HDAC6 Modulates Hsp90 Chaperone Activity and Regulates Activation of Aryl Hydrocarbon Receptor Signaling*

The aryl hydrocarbon receptor (AhR), a ligand-activated member of the basic helix-loop-helix family of transcription factors, binds with high affinity to polycyclic aromatic hydrocarbons (PAH) and the environmental toxin 2,3,7,8-tetrachlorodibenzo-p-dioxin (dioxin). Most of the biochemical, biological, and toxicological responses caused by exposure to PAHs and polychlorinated dioxins are mediated, at least in part, by the AhR. The AhR is a client protein of Hsp90, a molecular chaperone that can be reversibly acetylated with functional consequences. The main objective of this study was to determine whether modulating Hsp90 acetylation would affect ligand-mediated activation of AhR signaling. Trichostatin A and suberoylanilide hydroxamic acid, two broad spectrum HDAC inhibitors, blocked PAH and dioxin-mediated induction of CYP1A1 and CYP1B1 in cell lines derived from the human aerodigestive tract. Silencing HDAC6 or treatment with tubacin, a pharmacological inhibitor of HDAC6, also suppressed the induction of CYP1A1 and CYP1B1. Inhibiting HDAC6 led to hyperacetylation of Hsp90 and loss of complex formation with AhR, cochaperone p23, and XAP-2. Inactivation or silencing of HDAC6 also led to reduced binding of ligand to the AhR and decreased translocation of the AhR from cytosol to nucleus in response to ligand. Ligand-induced recruitment of the AhR to the CYP1A1 and CYP1B1 promoters was inhibited when HDAC6 was inactivated. Mutation analysis of Hsp90 Lys294 shows that its acetylation status is a strong determinant of interactions with AhR and p23 in addition to ligand-mediated activation of AhR signaling. Collectively, these results show that HDAC6 activity regulates the acetylation of Hsp90, the ability of Hsp90 to chaperone the AhR, and the expression of AhR-dependent genes. Given the established link between activation of AhR signaling and xenobiotic metabolism, inhibitors of HDAC6 may alter drug or carcinogen metabolism.

The aryl hydrocarbon receptor (AhR), 2 a ligand-activated member of the basic helix-loop-helix family of transcription factors (1), binds with high affinity to the environmental toxin 2,3,7,8-tetrachlorodibenzeno-p-dioxin (TCDD or dioxin) (2). Polycyclic aromatic hydrocarbons (PAH) in tobacco smoke, automobile exhaust, coal tar, and charbroiled food also bind to and activate the AhR. The AhR mediates most of the biological and toxicological responses caused by exposure to PAHs and dioxins (3). Recent studies in genetically engineered mice suggest a role for the AhR in carcinogenesis (4 -6). In transgenic mice, a constitutively active AhR induced tumors in the glandular stomach (4) and rendered mice more susceptible to carcinogens (5). Furthermore, mice engineered to be AhR-deficient were protected against benzo[a]pyrene (B[a]P)/PAH-induced skin tumors (6).
In the absence of ligand, the AhR is present in the cytosol as a component of a complex with a dimer of the chaperone heat shock protein (Hsp90) (7,8), the cochaperone p23 (9,10), and XAP-2 (11). The association with Hsp90 is required for the AhR to assume a conformation that is optimal for ligand binding (12). Upon ligand binding, the AhR translocates from the cytosol into the nucleus and forms a heterodimer with the AhR nuclear transporter (13,14). The AhR/AhR nuclear transporter heterodimer binds to the upstream regulatory region of genes containing xenobiotic-responsive elements (XRE) (14), resulting in the transcriptional activation of a network of genes encoding proteins including CYP1A1 and CYP1B1 (14 -17). Activation of AhR-mediated signaling leading to induction of xenobiotic metabolizing enzymes can both increase the production of highly carcinogenic metabolites (18 -20) and alter the metabolism of drugs (21,22). The connection between AhR signaling and carcinogenesis has led to a search for agents that antagonize the actions of the AhR (23,24).
