help button home button Endocrine Society Molecular Endocrinology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Molecular Endocrinology, doi:10.1210/me.2004-0255
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
18/12/2924    most recent
Author Manuscript (PDF)
Right arrow Purchase Article
Right arrow View Shopping Cart
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow NURSA Molecule Pages Link
Right arrow Request Copyright Permission
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Kim, S.-W.
Right arrow Articles by Larsen, P. R.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Kim, S.-W.
Right arrow Articles by Larsen, P. R.
Molecular Endocrinology 18 (12): 2924-2936
Copyright © 2004 by The Endocrine Society

A Novel Cell Type-Specific Mechanism for Thyroid Hormone-Dependent Negative Regulation of the Human Type 1 Deiodinase Gene

Sung-Woo Kim, Seong-June Hong, Kyung Min Kim, Sung-Chul Ho, Edward C. So, John W. Harney and P. Reed Larsen

Thyroid Section, Division of Endocrinology, Diabetes and Hypertension, Department of Medicine, Brigham and Women’s Hospital, Harvard Institute of Medicine, Boston, Massachusetts 02115

Address all correspondence and requests for reprints to: S.-W. Kim, Thyroid Section, Division of Endocrinology, Diabetes and Hypertension, Department of Medicine, Brigham and Women’s Hospital, Harvard Institute of Medicine, 77 Avenue Louis Pasteur, Boston, Massachusetts 02115. E-mail: swkim{at}rics.bwh.harvard.edu.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 
We have identified a cell type-specific, negative thyroid hormone-responsive element in the human type 1 iodothyronine deiodinase (hdio1) gene. This fragment, termed a JEG response element, bound tightly to a JEG-cell nuclear protein [JEG cell-specific transcription factor (JTF)] also present in placenta but not in COS-7, HeLa, or human embryonic kidney-293 cells. In JEG-3 cells, three copies of the JEG response element conferred a more than 40-fold transcriptional stimulation to the heterologous rat GH promoter which was further increased 2-fold by apo-thyroid hormone receptor (TR) and reduced 3-fold by T3. Dimethyl sulfide footprinting showed overlapping contact sites for the high-affinity interaction of JTF and low-affinity binding of TR-retinoid X receptor. Expression of the same construct was unaffected by TR or T3 in COS cells, indicating JTF was required for negative regulation by T3-TR. Mutations of the critical thyroid hormone responsive element binding P box amino acids EG to GS in TR{alpha}1 or TRß2 eliminated the apo-TR and T3-TR effects. These studies identify a novel mechanism for cell type-specific, promoter-independent negative regulation by T3.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 
THYROID HORMONE PLAYS important roles in growth, development, differentiation, metamorphosis, metabolic regulation, and thermogenesis. The biologically active hormone, T3, is produced from the prohormone T4 by the type 1 and 2 iodothyronine deiodinases (D1 and D2), which remove an iodine from the outer ring. D1 is abundantly expressed in liver, kidney, and thyroid in all species studied (1). The human, rat, and mouse dio1 genes are positively regulated by T3 in the liver and kidney (1). Two thyroid hormone-response elements (TREs) have been identified within 700 bp of the human dio1 transcription start site (TSS) that confer the positive response of this gene to T3 (2, 3). Positive transcriptional regulation by T3 requires thyroid hormone receptor (TR)-DNA binding either as a homodimer or as a heterodimer with a retinoid X receptor (RXR). Binding of T3 changes the conformation of the TR causing a dissociation of corepressors such as the nuclear receptor corepressor (NCoR), the silencing mediator for retinoid and TRs, and histone deacetylases (HDACs) but recruiting coactivators such as cAMP response element binding protein (CREB)-binding protein/p300 and the CREB-binding protein-associated factor, which contain histone acetyl transferase activities (4, 5, 6, 7, 8, 9, 10, 11, 12, 13). Subsequent acetylation of histone molecules is thought to increase accessibility of chromatin to RNA polymerase II (14).

