Activated glucocorticoid receptor interacts with the INHAT component Set/TAF-Ibeta and releases it from a glucocorticoid-responsive gene promoter, relieving repression: implications for the pathogenesis of glucocorticoid resistance in acute undifferentiated leukemia with Set-Can translocation

Abstract

Set/template-activating factor (TAF)-Ibeta, part of the Set-Can oncogene product found in acute undifferentiated leukemia, is a component of the inhibitor of acetyltransferases (INHAT) complex. Set/TAF-Ibeta interacted with the DNA-binding domain of the glucocorticoid receptor (GR) in yeast two-hybrid screening, and repressed GR-induced transcriptional activity of a chromatin-integrated glucocorticoid-responsive and a natural promoter. Set/TAF-Ibeta was co-precipitated with glucocorticoid response elements (GREs) of these promoters in the absence of dexamethasone, while addition of the hormone caused dissociation of Set/TAF-Ibeta from and attraction of the p160-type coactivator GRIP1 to the promoter GREs. Set-Can fusion protein, on the other hand, did not interact with GR, was constitutively co-precipitated with GREs and suppressed GRIP1-induced enhancement of GR transcriptional activity and histone acetylation. Thus, Set/TAF-Ibeta acts as a ligand-activated GR-responsive transcriptional repressor, while Set-Can does not retain physiologic responsiveness to ligand-bound GR, possibly contributing to the poor responsiveness of Set-Can-harboring leukemic cells to glucocorticoids.

Figure 1. Set/TAF-Iβ represses GR-induced transcriptional activity

A: TAF-Iα and Set/TAF-Iβ repress GRα-induced transcriptional activity of the MMTV promoter stably incorporated in host cell chromatin in HCT116/MMTV cells. HCT116/MMTV cells were transfected with the indicated amounts of the TAF-Iα-or Set/TAF-Iβ-expressing plasmid, together with pRShGRα and pSV40-β-Gal. Left panel: Dose effects of TAF-Iα and Set/TAF-Iβ are shown. Bars represent mean ± S.E. values of luciferase activity normalized for β-galactosidase activity in the absence or presence of 10⁻⁶ M of dexamethasone (Dex). *: p<0.01 (ANOVA, followed by Bonferroni correction), compared to the baseline (the value obtained in the absence of TAF-Iα and Set/TAF-Iβ, but in the presence of dexamethasone). Right panel: Set/TAF-Iβ shifts downward the dexamethasone titration curve of luciferase activity from the chromatin-integrated MMTV promoter. Circles represent mean ± S.E. values of luciferase activity normalized for β-galactosidase activity in the indicated concentrations of dexamethasone (Dex). *: p<0.01 (ANOVA, followed by Bonferroni correction), obtained by comparing the results in the absence and presence of Set/TAF-Iβ transfection. B: Endogenous Set/TAF-Iβ is a negative regulator of GR-induced transcriptional activity in HTC cells. HTC cells were transfected with Set/TAF-Iβ siRNA or control siRNA. The cells were stimulated with 10⁻⁶M of dexamethasone (Dex). Bars indicate the mean ± S.E. of the tyrosine aminotransferase (TAT) activity, mRNA levels of TAT or those of Set/TAF-Iβ. *: p<0.01 (unpaired two-tailed Student t test), by comparing the two values indicated. C: Set/TAF-Iβ represses the transcriptional activity of the mineralocorticoid (MR) and progesterone receptor (PR)-A in addition to that of the GR. HCT116 cells were transfected with Set/TAF-Iβ-expressing plasmid together with GRα-, MR- or PR-A-expressing plasmid, pMMTV-Luc or (κB)₃-Luc, and pSV40-β-Gal. Bars represent mean ± S.E. values of luciferase activity normalized for β-galactosidase activity in the absence or presence of 10⁻⁶ M of dexamethasone (Dex) or progesterone (Prog), 10⁻⁸ M of aldosterone (Aldo), or p65/p50 transfection. *: p<0.01, n.s.: not significant (unpaired two-tailed Student t test), compared to the baseline (“(-)” in the presence of ligand or p65/p50).