Reversible acetylation has been implicated as a functionally important posttranslational modification of Hsp90 (25). Treatment of cells with a histone deacetylase (HDAC) inhibitor causes acetylation of Hsp90 and depletion of several Hsp90 client proteins (26). Recently, HDAC6, an HDAC found in the cytoplasm, was identified to be an Hsp90 deacetylase (27,28). Inactivation of HDAC6 causes Hsp90 hyperacetylation and subsequent dissociation from the cochaperone, p23, with a loss of chaperone activity (27). In HDAC6-deficient cells, Hsp90dependent maturation of the glucocorticoid receptor (GR) was inhibited, resulting in functionally defective GR with reduced This article has been withdrawn by the authors. The actin immunoblots in ability to both bind ligand and activate transcription (27,29). An effort has been made to identify a functionally important Hsp90 acetylation site. Acetylation/deacetylation of Hsp90 Lys 294 was found to play an important role in regulating the Hsp90 chaperone cycle (30).
In the current study, our primary objective was to determine whether modulating Hsp90 acetylation would affect ligandmediated activation of AhR signaling. We demonstrate for the first time that HDAC6 activity is important for ligand-mediated activation of AhR signaling. Silencing or inhibition of HDAC6 resulted in an immature form of AhR with a reduced capacity to bind PAH, translocate to the nucleus, and activate the transcription of CYP1A1 and CYP1B1. Acetylation of the Lys 294 residue of Hsp90 was important for regulating the activation of AhR signaling. Taken together, these findings suggest that inhibitors of HDAC6 will suppress the activation of AhR-dependent genes, which could in turn impact on both chemical carcinogenesis and drug metabolism.

EXPERIMENTAL PROCEDURES
Reagents-TSA and SAHA were from Biomol (Plymouth Meeting, PA). Antibodies to Hsp90, AhR, XAP-2, CYP1A1, and FLAG were from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA). B[a]P, protein quantitation assay kits, horseradish peroxidase-conjugated secondary antibody, and antibodies to acetylated tubulin, nucleobindin, and ␤-actin were from Sigma. Antibodies to acetylated lysine and HDAC6 were from Cell Signaling Technology (Beverly, MA). Nonspecific and HDAC6specific siRNAs were from Dharmacon RNA Technologies (Lafayette, CO). Western blotting detection reagents (ECL) were from Amersham Biosciences. Reagents for the luciferase assay were from Analytical Luminescence (San Diego, CA  (34), and MSK-Leuk1 (oral leukoplakia) cells (35) were grown as previously described. A549 cells stably expressing pSuper control plasmid or HDAC6 siRNA (knockdown (KD)) were a kind gift from Dr. Tso-Pang Yao of Duke University (Durham, NC) (27). In all of the experiments, the cells were incubated in serum-free medium for 24 h before treatment. The treatments were carried out in serum-free medium.
Tobacco Smoke Preparation-Cigarettes (2R4F; Kentucky Tobacco Research Institute) were smoked in a Borgwaldt piston-controlled apparatus (model RG-1) using the Federal Trade Commission standard protocol. Cigarettes were smoked one at a time in the apparatus, and the smoke was drawn under sterile conditions into premeasured amounts of sterile phosphate-buffered saline, pH 7.4, representing whole, trapped mainstream smoke (TS). Quantitation of smoke content is expressed in puffs/ml of phosphate-buffered saline with 1 cigarette yielding about 8 puffs drawn into a 5 ml volume. The final concentration of TS in the cell culture medium is expressed as puffs/ml medium. All of the treatments were carried out with 0.03 puffs/ml because this concentration has been used to activate AhR signaling in previous studies (16,36,37).
Northern Blotting-Total RNA was prepared from cell monolayers using an RNA isolation kit from Qiagen. Ten micrograms of total RNA/lane were electrophoresed in a formaldehyde-containing 1% agarose gel and transferred to nylonsupported membranes. CYP1A1, CYP1B1, and 18 S rRNA probes were labeled with [ 32 P]CTP by random priming. The blots were then probed as previously described (16). All of the experiments were repeated three times, and representative Northern blot data are shown.
Transient Transfections-The cells were seeded at a density of 5 ϫ 10 4 /well in 6-well dishes and grown to 50 -60% confluence. For each well, 2 g of plasmid DNA were introduced into cells using 8 g of Lipofectamine as per the manufacturer's instructions. After 24 h of incubation, the medium was replaced with serum-free medium. The activities of luciferase and ␤-galactosidase were measured in cellular extracts. FLAG-Hsp90 constructs were transfected into A549 cells using Amaxa nucleofection technology following the manufacturer's instructions.
