.K.).

REFERENCES 1. Mangelsdorf DJ, Thummel C, Beato M, Herrlich P, Schutz G, Umesono K, Blumberg B, Kastner P, Mark M, Chambon P, Evans RM 1995 The nuclear receptor superfamily: the second decade. Cell 83:835–839 2. McKenna NJ, Lanz RB, O’Malley BW 1999 Nuclear receptor coregulators: cellular and molecular biology. Endocr Rev 20:321–344 3. Lee JW, Lee YC, Na SY, Jung DJ, Lee SK 2001 Transcriptional coregulators of the nuclear receptor superfamily: coactivators and corepressors. Cell Mol Life Sci 58:289–297 4. Lee SK, Anzick SL, Choi JE, Bubendorf L, Guan XY, Jung YK, Kallioniemi OP, Kononen J, Trent JM, Azorsa D, Jhun BH, Cheong JH, Lee YC, Meltzer PS, Lee JW 1999 A nuclear factor, ASC-2, as a cancer-amplified transcriptional coactivator essential for ligand-dependent transactivation by nuclear receptors in vivo. J Biol Chem 274: 34283–34293



5. Lee SK, Na SY, Jung SY, Choi JE, Jhun BH, Cheong J, Meltzer PS, Lee YC, Lee JW 2000 Activating protein-1, nuclear factor-␬B, and serum response factor as novel target molecules of the cancer-amplified transcription coactivator ASC-2. Mol Endocrinol 14:915–925 6. Tanner MM, Tirkkonen M, Kallioniemi A, Isola J, Kuukasjarvi T, Collins C, Kowbel D, Guan XY, Trent J, Gray JW, Meltzer P, Kallioniemi OP 1996 Independent amplification and frequent co-amplification of three nonsyntenic regions on the long arm of chromosome 20 in human breast cancer. Cancer Res 56:3441–3445 7. Zhu Y, Kan L, Qi C, Kanwar YS, Yeldandi AV, Rao MS, Reddy JK 2000 Isolation and characterization of peroxisome proliferator-activated receptor (PPAR) interacting protein (PRIP) as a coactivator for PPAR. J Biol Chem 275:13510–13516 8. Mahajan MA, Samuels HH 2000 A new family of nuclear receptor coregulators that integrate nuclear receptor signaling through CREB-binding protein. Mol Cell Biol 20: 5048–5063 9. Ko L, Cardona GR, Chin WW 2000 Thyroid hormone receptor-binding protein, an LXXLL motif-containing protein, functions as a general coactivator. Proc Natl Acad Sci USA 97:6212–6217 10. Caira F, Antonson P, Pelto-Huikko M, Treuter E, Gustafsson JA 2000 Cloning and characterization of RAP250, a novel nuclear receptor coactivator. J Biol Chem 275: 5308–5317 11. Lee SK, Jung SY, Kim YS, Na SY, Lee YC, Lee JW 2001 Two distinct nuclear receptor-interaction domains and CREB-binding protein-dependent transactivation function of activating signal cointegrator-2. Mol Endocrinol 15:241–254 12. Goo YH, Na SY, Zhang H, Xu J, Hong S, Cheong J, Lee SK, Lee JW 2004 Interactions between activating signal cointegrator-2 and the tumor suppressor retinoblastoma in androgen receptor transactivation. J Biol Chem 279: 7131–7135 13. Goo YH, Sohn YC, Kim DH, Kim SW, Kang MJ, Jung DJ, Kwak E, Barlev NA, Berger SL, Chow V Y, Roeder RG, Azorsa DO, Meltzer PS, Suh PG, Song EJ, Lee KJ, Lee YC, Lee JW 2003 Activating signal cointegrator 2 belongs to a novel steady-state complex that contains a subset of trithorax group proteins. Mol Cell Biol 23: 140–149 14. Heery DM, Kalkhoven E, Hoare S, Parker MG 1997 A signature motif in transcriptional co-activators mediates binding to nuclear receptors. Nature 387:733–736 15. Torchia J, Rose DW, Inostroza J, Kamei Y, Westin S, Glass CK, Rosenfeld MG 1997 The transcriptional coactivator p/CIP binds CBP and mediates nuclear-receptor function. Nature 387:677–684 16. Tzameli I, Moore DD 2001 Role reversal: new insights from new ligands for the xenobiotic receptor CAR. Trends Endocrinol Metab 12:7–10 17. Sonoda J, Rosenfeld JM, Xu L, Evans RM, Xie W 2003 A nuclear receptor-mediated xenobiotic response and its implication in drug metabolism and host protection. Curr Drug Metab 4:59–72 18. Forman BM, Tzameli I, Choi HS, Chen J, Simha D, Seol W, Evans RM, Moore DD 1998 Androstane metabolites bind to and deactivate the nuclear receptor CAR-␤. Nature 395:612–615 19. Tzameli I, Pissios P, Schuetz EG, Moore DD 2000 The xenobiotic compound 1,4-bis[2-(3,5-dichloropyridyloxy)] benzene is an agonist ligand for the nuclear receptor CAR. Mol Cell Biol 20:2951–2958 20. Gill RQ, Sterling RK 2001 Acute liver failure. J Clin Gastroenterol 33:191–198 21. Prescott LF 1983 Paracetamol overdosage. Pharmacological considerations and clinical management. Drugs 25:290–314

