Role of the Coiled-coil Coactivator (CoCoA) in Aryl Hydrocarbon Receptor-mediated Transcription*

The aryl hydrocarbon receptor (AHR) and AHR nuclear translocator (ARNT) are DNA binding transcription factors with basic helix-loop-helix/Per-Arnt-Sim (bHLH-PAS) domains. These two proteins form a heterodimer that mediates the toxic and biological effects of the environmental contaminant and AHR ligand 2,3,7,8-tetrachlorodibenzo-p-dioxin. The coiled-coil protein coiled-coil coactivator (Co-CoA) is a secondary coactivator for nuclear receptors and enhances nuclear receptor function by interacting with the bHLH-PAS domain of p160 coactivators. We report here that CoCoA also binds the bHLH-PAS domains of AHR and ARNT and functions as a potent primary coactivator for them; i.e. CoCoA does not require p160 coactivators for binding to and serving as a coactivator for AHR and ARNT. Endogenous CoCoA was recruited to a natural AHR target gene promoter in a 2,3,7,8-tetrachlorodibenzo-p-dioxin -dependent manner. Moreover, reduction of CoCoA mRNA levels by small interfering RNA inhibited the transcriptional activation by AHR and ARNT. Our data support a physiological role for CoCoA as a transcriptional coactivator in AHR/ARNT-mediated transcription.

Polycyclic or halogenated aromatic hydrocarbons and other related planar organochlorinated compounds elicit many adverse biological effects, including immunosuppression, teratogenesis, tumor promotion, hormonal disregulation, and cardiovascular disease. All of these biological effects are believed to be mediated by the sustained activation of the aryl hydrocarbon receptor (AHR) 1 (1,2). AHR is a ligand-dependent transcription factor belonging to the basic helix-loop-helix/Per-Arnt-Sim (bHLH-PAS) gene family (3). bHLH-PAS proteins share a conserved N-terminal structural motif. The bHLH domain of most (but not all) bHLH-PAS proteins is used for specific DNA binding and/or heterodimerization with other bHLH-PAS proteins. The PAS domain located immediately after the bHLH domain harbors two conserved hydrophobic repeats termed A and B and functions as a sensor of specific environmental signals, a ligand binding surface, and a proteinprotein interaction surface in various bHLH-PAS transcription factors, such as AHR and hypoxia-inducible factors. Another member of this family, AHR nuclear translocator (ARNT) is an indispensable heterodimer partner for AHR and hypoxia-inducible factors. In addition to the heterodimerization with AHR or hypoxia-inducible factor 1␣, ARNT homodimers are likely to play physiological roles by binding to the E-box core sequence found in some types of enhancer elements (4).
AHR is found in the cytosol in a heterotetrameric complex with two Hsp90 molecules and one X-associated protein 2 (XAP2) molecule (5,6). Upon binding of xenobiotic compounds such as TCDD, the complexes translocate to the nucleus, where the AHR heterodimerizes with ARNT after Hsp90 dissociation (2,6). Heterodimeric AHR⅐ARNT complexes bind to specific enhancer elements called xenobiotic response elements (XREs) found in the regulatory domains of numerous genes (7). Genes transcriptionally activated by AHR⅐ARNT encode several enzymes that metabolize xenobiotic compounds (e.g. cytochrome P-450 enzymes such as CYP1A1) and proto-oncogenes c-jun and c-fos (8,9). AHR is also apparently involved in hepatic growth and development of the immune system, based on the phenotype of AHR knock-out mice, and may play a role in the cell cycle (10,11).
Despite the structural and functional differences between AHR⅐ARNT and nuclear receptors (NRs), both are ligand-regulated transcription factors and share common coregulators to mediate their transcription-enhancing activities. For example, the NR corepressor SMRT binds to AHR⅐ARNT and inhibits AHR⅐ARNT activity (12). The AHR⅐ARNT heterodimer also interacts with NR coactivators such as the p160 coactivators SRC-1, GRIP-1, and p/CIP (13,14), which enhance AHR/ ARNT-mediated transcription in transient reporter gene assays. Although they have not been shown to bind DNA, the p160 coactivators also contain N-terminal bHLH-PAS domains (3,15). Furthermore, p160 coactivators are recruited to the CYP1A1 enhancer region in a TCDD-dependent fashion (13).