RNA Interference-The cells were seeded in KGM for 24 h before transfection. One hundred pmol/ml of siRNA to HDAC6 or nonspecific siRNA were transfected using Dharma-FECT 4 transfection reagent according to the manufacturer's instructions.
Immunoprecipitation-This was performed with a kit from Upstate Biotechnology, Inc. (Lake Placid, NY) according to the manufacturer's instructions. Two hundred fifty g of cell lysate protein were used for immunoprecipitation at room temperature. The immunoprecipitates were then analyzed by SDS-PAGE.
Western Blotting-The lysates were prepared by treating cells with lysis buffer (150 mM NaCl, 100 mM Tris, pH 8.0, 1% Tween 20, 50 mM diethyldithiocarbamate, 1 mM EDTA, 1 mM phenylmethylsulfonyl fluoride, 10 g/ml aprotinin, 10 g/ml trypsin inhibitor, and 10 g/ml leupeptin). The lysates were sonicated for 20 s on ice and centrifuged at 10,000 ϫ g for 10 min to sediment the particulate material. The protein concentration of the supernatant was measured by the method of Lowry et al. (38). SDS-PAGE was performed under reducing conditions as described by Laemmli (39). The resolved proteins were transferred onto nitrocellulose sheets as detailed by Towbin et al. (40), and the membrane was incubated with primary antisera.
Secondary antibody to IgG conjugated to horseradish peroxidase was used. The blots were probed with the ECL Western blot detection system.
Ligand Binding Assay-The cells were lysed in 1.5 volumes of buffer (10 mM HEPES, pH 7.3, 1 mM EDTA, and 20 mM sodium molybdate) and centrifuged at 100,000 ϫ g. Aliquots of cytosol were incubated overnight at 4°C with 100 nM [  Chromatin Immunoprecipitation (ChIP) Assay-ChiP assay was performed with a kit (Upstate Biotechnology) according to the manufacturer's instructions. The cells (2 ϫ 10 6 ) were crosslinked in a 1% formaldehyde solution for 10 min at 37°C. The cells were then lysed in 200 l of SDS buffer and sonicated to generate 200 -1000-bp DNA fragments. After centrifugation, the cleared supernatant was diluted 10-fold with ChIP buffer and incubated with 1.5 g of the indicated antibody at 4°C. Immune complexes were precipitated, washed, and eluted as recommended. DNA-protein cross-links were reversed by heating at 65°C for 4 h, and the DNA fragments were purified and dissolved in 50 l of water. Ten microliters of each sample were used as a template for PCR amplification. The forward and reverse primers used for amplifying the CYP1A1 promoter are: 5Ј-ACCCGCCACCCTTCGACAGTTC-3Ј and 5Ј-TGC-CCAGGCGTTGCGTGAGAAG-3Ј (41). Forward and reverse primers used to amplify the CYP1B1 promoter are: 5Ј-GTT-CCCTTATAAAGGGAG-3Ј and 5Ј-CTGCGATGGAAGC-CGTTG-3Ј (42). PCR was performed at 94°C for 30 s, 60°C for 30 s, and 72°C for 45 s for 30 cycles. The PCR products generated from the ChIP template were sequenced, and the identity of the CYP1A1 and CYP1B1 promoters was confirmed.
Statistics-Comparisons between groups were made by Student's t test. A difference between groups of p Ͻ 0.05 was considered significant.