Figure 2. The Set/TAF-Iβ region included between amino acids 181 and 225 (GR-binding domain) is responsible for binding to the GR, while the INHAT domain (amino acids 225–277) is necessary and sufficient for the suppression of GR-induced transcriptional activity

A: Linearized Set/TAF-Iβ fragments employed. The GR-binding (amino acids 181–225) and INHAT (amino acids 225–277) domains are shown in gray and black, respectively. B: The Set/TAF-Iβ-(181–225) binds to the GR DBD in a yeast two-hybrid assay. EGY48 yeast cells were transformed with p8OP-LacZ, pLexA-GRα-(420–489) and the indicated Set/TAF-Iβ-fragment-expressing plasmids. Bars represent the mean ± S.E. of fold activation compared to the baseline (the value obtained in the absence of Set/TAF-Iβ expression). *: p<0.01 (unpaired two-tailed Student t test ), compared to the baseline. C: The INHAT domain (amino acids 225–277) is sufficient for the suppression of GR-induced transcriptional activity in HCT116 cells. HCT116 cells were transfected with the indicated Set/TAF-Iβ fragment-expressing plasmid together with pRShGRα, pMMTV-Luc and pSV40-β-Gal. Open and solid bars indicate the mean ± S.E. of luciferase activity in the absence or presence of 10⁻⁶ M dexamethasone (Dex). *: p<0.01(unpaired two-tailed Student t test), compared to baseline (value obtained in the absence of Set/TAF-Iβ expression). #: p<0.01 (unpaired two-tailed Student t test), compared to “2–277” in the presence of dexamethasone. n.d.: not determined. D: Set-TAF-Iβ is associated through amino acids 181 to 225 with the GR in vivo. HCT116/MMTV cells were transfected with the wild type His-Set-TAF-Iβ-or His-Set-TAF-Iβ(Δ181–225)-expressing plasmid together with pRShGRα, treated with 10⁻⁶ M of dexamethasone, and GR- Set-TAF-Iβ complexes were precipitated with control or anti-GR antibody. His-Set-TAF-Iβ-, His-Set-TAF-Iβ(Δ181–225) and GR were visualized with anti-His and anti-GR antibodies. IP results are shown in the top panel, whereas expressed GR, His-Set-TAF-Iβ and His-Set-TAF-Iβ(Δ181–225) are demonstrated in the bottom 2 panels.

Figure 3. Binding of GR to GREs releases pre-existing INHAT complex from GREs via physical interaction between the GR and Set/TAF-Iβ

A: Ligand-activated GR releases INHAT components Set/TAF-Iβ and pp32 from the MMTV GREs in HCT116/MMTV cells (left panels) and TAT GREs in HTC cells (right panels) HCT116/MMTV cells transfected with pRShGRα or the control plasmid and HTC cells were treated with 10⁻⁶ M of dexamethasone (Dex). They were fixed with 1% formaldehyde and ChIP assays were performed by using anti-Set/TAF-Iβ, anti-pp32, anti -GRα or control antibodies. B: Ligand-activated GR releases the wild type Set/TAF-Iβ but not the GR-binding domain-defective Set/TAF-Iβ-(Δ181–225) from the MMTV-GREs in HCT116/MMTV cells. HCT116/MMTV cells were transfected with His-Set/TAF-Iβ-expressing plasmids and/or pRShGRα. The cells were treated with 10⁻⁶ M of dexamethasone (Dex), fixed with 1% formaldehyde and ChIP assays were performed by using anti-His, anti-GRαor control antibodies. C: Overexpression of Set/TAF-Iβ-(181–225) attenuates GR-induced release of Set/TAF-Iβ from GREs in HCT116/MMTV cells. HCT116/MMTV cells were transfected with 0.5 μg/well of His-Set/TAF-Iβ-expressing plasmid (as “1”), pRShGRα and the indicated amounts (ratios) of the Set/TAF-Iβ-(Δ181–225)-expressing plasmid. The cells were treated with 10⁻⁶ M of dexamethasone (Dex), fixed with 1% formaldehyde and ChIP assays were performed using anti-His, anti-GRαor control antibodies.