22. Hinson JA, Roberts DW, Benson RW, Dalhoff K, Loft S, Poulsen HE 1990 Mechanism of paracetamol toxicity. Lancet 335:732 23. Snawder JE, Roe AL, Benson RW, Roberts DW 1994 Loss of CYP2E1 and CYP1A2 activity as a function of acetaminophen dose: relation to toxicity. Biochem Biophys Res Commun 203:532–539 24. Patten CJ, Thomas PE, Guy RL, Lee M, Gonzalez FJ, Guengerich FP, Yang CS 1993 Cytochrome P450 enzymes involved in acetaminophen activation by rat and human liver microsomes and their kinetics. Chem Res Toxicol 6:511–518 25. Raucy JL, Lasker JM, Lieber CS, Black M 1989 Acetaminophen activation by human liver cytochromes P450IIE1 and P450IA2. Arch Biochem Biophys 271: 270–283 26. Thummel KE, Lee CA, Kunze KL, Nelson SD, Slattery JT 1993 Oxidation of acetaminophen to N-acetyl-p-aminobenzoquinone imine by human CYP3A4. Biochem Pharmacol 45:1563–1569 27. Rogers LK, Moorthy B, Smith CV 1997 Acetaminophen binds to mouse hepatic and renal DNA at human therapeutic doses. Chem Res Toxicol 10:470–476 28. Arnaiz SL, Llesuy S, Cutrin JC, Boveris A 1995 Oxidative stress by acute acetaminophen administration in mouse liver. Free Radic Biol Med 19:303–310 29. Coles B, Wilson I, Wardman P, Hinson JA, Nelson SD, Ketterer B 1998 The spontaneous and enzymatic reaction of N-acetyl-p-benzoquinonimine with glutathione: a stopped-flow kinetic study. Arch Biochem Biophys 264: 253–260 30. Henderson CJ, Wolf CR, Kitteringham N, Powell H, Otto D, Park BK 2000 Increased resistance to acetaminophen hepatotoxicity in mice lacking glutathione S-transferase Pi. Proc Natl Acad Sci USA 97:12741–12745 31. Zhang J, Huang W, Chua SS, Wei P, Moore DD 2002 Modulation of acetaminophen-induced hepatotoxicity by the xenobiotic receptor CAR. Science 298:422–424 32. Kuang SQ, Liao L, Zhang H, Pereira FA, Yuan Y, DeMayo FJ, Ko L, Xu J 2002 Deletion of the cancer-amplified coactivator AIB3 results in defective placentation and embryonic lethality. J Biol Chem 277:45356–45360 33. Zhu YJ, Crawford SE, Stellmach V, Dwivedi RS, Rao MS, Gonzalez FJ, Qi C, Reddy JK 2003 Coactivator PRIP, the peroxisome proliferator-activated receptorinteracting protein, is a modulator of placental, cardiac, hepatic, and embryonic development. J Biol Chem 278:1986–1990 34. Antonson P, Schuster GU, Wang L, Rozell B, Holter E, Flodby P, Treuter E, Holmgren L, Gustafsson JA 2003 Inactivation of the nuclear receptor coactivator RAP250 in mice results in placental vascular dysfunction. Mol Cell Biol 23:1260–1268 35. Mahajan MA, Das S, Zhu H, Tomic-Canic M, Samuels HH 2004 The nuclear hormone receptor coactivator NRC is a pleiotropic modulator affecting growth, development, apoptosis, reproduction, and wound repair. Mol Cell Biol 24:4994–5004 36. Kim SW, Cheong C, Sohn YC, Goo YH, Oh WJ, Park JH, Joe SY, Kang HS, Kim DK, Kee CW, Lee JW, Lee HW 2002 Multiple developmental defects derived from impaired recruitment of ASC-2 to nuclear receptors in mice: implication for posterior lenticonus with cataract. Mol Cell Biol 22:8409–8414 37. Kim SW, Park K, Kwak E, Choi E, Lee S, Ham J, Kang H, Kim JM, Hwang SY, Kong YY, Lee K, Lee JW 2003 Activating signal cointegrator 2 required for liver lipid metabolism mediated by liver X receptors in mice. Mol Cell Biol 23:3583–3592 38. Peet DJ, Turley SD, Ma W, Janowski BA, Lobaccaro JM, Hammer RE, Mangelsdorf DJ 1998 Cholesterol and bile acid metabolism are impaired in mice lacking the nuclear oxysterol receptor LXR ␣. Cell 93:693–704



39. Acevedo ML, Lee KC, Stender JD, Katzenellenbogen BS, Kraus WL 2004 Selective recognition of distinct classes of coactivators by a ligand-inducible activation domain. Mol Cell 13:725–738 40. Mithani SK, Balch GC, Shiou SR, Whitehead RH, Datta PK, Beauchamp RD 2004 Smad3 has a critical role in TGF-␤-mediated growth inhibition and apoptosis in colonic epithelial cells. J Surg Res 117:296–305 41. Goodwin B, Hodgson E, D’Costa DJ, Robertson GR, Liddle C 2002 Transcriptional regulation of the human CYP3A4 gene by the constitutive androstane receptor. Mol Pharmacol 62:359–365 42. Sueyoshi T, Kawamoto T, Zelko I, Honkakoski P, Negishi M 1999 The repressed nuclear receptor CAR responds to

phenobarbital in activating the human CYP2B6 gene. J Biol Chem 274:6043–6046 43. Wei P, Zhang J, Egan-Hafley M, Liang S, Moore DD 2000 The nuclear receptor CAR mediates specific xenobiotic induction of drug metabolism. Nature 407: 920–923 44. Sohn YC, Kwak E, Na Y, Lee JW, Lee SK 2001 Silencing mediator of retinoid and thyroid hormone receptors and activating signal cointegrator-2 as transcriptional coregulators of the orphan nuclear receptor Nur77. J Biol Chem 276:43734–43739 45. Shang Y, Myers M, Brown M 2002 Formation of the androgen receptor transcription complex. Mol Cell 9:601–610

Molecular Endocrinology is published monthly by The Endocrine Society (http://www.endo-society.org), the foremost professional society serving the endocrine community.