HDAC Inhibitors Suppress Ligand-mediated Induction of
AhR-dependent Gene Expression-Western blot analysis was performed to examine the effect of TSA and SAHA, structurally related broad spectrum HDAC inhibitors, on TS-mediated induction of CYP1A1 and CYP1B1. Both HDAC inhibitors caused dose-dependent inhibition of TS-mediated induction of CYP1A1 (Fig. 1A) and CYP1B1 (Fig. 1B) in MSK-Leuk1 cells, a human cell line derived from oral leukoplakia. Suppressive effects were also observed in several other cell lines derived from the human aerodigestive tract (Fig. 1). CYP1B1 but not CYP1A1 was readily detected in TS-treated A549 cells. Next we evaluated whether HDAC inhibitors could suppress the inductive effect of B[a]P, a prototypic AhR ligand. As shown in Fig. 2, both TSA and SAHA suppressed B[a]P-mediated induction of CYP1A1 and CYP1B1 in KYSE450 ( Fig. 2A) and MSK-Leuk1 cells (Fig.  2B). To further elucidate the mechanism underlying these effects, Northern blotting was carried out. TSA and SAHA suppressed TS-mediated induction of CYP1A1 and CYP1B1 mRNAs (Fig. 3). The HDAC inhibitors also blocked B[a]P-and TCDD-mediated induction of CYP1A1 and CYP1B1 mRNAs (data not shown). HDAC6 Is Important for Ligandmediated Induction of AhR-dependent Gene Expression-Next, experiments were conducted to identify  the HDAC that regulates ligand-mediated activation of AhR signaling. Previously, Kovacs et al. (27) reported that Hsp90 is a substrate of HDAC6 and that its chaperone activity is regulated by acetylation. Because the AhR is a client protein of Hsp90, we hypothesized that HDAC6 could be important for ligand-mediated induction of CYP1A1 and CYP1B1. To test this possibility, a series of experiments were carried out in which HDAC6 was silenced. Initially, A549 cells in which HDAC6 was stably knocked down were used (Fig. 4A). Treatment with TS (Fig.  4B), TCDD (Fig. 4C), or B[a]P (data not shown) induced CYP1B1 in vector-expressing cells but not in A549 HDAC6 knockdown (KD) cells. Silencing of HDAC6 also suppressed TS-mediated induction of XRE-luciferase activity in these cells (Fig. 4D). Consistent with the findings in A549 cells, silencing of HDAC6 (Fig. 4E) also suppressed the induction of CYP1A1 and CYP1B1 by TS (Fig. 4F) and B[a]P (data not shown) in MSK-Leuk1 cells. To complement this genetic strategy, a pharmacological approach was utilized to confirm the role of HDAC6 in ligand-mediated activation of AhR signaling. HDAC6 possesses tubulin deacetylase activity. Tubacin, a targeted inhibitor of HDAC6 (43), caused dose-dependent increases in tubulin acetylation consistent with its ability to inhibit HDAC6 (Fig. 5A).
Tubacin also suppressed TS-and TCDD-mediated induction of CYP1A1 and CYP1B1 and blocked TS-mediated stimulation of XREluciferase activity in MSK-Leuk1 cells (Fig. 5, B-D). Additionally, tubacin caused dose-dependent suppression of both TS-and TCDD-mediated induction of CYP1B1 in A549 cells (Fig. 5, E and  F). Collectively, these data imply   MARCH 20, 2009 • VOLUME 284 • NUMBER 12 that HDAC6 is important for regulating the ability of AhR ligands to activate the expression of target genes. We next investigated the mechanism by which HDAC inhibitors suppress ligand-mediated activation of AhR signaling. Previously, HDAC6 was found to interact with Hsp90 and regulate its acetylation (27). Immunoprecipitation of HDAC6 coprecipitated Hsp90 in both MSK-Leuk1 and A549 cells (Fig.  6A). Silencing HDAC6 or treatment with TSA, SAHA, or tubacin prevented the coprecipitation of HDAC6 with Hsp90 (Fig.  6, B-D). Levels of Hsp90, AhR, p23, and XAP-2 were unaffected by these treatments (supplemental Fig. S1, A-D). Because HDAC6 is a Hsp90 deacetylase (27), we next determined whether HDAC inhibition caused an increase in Hsp90 acety-lation. Treatment with a broad spectrum HDAC inhibitor (TSA or SAHA), tubacin, or silencing of HDAC6 stimulated the acetylation of Hsp90 (Fig. 6, E-G, and supplemental Fig. S2). In cells in which HDAC6 was silenced, further treatment with TSA or SAHA (data not shown) did not lead to an additional enhancement of Hsp90 acetylation (Fig. 6, F and G). These results suggest that HDAC6 is the dominant TSA/SAHA-sensitive Hsp90 deacetylase.