Figure 4. Set/TAF-Iβ and GRIP1 act antagonistically on GR-induced transcriptional activity

A: Set/TAF-Iβ attenuates HAT coactivator-induced enhancement of GR-stimulated transcriptional activity in HCT116 cells. HCT116 cells were transfected with pRShGRα together with indicated coactivator-expressing plasmid, and pMMTV-Luc in the absence or presence of Set/TAF-Iβ-expressing plasmid. Bars represent mean ± S.E. values of luciferase activity normalized for β-galactosidase activity in the absence or presence of 10⁻⁶ M of dexamethasone (Dex). *: p<0.01 (unpaired two-tailed Student t test), obtained by comparing the two values indicated. B: Ligand-activated GR releases MMTV promoter-bound Set/TAF-Iβ from GREs, while it attracts the HAT coactivator GRIP1 to GREs in HCT116/MMTV cells. HCT116/MMTV cells were transfected with GRIP1-expressing plasmid, His-Set/TAF-Iβ-expressing plasmid and/or pRShGRα. The cells were treated with 10⁻⁶ M of dexamethasone (Dex), fixed with 1% formaldehyde and ChIP assays were performed by using anti-His, anti-GRα, anti-GRIP1 or control antibodies. C: Set/TAF-Iβ-(Δ181–225), a mutant defective in the GR-binding domain, is constitutively associated with GREs but does not inhibit dexamethasone-induced attraction of GRIP1 to GREs. HCT116/MMTV cells were transfected with His-Set/TAF-Iβ-expressing plasmid, GRIP1-expressing plasmid, and/or pRShGRα. The cells were treated with 10⁻⁶ M of dexamethasone (Dex), fixed with 1% formaldehyde and ChIP assays were performed by using anti-His, anti-GRIP1, and anti-GRα antibodies. D: Endogenous Set/TAF-Iβ suppresses dexamethasone-induced acetylation of histone H3 (K14) associated with GREs. HCT116/MMTV cells were transfected with control or Set/TAF-Iβ siRNA, together with pRShGRα. The cells were treated with 10⁻⁶ M of dexamethasone (Dex), fixed with 1% formaldehyde and ChIP assays were performed by using anti-acetylated H3 (K14) antibody. Bars indicate fold acetylation of histone H3 (K14) (left panel) and fold induction of the Set/TAF-Iβ mRNA expression compared to that of the baseline. *: p<0.01, (unpaired two-tailed Student t test), compared to baseline (“Set/TAF-Iβ siRNA (−)” in presence of dexamethasone). E: Set/TAF-Iβ-(Δ181–225), a mutant defective in the GR-binding domain, suppresses GRIP1-induced enhancement of dexamethasone-stimulated acetylation of histone H3 (K14) more strongly than the wild type Set/TAF-Iβ. HCT116/MMTV cells were transfected with His-Set/TAF-Iβ wild type or (Δ181–225)-expressing plasmid and/or GRIP1-expressing plasmid, together with pRShGRα. The cells were treated with 10⁻⁶ M of dexamethasone (Dex), fixed with 1% formaldehyde and ChIP assays were performed using anti-acetylated H3 (K14) antibody. Bars indicate fold acetylation of histone H3 (K14) above the baseline. *: p<0.01 (unpaired two-tailed Student t test), compared to the baseline (the value obtained in the absence of plasmid transfection) or obtained by comparing the two values indicated.

Figure 5. The pathologic Set-Can fusion oncoprotein represses dexamethasone-stimulated transcriptional activity of the MMTV promoter by antagonizing GRIP1-induced acetylation of histone H3