We previously isolated a novel NR coactivator, CoCoA, using the bHLH-PAS domain of GRIP1 as bait in a yeast two-hybrid screen (16,17). CoCoA is a new type of NR coactivator with a potent C-terminal activation domain and a central coiled-coil domain that binds the bHLH-PAS motif of p160 coactivators. Thus, CoCoA acts as a secondary coactivator for NRs, interacting with them indirectly through the primary p160 coactivators. Given the shared bHLH-PAS domains among p160 coactivators and bHLH-PAS transcription factors, we tested whether CoCoA could physically and functionally interact with AHR and ARNT as a coactivator in TCDD-dependent gene activation.
Coimmunoprecipitation and Immunoblotting-COS-7 cell transfection, coimmunoprecipitation, and immunoblotting were performed as described previously (17). For coimmunoprecipitation, cell lysate containing 1 mg of protein was incubated with 1 g of the anti-V5 antibody R960 -25 (Invitrogen), mouse normal IgG (Santa Cruz Biotechnology), 10 l of an equal mixture of the two rabbit antisera against CoCoA (17) or preimmune sera, and 20 l of protein A/G-agarose suspension (Santa Cruz Biotechnology) overnight at 4°C. Immunoblots were then performed on the precipitated proteins with anti-HA antibody 3F10 (Roche Applied Science). Subsequently, the membrane was stripped of bound antibody and re-probed with the anti-V5 antibody.
GST Pull-down Assay-HA epitope-tagged AHR, ARNT, and their fragments were synthesized in vitro by using TNT-Quick coupled transcription/translation system (Promega) according to the manufacturer's protocol. GST pull-down assays were performed as described previously (17). Bound proteins were analyzed by immunoblot with anti-HA antibody.
Cell Culture and Transient Transfection-Hepa1c1c7 cells, hereafter referred to as Hepa-1 cells, were maintained in ␣ minimal essential medium (Irvine Scientific) supplemented with 10% fetal bovine serum. HEK 293T cells and COS-7 cells were grown in Dulbecco's modified Eagle's medium with 10% fetal bovine serum. For reporter gene assays, HEK 293T and Hepa-1 cells were plated at 2 ϫ 10 5 cells/well in 12-well plates. HEK 293T cells were transiently transfected using Lipofectamine 2000 (Invitrogen). 48 h after transfection, cell extracts were prepared and assayed for luciferase activity as described previously (17). Hepa-1 cells were transiently transfected by Superfect transfection reagent (Qiagen) according to the manufacturer's protocol. The total amount of plasmid DNA added to each well was adjusted to 1.0 g by adding the necessary amount of pSG5.HA empty vector. Three hours after transfection cells were treated with either 10 nM TCDD or 0.1% dimethyl sulfoxide (Me 2 SO) vehicle and incubated for a further 18 h before luciferase assays. The results shown are the means and S.D. of triplicate points.

RESULTS
CoCoA Interacts with AHR and ARNT-Because the bHLH-PAS domain of the p160 coactivators is also present in AHR, ARNT, and other bHLH-PAS transcription factors, we tested whether CoCoA has the ability to interact with AHR and ARNT. In an in vitro GST pull-down assay GST-CoCoA efficiently bound full-length AHR and ARNT synthesized in vitro (Fig. 1, A and B). The interaction with AHR was TCDD-independent. To study in vivo interactions by co-immunoprecipitation, COS-7 cells were transfected with expression plasmids for V5-tagged CoCoA and HA-tagged AHR or ARNT. Anti-V5 antibodies efficiently and specifically precipitated AHR and ARNT (Fig. 1, C and D). Again, the interaction between CoCoA and AHR was TCDD-independent (data not shown). Other AHR coactivators can also bind efficiently to AHR without its FIG. 1. CoCoA interacts with AHR and ARNT in vitro and in vivo. A, HAtagged AHR protein synthesized in vitro was incubated with equal amounts of GST or GST-CoCoA fusion protein bound to glutathione-Sepharose beads in the absence or presence of 10 nM TCDD. Bound proteins were analyzed by 10% SDS-PAGE and detected by immunoblot analysis with anti-HA antibody. A portion of in vitro translated protein was loaded directly (10% input). B, GST pull-down assays using in vitro translated HA-tagged ARNT were performed as in A without TCDD. C, co-immunoprecipitation of V5-CoCoA and HA-AHR was performed with COS-7 cells in 100-mm dishes transfected with 2.5 g each of pcDNA3.1-CoCoA.V5 and pAC-TAG-2.mAHR. After 48 h cell extracts were immunoprecipitated with anti-V5 antibody or normal mouse IgG, and immunoblot (IB) analysis was performed with the indicated antibodies on the immunoprecipitates and a portion of the original cell extract (5% input). D, co-immunoprecipitation assays of V5-CoCoA and HA-ARNT were performed as described in C.
ligand (13, 18 -20). This is in contrast to the ligand-dependent interactions of NRs with many coactivators.