HDAC6 Regulates Hsp90 Chaperone Activity and AhR Signaling
Next, we investigated whether hyperacetylation of Hsp90 would alter its interaction with AhR, p23, and XAP-2. p23 and XAP-2 are normally found in a cytoplasmic complex with Hsp90 and AhR. As shown in Fig. 6, (H, MSK-Leuk1 cells, and I, A549 cells), immunoprecipitation of Hsp90 pulled down p23,  FIGURE 6. HDAC6 associates with Hsp90 and regulates its acetylation and chaperone complex formation with p23 and XAP-2. A, cell lysates (250 g) from MSK-Leuk1 and A549 cells were subjected to immunoprecipitation (IP) with antibody to HDAC6 or IgG. The immunoprecipitates were then subjected to immunoblotting (IB) and probed with antibodies to Hsp90 or HDAC6. B, MSK-Leuk1 cells were treated with vehicle (lanes C), TSA (500 nM), SAHA (20 M), or tubacin (20 M) for 1 h. The cell lysates were prepared and immunoprecipitated with an antibody to HDAC6. Immunoprecipitates were then subjected to immunoblotting and probed with antibodies to Hsp90 or HDAC6. C, cell lysates were prepared from A549 cells stably expressing pSUPER (Control) or HDAC6 siRNA (HDAC6 KD). D, cell lysates were prepared from MSK-Leuk1 cells transiently transfected with nonspecific (control) siRNA or HDAC6 siRNA. In C and D, cell lysates were subjected to immunoprecipitation with antibody to HDAC6. The immunoprecipitates were then subjected to immunoblotting and probed for Hsp90 and HDAC6. Additionally, cell lysates were directly subjected to immunoblotting for Hsp90. In E-G, MSK-Leuk1 and A549 cells were treated as indicated with vehicle (lanes C), TSA (500 nM), SAHA (20 M), or tubacin (20 M) for 1 h. The cell lysates were then subjected to immunoprecipitation with antibody to acetyl lysine (AcK). The immunoprecipitates were then subjected to immunoblotting and probed for Hsp90. Cell lysates were also directly subjected to immunoblotting for Hsp90. H, cell lysates were prepared from MSK-Leuk1 cells that received vehicle (control) or TSA (500 nM) for 1 h or were transfected with nonspecific siRNA (control siRNA) or HDAC6 siRNA. I, cell lysates were prepared from A549 cells that received vehicle or TSA (500 nM) for 1 h. In H and I, cell lysates were subjected to immunoprecipitation with antibody to Hsp90. The immunoprecipitates were then subjected to immunoblotting and probed for p23, AhR, HDAC6, and XAP-2. Additionally, cell lysates were directly subjected to immunoblotting for Hsp90.

HDAC6 Regulates Hsp90 Chaperone Activity and AhR Signaling
AhR, HDAC6, and XAP-2. This interaction was not observed in cells treated with an HDAC inhibitor or in which HDAC6 was silenced. Levels of Hsp90, AhR, p23, and XAP-2 were unaffected by treatment with TSA or silencing of HDAC6 (supplemental Fig. S1E). Hence, HDAC6 activity is necessary for Hsp90 complex formation, which in turn plays a significant role in regulating the induction of CYP1A1 and CYP1B1.
HDAC6 Is Important for AhR Ligand Binding, Translocation, and Transcriptional Activity-We next evaluated whether HDAC6-regulated Hsp90 acetylation was important for the chaperone function of Hsp90. Efficient binding of ligand and the subsequent translocation of activated AhR from cytosol to nucleus are known chaperone functions of Hsp90. Several experiments were conducted to determine whether HDAC6 was a determinant of ligand binding to the AhR, translocation of AhR from cytosol to nucleus, and activation of CYP1A1 and CYPB1 transcription. Silencing HDAC6 did not alter the levels of AhR or Hsp90 in A549 cells (Fig. 7A). The effect of silencing HDAC6 on the binding of ligand to the AhR was investigated. Cytosols prepared from A549 control and A549 HDAC6 KD cells were incubated with radiolabeled B[a]P, and ligand binding to the AhR was determined. The binding of [ 3 H]B[a]P to AhR was reduced in cytosols prepared from A549 HDAC6 KD cells compared with A549 control cells (Fig. 7A). Consistent with this finding, a dramatic decrease in the binding of [ 3 H]B[a]P was also found in cytosols prepared from A549 cells treated with TSA or SAHA (Fig. 7B). Subsequently, the same experiments were repeated in MSK-Leuk1 cells. Once again, silencing of HDAC6 did not lead to reduced amounts of AhR but did cause a significant decrease in the binding of [ 3 H]B[a]P to the AhR (Fig. 7C). Furthermore, reduced binding of B[a]P to the AhR was also found in MSK-Leuk1 cells treated with either TSA or SAHA (Fig. 7D). Although comparable amounts of AhR are present in the cytosols of cells in which HDAC6 was silenced, the binding of B[a]P was markedly reduced relative to control cells. These results suggest that the AhR produced in cells in which HDAC6 is silenced is defective in ligand binding. The chaperone function of Hsp90 is deficient in association with hyperacetylation of Hsp90.