A: Set-Can suppresses GR-induced transcriptional activity in HCT116 cells. HCT116 cells were transfected with TAF-Iβor Set/Can-expressing plasmid together with pRShGRα, pMMTV-Luc and pSV-40-β-Gal. Bars represent mean ± S.E. values of luciferase activity normalized for β-galactosidase activity in the absence or presence of 10⁻⁶ M of dexamethasone (Dex). *: p<0.01 (unpaired two-tailed Student t test), compared to the baseline (“Control” in the presence of dexamethasone), or obtained by comparing the two values indicated. B: Set-Can is not physically associated with the GR in vivo. HCT116/MMTV cells were transfected with the wild type His-Set-TAF-Iβ-or His-Set-Can-expressing plasmid together with pRShGRα and were treated with 10⁻⁶ M of dexamethasone. The protein complexes were precipitated with control or anti-GR antibody and His-Set-TAF-Iβ-, His-Set-Can and GR were visualized with anti-His and anti-GR antibodies. IP results are shown in the left panel, while expressed His-Set-TAF-Iβ and His-Set-Can were demonstrated in the right panel. GR expressed was also shown in the bottom panel. C: Set-Can represses GRIP1-induced enhancement of GR transcriptional activity in HCT116/MMTV cells. HCT116/MMTV cells were transfected with pRShGRα together with pMMTV-Luc and pSV40-β-Gal in the absence or presence of GRIP1- and/or Set/TAF-Iβ-expressing plasmids. Bars represent mean ± S.E. values of luciferase activity normalized for β-galactosidase activity in the absence or presence of 10⁻⁶ M of dexamethasone (Dex). Numbers on the bars indicate mean ± S.E values of GRIP1-induced fold enhancement of GR transcriptional activity in the absence and presence of Set-Can. D: Set-Can dose-dependently represses GRIP1-induced enhancement of GR transcriptional activity in HCT116/MMTV cells. HCT116/MMTV cells were transfected with the indicated amounts of Set/TAF-Iβ-expressing plasmid, together with 0.5 μg/well of the GRIP1-expressing plasmid, pRShGRα and pSV40-β-Gal, and pMMTV-Luc. Bars represent mean ± S.E. values of luciferase activity normalized for β-galactosidase activity in the absence or presence of 10⁻⁶ M of dexamethasone (Dex). *: p<0.01 (ANOVA, followed by Bonferroni correction), compared to the baseline (the value obtained in the presence of GRIP1 transfection and absence of Set-Can transfection). E: Set-Can is constitutively associated with GREs in HCT116/MMTV cells. HCT116/MMTV cells were transfected with His-Set/TAF-Iβ-or His-Set-Can-expressing plasmid, and pRShGRα. The cells were treated with 10⁻⁶ M of dexamethasone (Dex), fixed with 1% formaldehyde and ChIP assays were performed by using anti-His, anti-GRα or control antibodies. F: Set-Can is associated with GREs both in the absence and presence of dexamethasone. HCT116/MMTV cells were transfected with His-Set-Can-expressing plasmid and GRIP1-expressing plasmid, together with pRShGRα. The cells were treated with 10⁻⁶ M of dexamethasone (Dex), fixed with 1% formaldehyde, and ChIP assays were performed by using anti-His, anti-GRIP1 or anti-GRα antibodies. G: Set-Can suppresses GRIP1-induced enhancement of dexamethasone-stimulated acetylation of histone H3 (K14) more strongly than the wild type Set/TAF-Iβ. HCT116/MMTV cells were transfected with 0.5 μg/well of His-Set/TAF-Iβ or His-Set-Can-expressing plasmid and/or 0.5 μg/well of GRIP1-expressing plasmid, together with 0.5 μg/well of pRShGRα. The cells were treated with 10⁻⁶ M of dexamethasone (Dex), fixed with 1% formaldehyde, and ChIP assays were performed by using anti-acetylated H3 (K14) antibody. Bars indicate fold acetylation of histone H3 (K14) above the baseline. *: p<0.01 (unpaired two tailed Student t test), compared to the baseline (the value obtained in the absence of plasmid transfection), or obtained by comparing the two values indicated.

Figure 6. Activated GR releases INHAT from GREs through physical interaction with Set/TAF-Iβ, permitting HAT-induced acetylation of histone tails and hence gene transcription, while Set-Can suppresses GR transcriptional activity by being continuously associated with GREs and preventing GR coactivator-driven acetylation

A: INHAT is associated with a GRE-containing promoter at the baseline state, binding histones and protecting them from transcription factor/nuclear receptor coactivator HAT-mediated acetylation. Once ligand becomes available, activated GR induces dissociation of INHAT from histones through physical interaction with Set/TAF-Iβ, making these histones susceptible to HAT-induced acetylation. B: Pathologic Set-Can fusion protein remains associated with GREs in the presence of glucocorticoids, suppressing GR-induced transcriptional activity by preventing HAT-induced histone acetylation. INHAT: inhibitor of the histone acetyltransferases, HAT: histone acetyltransferase, GR: glucocorticoid receptor, GREs: glucocorticoid response elements.