To characterize the CoCoA interaction domains in AHR and ARNT, the N-terminal (AHR 1-374 and ARNT 1-458) and C-terminal (AHR 375-805 and ARNT 459 -799) portions ( Fig.  2A) were tested for binding to CoCoA. In GST pull-down assays full-length AHR and its N-terminal and C-terminal domains interacted with CoCoA (Fig. 2B). The N-terminal bHLH-PAS domain of AHR (which contains the ligand binding function) interacted with CoCoA in the absence and the presence of TCDD (data not shown). Because ARNT fragments were not generated efficiently by in vitro translation, ARNT deletion mutants were tested in co-immunoprecipitation assays. CoCoA interacted with the N-terminal bHLH-PAS region but not with the C-terminal activation domain of ARNT (Fig. 2C).
CoCoA Potentiates Dioxin-dependent AHR/ARNT-mediated Transcription-CoCoA functions as a secondary coactivator to enhance transactivation by NRs; i.e. CoCoA binds to NRs indirectly through p160 coactivators, and the ability of CoCoA to enhance NR function depends on the presence of p160 coactivators (17). In contrast, CoCoA appears to interact directly with AHR and ARNT ( Figs. 1 and 2). To investigate whether CoCoA can function as a primary coactivator for AHR⅐ARNT, we transfected AHR/ARNT-positive Hepa-1 cells with a luciferase reporter plasmid controlled by the XRE-containing CYP1A1 promoter together with expression vectors for GRIP1, CoCoA, or both. TCDD treatment alone enhanced CYP1A1 promoter-driven luciferase activity nearly 28-fold (Fig. 3A, assay 1). As expected (13), overexpression of GRIP1 enhanced AHR function in a dose-dependent and TCDD-dependent manner (assays 2-5). CoCoA expression increased TCDD-dependent reporter gene activity up to 5-fold (assays 6 -9). This stimulation of AHR⅐ARNT activity was dependent on the amount of CoCoA, and the magnitude of stimulation was similar to that produced by GRIP1 under the same conditions. Despite the fact that CoCoA binds to AHR in a ligand-independent manner (Fig. 1A), the effect of CoCoA on AHR-mediated transcription was TCDD-dependent. When CoCoA was co-expressed with GRIP1, an additive induction of AHR/ARNT-mediated transactivation was observed (assays 10 -11). These results suggest that GRIP1 and CoCoA both interact directly with AHR⅐ARNT, function as primary coactivators, and make distinct contributions to AHR/ARNT-mediated transcriptional activation.
Although p160 coactivators contain an N-terminal bHLH-PAS domain, they do not utilize this domain to interact with AHR and ARNT; instead a C-terminal region of p160 proteins, just proximal to the p300/CBP binding AD1 domain, binds to the bHLH-PAS domains of AHR and ARNT (13). Therefore, we tested whether a GRIP1⌬N mutant lacking the bHLH-PAS domain could enhance AHR⅐ARNT function and cooperate with CoCoA. The GRIP1⌬N mutant enhanced AHR⅐ARNT function in a dose-dependent manner (Fig. 3B, assays 4 and data not shown). Furthermore, coexpression of CoCoA with the GRIP1⌬N mutant further potentiated the activity of AHR (Fig.  3B, assays 8 -9). Thus, the bHLH-PAS domain of GRIP1 is not required for the coactivator function of CoCoA with AHR. These data are consistent with the conclusion that CoCoA interacts directly as a primary coactivator with AHR and ARNT. This result contrasts starkly with the secondary coactivator function of CoCoA with NRs, which depends entirely on the presence of a p160 coactivator with an intact N-terminal region (17).
The C-terminal segment of CoCoA (amino acids 501-691) possesses very strong autonomous transcriptional activation activity when fused to the GAL4 DNA binding domain and is necessary for the coactivator function of CoCoA with NRs (17). The CoCoA mutant lacking the C-terminal activation domain failed to enhance AHR function (Fig. 3B, assay 5).