Ligand-induced nuclear accumulation of AhR was also evaluated. In untreated MSK-Leuk1 cells, the AhR is located predominantly in the cytosol. Treatment with SAHA (Fig. 8, A and B) or silencing HDAC6 (data not shown) blocked TS-and B[a]P-mediated translocation of the AhR from cytosol to nucleus. ChIP assays were conducted to determine whether recruitment of AhR to CYP1A1 and CYP1B1 promoters was altered by inhibition of HDAC6. Treatment with SAHA (Fig. 8, C-E) or silencing HDAC6 (data not shown) blocked TS-, TCDD-, and B[a]P-mediated recruitment of the AhR to the CYP1A1 promoter. Similar effects were observed for the CYP1B1 promoter (data not shown). In A549 cells, TS-mediated recruitment of AhR to the CYP1B1 promoter was blocked by SAHA (Fig. 8F) or silencing of HDAC6 (Fig. 8G). Thus, Hsp90-dependent binding of ligand to the AhR, nuclear translocation, and stimulation of gene transcription are all suppressed when HDAC6 is inhibited.
Acetylation of Lys 294 Weakens the Interaction with AhR-The above findings indicate that Hsp90 acetylation inhibits its activity. Recently, the acetylation state of Hsp90 Lys 294 was found to be A, cytosols from A549 cells stably expressing pSUPER control plasmid (Control) or HDAC6 siRNA (HDAC6 KD) were subjected to ligand binding assay. Inset, cell lysates were subjected to immunoblotting and probed for HDAC6, AhR, Hsp90, and ␤-actin. B, cytosols were prepared from A549 cells that were treated with the indicated concentrations of TSA and SAHA for 1 h prior to ligand binding assay. C, cytosols were prepared from MSK-Leuk1 cells that were transiently transfected with control siRNA or HDAC6 siRNA prior to ligand binding assay. Inset, cell lysates were subjected to immunoblotting and probed as indicated. D, cytosols were prepared from MSK-Leuk1 cells that were treated with the indicated concentrations of TSA and SAHA for 1 h prior to ligand binding assay. The data are plotted as the means Ϯ S.D. (n ϭ 3). *, p Ͻ 0.01; **, p Ͻ 0.001. The data shown are representative of three independent experiments. critical for regulating its chaperone function (30). Hence, to determine whether the acetylation state of Lys 294 affects the interaction between Hsp90 and AhR, we transfected FLAG-tagged WT and mutant Hsp90 constructs (Fig. 9A) into A549 cells and immunoprecipitated FLAG-Hsp90 complexes. A series of Hsp90 mutants that mimic unacetylated (arginine, K294R) or acetylated (glutamine, K294Q; alanine, K294A) lysine were used (30). Endogenous AhR interacted with WT Hsp90 and a K294R mutant of Hsp90, but almost no interaction was detected when K294Q or K294A mutants of Hsp90 were overexpressed (Fig. 9B). Because Hsp90 functions in association with the cochaperone p23, we next determined whether the acetylation status of Lys 294 affected the interaction between Hsp90 and p23. An interaction between p23 and FLAG-Hsp90 WT and K294R was detected, but its interaction with K294Q or K294A was markedly reduced (Fig. 9C). Consistent with these findings, TS and B[a]P induced CYP1B1 in WT and K294R-expressing cells but not in K294Q-or K294A-expressing cells (Fig. 9, D and E). Thus, our data identify the acetylation of Lys 294 , a residue conserved in eukaryotic Hsp90, as a key regulator of Hsp90 function leading to modulation of AhR signaling.