CoCoA Enhances the Transcriptional Activity of ARNT-To further investigate a functional interaction between ARNT and CoCoA, we tested the effect of CoCoA expression on ARNTmediated transactivation in a mammalian one-hybrid system. In transiently transfected HEK 293T cells the strong autonomous transactivation activity of a Gal4 DBD-ARNT fusion protein was further enhanced (up to 4-fold) by co-expression of CoCoA (Fig. 4, assays 4 -6). Because the bHLH-PAS domain of ARNT mediates protein interaction with CoCoA (Fig. 2C), we tested the effect of CoCoA overexpression on the activity of Gal4-ARNT-N, which harbors the bHLH-PAS motif. Gal4-ARNT-N possesses very little autonomous transcriptional activation activity on a transiently transfected Gal4-responsive reporter plasmid in HEK 293T cells (Fig. 4, assays 7 and 9),

FIG. 2. The bHLH-PAS domains of AHR and ARNT and the activation domain of AHR interact with CoCoA.
A, domain structure of the bHLH-PAS proteins. B, HA-tagged AHR, AHR N (amino acids 1-374), and AHR C (375-805) were translated in vitro and incubated with GST or GST-CoCoA fusion protein bound to beads. Bound proteins were analyzed by immunoblot analysis with anti-HA antibody. C, co-immunoprecipitation of V5-tagged CoCoA and HAtagged ARNT fragments was performed with COS-7 cells in 100-mm dishes transfected with 2.5 g each of pcDNA-3.1-CoCoA.V5 and pSG5.HA-ARNT N (1-458) or pSG5.HA-ARNT C (459 -799). After 48 h cell extracts were immunoprecipitated with antisera against CoCoA or control preimmune sera, and immunoprecipitated V5-CoCoA (panel ii) and co-precipitated HA-ARNT fragments (panel i) were detected by immunoblot (IB) analysis with antibodies against the epitope tags. A portion of the original cell extract was also analyzed (5% input). consistent with studies indicating the requirement for the ARNT C-terminal activation domain for fully activated transcription (21). Nevertheless, co-expression of CoCoA increased luciferase activity by up to 7-fold (assays 9 -11). Thus, CoCoA interacts with the bHLH-PAS domain of ARNT and is a potent coactivator for ARNT.
CoCoA Is Specifically Targeted to the CYP1A1 XRE Region-ChIP assays were used to test whether CoCoA is recruited to the XREs of known AHR target genes. The CYP1A1 gene from mouse Hepa-1 cells has a well characterized regulatory region that responds to TCDD. Hepa-1 cells were treated either with 10 nM TCDD in Me 2 SO or with Me 2 SO vehicle. The crosslinked, sheared chromatin preparations were subjected to im-munoprecipitation with various antibodies, and the precipitated DNA was analyzed by PCR amplification of the XRE region of the CYP1A1 promoter. TCDD treatment caused recruitment of AHR and GRIP1 to the CYP1A1 promoter region (Fig. 5), as shown previously (13,22). In addition, antibodies against CoCoA efficiently immunoprecipitated the XRE region of the CYP1A1 promoter from the TCDD-treated but not from the control cells, indicating the dioxin-dependent recruitment of CoCoA to the CYP1A1 promoter. Normal IgG failed to precipitate the CYP1A1 promoter. In contrast to the CYP1A1 promoter, the ␤-actin promoter region was not detected in association with AHR, GRIP1, or CoCoA. Thus, endogenous CoCoA was recruited to a native AHR-regulated promoter in a TCDD-dependent fashion, demonstrating a functional interaction between AHR and CoCoA occurring in an in vivo setting.

Requirement for Endogenous CoCoA for Transcriptional Activation by AHR and ARNT-If
CoCoA is a coactivator for AHR⅐ARNT, reducing the endogenous level of CoCoA should decrease the transcriptional activity of endogenous AHR and ARNT. In RNA interference experiments using a CoCoA-specific siRNA that was previously demonstrated to be effective in reducing CoCoA expression level (17), transfection of the Co-CoA-directed siRNA, but not a control scrambled-sequence siRNA, into Hepa-1 cells reduced the level of endogenous Co-CoA mRNA but had no effect on the ␤-actin mRNA level (Fig.  6A). TCDD-induced expression of the endogenous CYP1A1 gene was inhibited (more than 40%) by the CoCoA-directed siRNA, but the scrambled-sequence siRNA had no effect.