DISCUSSION
In this study, we showed that HDAC6 activity is important for ligand-mediated activation of AhR signaling. Several lines of evidence support this point. First, TSA and SAHA, two broad spectrum HDAC inhibitors, suppressed both TS-and B[a]P-mediated induction of CYP1A1 and CYP1B1 in cell lines derived from the aerodigestive tract (Figs. 1-3). Moreover, silencing of HDAC6 suppressed the induction of CYP1A1 and CYP1B1 by ligands of the AhR (Fig. 4). Consistent with this finding, tubacin, a HDAC6 inhibitor, also inhibited AhR ligand-mediated induction of CYP1A1 and CYP1B1 (Fig. 5).
Previously, HDAC6 was found to function as an Hsp90 deacetylase (27,28). The known chaperone functions of Hsp90 include maintaining the AhR in a high affinity ligand binding conformation (12,44). Hence, we next investigated whether inhibiting HDAC6 caused hyperacetylation of Hsp90, leading in turn to reduced chaperone function. In agreement with a previous report (27), we found that HDAC6 physically interacted with Hsp90. Inhibiting or silencing HDAC6 caused hyperacetylation of Hsp90, resulting in the loss of interaction between Hsp90 and HDAC6 (Fig. 6). Interestingly, neither treatment with TSA nor SAHA caused a significant increase in Hsp90 acetylation in HDAC6 knockdown cells. This finding is consistent with another report (27) and suggests that HDAC6 may be the principal deacetylase involved in the regulation of Hsp90 acetylation. In untreated cells, Hsp90 exists in a cytosolic complex that includes the AhR, p23, and XAP-2 (7)(8)(9)(10)(11). The interaction between Hsp90 and AhR, p23 and XAP-2 was lost when HDAC6 was silenced or cells were treated with an HDAC inhibitor (Fig. 6). Thus, the accumulation of hyperacetylated Hsp90 in HDAC6-deficient cells prevents stable complexes from forming with both the cochaperone p23 and the AhR. As . In A and B, the cells were lysed, nuclear (lanes N) and cytosolic (lanes C) fractions were isolated, and protein (100 g/lane) was subjected to immunoblotting. The immunoblots were probed with the indicated antibodies. In C-G, ChIP assays were performed. Chromatin fragments were immunoprecipitated with antibodies against AhR, and the CYP1A1 (C-E) and CYP1B1 promoters (F and G) were amplified by PCR. DNA sequencing was carried out, and the PCR product was confirmed to be the correct promoter. The CYP1A1 and CYP1B1 promoters were not detected when normal IgG was used or when antibody was omitted from the immunoprecipitation step (data not shown). mentioned above, the AhR must be in a complex with Hsp90 and p23 to exhibit high affinity ligand binding activity. Hence, we also investigated whether hyperacetylation of Hsp90 altered the ligand binding capacity of the AhR. Silencing HDAC6 or treatment with HDAC inhibitors markedly suppressed the binding of B[a]P to AhR without altering levels of the receptor (Fig. 7). Consistent with this finding, the ability of ligand to induce the nuclear translocation of AhR and activate the transcription of CYP1A1 and CYP1B1 was suppressed when HDAC6 was silenced or cells were treated with HDAC inhibitors (Fig. 8). Taken together, these results indicate that HDAC6 activity regulates the acetylation of Hsp90, the chaperone function of Hsp90, and the ability of ligand to activate AhR-dependent gene expression. Although our study is the first to show the importance of the HDAC6-Hsp90 axis for regulating the activation of AhR signaling, similar results have been reported for the GR (27). More specifically, in HDAC6-deficient cells, Hsp90dependent maturation of the GR was defective, resulting in reduced ligand binding, nuclear translocation, and gene activation. If HDAC6 regulates the Hsp90-dependent maturation of client proteins including the AhR and GR, it is reasonable to speculate that other nuclear hormone receptors, which are client proteins, e.g. progesterone receptor, will also be affected. Additional studies will be needed to evaluate this possibility.
The above findings clearly show that hyperacetylation of Hsp90 inhibits ligand-mediated activation of AhR signaling. Recently, Scroggins et al. (30) found that acetylation of a specific lysine residue (Lys 294 ) in the beginning of the middle domain of Hsp90 was critical for both cochaperone and client protein binding. In our study, conservative mutation of Lys 294 to amino acids that mimic the unacetylated (arginine) or acetylated (alanine or glutamine) state indicated that the acetylation status of Lys 294 affected the interaction between Hsp90, the cochaperone p23, and AhR (Fig.  9). Consistent with the behavior of hyperacetylated Hsp90, point mutations of Lys 294 affected the interactions with p23 and AhR. More specifically, when acetylation of this residue was mimicked, a decrease in both p23 and AhR binding was observed. Importantly, mutation of Lys 294 to mimic the acetylated state of Hsp90 also blocked both TS-and B[a]P-mediated induction of CYP1B1. Therefore, reversible acetylation of Lys 294 and possibly other sites in Hsp90 appears to provide an important level of posttranslational control that may be important for cellular responses to PAH.
Inhibitors of Hsp90 including geldanamycin and its derivatives bind within the ATP-binding pocket of the NH 2 -terminal domain of Hsp90. These agents cause a rapid decrease in amounts of AhR and thereby suppress PAH-mediated activation of CYP1A1 and CYP1B1 transcription (45,46). In the present study, we found that HDAC6 inhibition caused hyperacetylation of Hsp90 that in turn suppressed PAH-mediated induction of CYP1A1 and CYP1B1. In contrast to the findings with Hsp90 inhibitors, levels of AhR were stable in HDAC6 knockdown cells, but the ability of AhR to bind ligand and thereby activate gene expression was compromised. This result is consistent with previous findings for the GR (27). Thus, Hsp90 inhibitors and HDAC inhibitors, two classes of molecularly targeted agents, both inhibit Hsp90 function and AhR-dependent gene expression but by different mechanisms. In contrast to the findings for AhR and GR, depletion of HDAC6 has been reported to enhance the proteasomal degradation of other Hsp90 client proteins (28). It seems likely, therefore, that HDAC6 inhibition modulates the function of Hsp90 client proteins by different mechanisms including both inhibition of receptor maturation and increased proteasomal degradation.
Recently, HDAC6 null mice were found to be resistant to PAH-induced skin tumors (47). The results of the current study may help to explain this finding. The induction of CYP1A1 and CYP1B1 by ligand-activated AhR can increase the production of highly carcinogenic metabolites, creating a link between the AhR and chemical carcinogenesis. For example, B[a]P, a potent ligand of the AhR, induces its own metabolism to a toxic metabolite, anti-7,8-dihydroxy-9,10-epoxy-7,8,9,10-tetrahydroxy-benzo[a]pyrene, which covalently binds to DNA, forming bulky adducts that induce mutations (20,49). We showed that inhibiting or silencing HDAC6 suppressed PAH-mediated activation of AhR signaling and thereby blocked the induction of CYP1A1 and CYP1B1. It is reasonable FIGURE 9. Lys 294 is important for Hsp90 interaction with AhR. A represents the structure of Hsp90 protein. B and C, A549 cells were stably transfected with the indicated FLAG-Hsp90 constructs. The cells were lysed, and FLAG-Hsp90 immunoprecipitates (IP) were then subjected to immunoblotting and probed for AhR (B) and p23 (C). In D and E, the cells transfected with the indicated FLAG-Hsp90 constructs were treated for 5 h with vehicle, TS, or B[a]P (1 M). Following treatment, cell lysates (100 g/lane) were subjected to immunoblotting for CYP1B1 and ␤-actin, respectively.
to speculate, therefore, that HDAC6 deficiency will suppress carcinogen activation, which would help to explain why HDAC6 null mice are resistant to PAH induced skin tumors. Elevated levels of CYP1A1 and CYP1B1 are found in the upper aerodigestive tracts of human smokers (16,37,50). Our results suggest that inhibiting HDAC6 should reduce the expression of CYP1A1 and CYP1B1 in smokers, which could reduce the procarcinogenic effects of tobacco smoke. In addition to being able to activate carcinogens, CYP1A1 and CYP1B1 are important for the metabolism and clearance of a variety of drugs, xenobiotics, and endogenous substrates (22,48). The present findings raise the possibility that both broad spectrum HDAC inhibitors and HDAC6 inhibitors will alter the metabolism and clearance of therapeutic drugs, which could affect both their efficacy and toxicity. The metabolism and clearance of xenobiotics and endogenous substrates could also be affected. In future studies of HDAC6 inhibitors, our findings suggest that it will be important to monitor blood levels of coadministered drugs that are metabolized by enzymes encoded by AhR-dependent genes.