To our surprise we also found that dioxin up-regulated the CoCoA mRNA level greater than 2-fold (Fig. 6A, lanes 1 and  2). Analysis of the mouse CoCoA gene (LocusLink locus ID 67488) for core AHR/ARNT binding sequences using the webbased TRANSFAC data base (23) revealed one such sequence (tgcCACGCtgagtcca) and two putative ARNT binding sequences (CACGTG) within 2 kilobases upstream of the Co-CoA transcription start site. Thus, the CoCoA gene is apparently dioxin-responsive.
The physiological significance of CoCoA in ARNT function was further tested by RNA interference in COS-7 cells. The siRNA against CoCoA, but not the scrambled-sequence siRNA, specifically reduced the level of endogenous CoCoA mRNA but not ␤-actin mRNA (Fig. 6B, inset). Reduction of CoCoA levels in COS-7 cells also reduced activity of a Gal4 DBD-ARNT fusion protein by more than 65% but had little or no effect on the activity of Gal4 DBD (Fig. 6B) Thus, endogenous CoCoA is required for efficient transcriptional activation by AHR and ARNT. The results from ChIP and RNAi assays strongly support a physiological role for CoCoA in AHR/ARNT-dependent transcription.

DISCUSSION
CoCoA Is a Primary Coactivator for AHR⅐ARNT-CoCoA was initially identified as a novel type of NR coactivator with a coiled-coil domain (17). CoCoA binds to the bHLH-PAS domain of p160 coactivators but not directly to NRs, and the ability of CoCoA to enhance the activity of steroid receptors is highly dependent on the presence of a p160 coactivator with an intact N-terminal bHLH-PAS domain. By these characteristics, Co-CoA is classified as a secondary coactivator for nuclear receptors (16). In contrast, CoCoA bound to AHR and ARNT in vitro and in vivo and functioned as a coactivator for AHR and ARNT without co-expression of p160 coactivators or any other exogenously expressed proteins (Figs. 1 and 3). Thus, CoCoA binds indirectly to NRs and functions as a secondary coactivator for them but appears to bind directly to AHR⅐ARNT and serves as a primary coactivator for AHR⅐ARNT.
The transient transfection data indicate that CoCoA has the activity of a coactivator for AHR⅐ARNT under those conditions. Under more physiologically relevant conditions endogenous CoCoA was efficiently recruited together with AHR to the endogenous CYP1A1 promoter in a TCDD-dependent fashion (Fig. 5) and was required for efficient TCDD induction of the endogenous CYP1A1 gene by AHR and for the autonomous transcriptional activation activity of ARNT (Fig. 6). These findings demonstrate that CoCoA is a physiologically relevant part of transcriptional activation by AHR⅐ARNT.
Interestingly, we observed that the CoCoA mRNA level was up-regulated more than 2-fold by TCDD. Obviously, the resulting increase in CoCoA may further enhance AHR⅐ARNT activity. The regulation of CoCoA by ligand-activated AHR⅐ARNT is an addi-tional physiological link between CoCoA and AHR⅐ARNT. The mechanism by which AHR⅐ARNT regulates CoCoA levels may involve direct binding of AHR⅐ARNT to the CoCoA gene promoter, since a putative AHR⅐ARNT binding sequence occurs within the proximal CoCoA promoter. Thus CoCoA and AHR⅐ARNT mutually regulate each other; AHR⅐ARNT regulates CoCoA levels, and Co-CoA serves as a coactivator for AHR⅐ARNT.
Interaction of CoCoA with Mulitple Classes of bHLH-PAS Proteins-The interaction between CoCoA and AHR appears to be direct and involves two distinct regions within AHR, the N-terminal bHLH-PAS domain and a second site within the C-terminal activation domain (Fig. 2B). AHR bound CoCoA in a ligand-independent manner, suggesting that CoCoA may exist in a complex with AHR before ligand activation and after ligand binding travels with AHR to its target promoter. CoCoA and AHR are both recruited to the CYP1A1 promoter in a TCDD-dependent manner (Fig. 5), and CoCoA fails to enhance the activity of the CYP1A1 promoter in the absence of TCDD (Fig. 3). The ligand-independent association of AHR with Co-CoA and other coactivators (13, 18 -20) is in contrast to the interactions of many coactivators with NRs and may result from different structural organizations of AHR and NR proteins. Unlike the steroid receptors, where the ligand binding pocket and the important AF-2 activation function are part of the same structural domain, the major C-terminal activation domain of AHR is well separated from the N-terminal ligand binding domain; this may allow ligand-independent interactions between activation domain and coactivators.
In contrast to the situation with AHR, CoCoA interacts with the bHLH-PAS domain but not with the activation domain of ARNT (Fig. 2C). These observations are consistent with previous reports which indicate that deletion of the C-terminal activation domain of ARNT failed to have a significant effect on activation of a reporter gene under control of the CYP1A1 enhancer and promoter, leading to the conclusion that the activation domain of AHR is dominant over that of ARNT during transcriptional activation (24,25).
In general, the bHLH-PAS superfamily can be divided into three subgroups; members of one group are regulated in their expression or activity by binding of ligands or by cellular con- FIG. 5. Endogenous CoCoA is recruited to the native CYP1A1 promoter in a dioxin-dependent manner. Hepa-1 cells were untreated or treated with 10 nM TCDD for 60 min. For ChIP assays sheared chromatin was immunoprecipitated with anti-AHR, anti-GRIP1, anti-CoCoA, or rabbit normal IgG. The co-precipitated DNA was amplified by PCR using primers to amplify the CYP1A1 promoter or the ␤-actin promoter. PCR was performed with a serial dilution of input (1, 1:10, 1:50, 1:100, and 1:500) and precipitated DNA samples (5, 1, 1:10, 1:50, and 1:100) to determine the linear range for the amplification (data not shown); data shown were obtained with 1 or 5 l of 1:10-diluted samples (input) and 1 or 5 l of the undiluted samples ditions (AHR, hypoxia-inducible factor, Single-minded (SIM), and Trachealess (TRH)); the second subgroup is ARNT, which does not need activation and has a central role in dimerization with members of group 1; the third subgroup is the p160 coactivators, which are not known to bind DNA directly but rather function as coactivators for diverse types of DNA binding transcription factors. CoCoA interacts with members of all three subgroups and enhances their activities as transcription factors or coactivators. These findings suggest that CoCoA may function as a general coactivator for all the bHLH-PAS transcriptional activators and may help us to understand the transactivation mechanisms of the bHLH-PAS protein family. Future studies of the interaction between CoCoA and the bHLH-PAS proteins may generate insights into regulation of genes involved in responses to hypoxia, circadian rhythms, development, and other pathways.
Implications of CoCoA as a Common Coactivator for NRs and for AHR⅐ARNT-The fact that CoCoA serves as a common coactivator for two different classes of DNA binding transcription factors suggests a novel avenue for cross-talk between these two signaling pathways. It is interesting to note that TCDD inhibits several estrogen-dependent biological responses, decreases cellular estrogen receptor (ER) and progesterone receptor levels (26), and inhibits transcription of estrogen-regulated genes (27). Conversely, 17␤-estradiol (E2) can inhibit TCDD-induced reporter gene activity (28). The possible mechanisms proposed for inhibitory AHR-ER cross-talk include induction of inhibitory factors, enhanced E2 metabolism, proteasome-dependent degradation of ER␣ by non-genomic inhibitory effects of AHR, and competition for common nuclear coactivators (29). Recent studies report that TCDD-induced interactions between AHR and ER␣ enhance ubiquitylation of ER␣ and subsequent proteasome-dependent degradation of ER␣ (30).
NRs and AHR⅐ARNT have been reported to interact with many common coactivators and components of the transcription machinery, including CoCoA, p160 coactivators (13,14), p300/CBP (31), BRG-1 (22), ERAP140 (12), RIP140 (19), basal transcription factors (32), and mediator complex (33). In addition, a recent study has shown that ARNT, the obligatory heterodimerization partner for AHR, also functions as a coactivator of ER␣ and ER␤ (34). Some transcription factors have been reported to compete with one another for binding with limiting coactivators, resulting in mutually antagonistic effects (35,36). Competition for binding and squelching of these limiting common coactivators could be one of the mechanisms for the anti-estrogenic activity of TCDD. Our results add an additional possibility that cellular availability of CoCoA may also be an important factor in regulating ER and AHR transcriptional activities.