The negative regulation of gene expression by thyroid hormone is common, but despite considerable efforts, no single mechanism that explains this process has been elucidated. There are several models for negative regulation by T3 (14). Most frequently studied have been those genes that are part of the feedback-regulatory loop controlling thyroid function, e.g. the genes encoding the TSH{alpha} and ß subunits and TRH (15, 16, 17). A direct TR-DNA-binding model proposes that binding of T3 to TR at so called "negative TREs" on target genes is required for negative regulation by T3. For example, the mutation of a TRE half-site sequence near the TSS of the TSHß gene abolishes hTRß binding or the use of a hTRß with a mutation in the zinc finger region of the receptor eliminates the T3-dependent repression of this gene (18, 19, 20). As a mechanism for the negative regulation by these DNA sequences, some have shown recruitment of HDACs to the TR-T3 complex (21). Other systems in which direct TR binding is required for negative regulation are in genes containing a binding site for the CCCTC-binding factor (CTCF). This is present in several negatively regulated genes including the lysozyme silencer F1, genomic DNA segment 144, and mammalian c-myc (22, 23, 24). Another mechanism requiring TR-DNA binding occurs in genes with TR binding sites that overlap those for transcriptional enhancers. Examples are the human epidermal growth factor receptor and the ß-amyloid precursor protein (APP) genes (25, 26).

Other proposals are that unoccupied TR enhances expression of negatively regulated genes by binding corepressors-HDAC complexes facilitating their expression in a indirect manner. Such a "squelching" mechanism has been reported for the TSH{alpha} subunit gene (17, 27). Yet another model proposes that TR interacts with DNA indirectly by protein-protein interaction such as in the jun-fos heterodimers of activator protein 1 sites (14, 28).

In the course of studies of hdio1 transcriptional regulation, we were surprised to find that in the JEG-3 choriocarcinoma cell line, an approximately 5-kb 5'-flanking region (FR) chloramphenicol acetyltransferase (CAT) construct was negatively regulated by T3 when coexpressed with either TR{alpha} or TRß in transient transfection studies whereas it is positively regulated in COS-7 and HEK-293 cells. This suggested that there were specific properties of the JEG cell that facilitated T3-dependent negative regulation. In this work, we have analyzed this cell type-specific mechanism for negative regulation by T3. Negative regulation of hdio1 in these cells requires TR-DNA binding but possibly not to the hdio1 gene. We have identified a nuclear protein in JEG-3 cells which binds with high affinity to a site in the 5'-FR of the human dio1 gene, markedly enhancing its transcription. In transient expression studies in JEG-3, but not in COS-7 cells, dio1 gene expression is further enhanced by coexpressing TR and expression is markedly reduced by addition of T3. There is no evidence for binding of TR-RXR to the JEG cell protein-DNA complex in vivo in COS-7 cells although it does bind with low affinity in vitro. Nonetheless, a mutation in the TR that blocks its capacity to bind to a TRE eliminates negative regulation. Such an example of negative regulation by T3 has not been described previously, suggesting yet another mechanism for thyroid hormone action.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 
TR-T3-Dependent, Promoter-Independent Negative Regulation Is Directed by Sequences from the 5-kb hdio1 Promoter in JEG-3 But Not COS-7, HEK-293, or HepG2 Cells
Analyses of hdio1 5'-FR-reporter constructs showed that when sequences of 3.5 kb or greater were included in the construct, a significant negative regulation of the hdio1 gene was conferred by T3 in the presence of coexpressed TR in JEG-3 cells (Fig. 1Go). This was further confirmed by showing a similar effect on a heterologous promoter-CAT [137-bp rat GH (rGH) CAT] plasmid using TRß1. Functional studies of this 5'-FR suggested that the critical sequences were in a 100-bp fragment located between –3.4 and –3.3 kb relative to the TSS (Fig. 1Go). The negative regulation of hdio1 by T3 was TR dependent as well as cell type specific. In the absence of TR, the 5-kb hdio1 construct was not significantly affected by T3 (Table 1Go). However, in the presence of TR{alpha} (or TRß, data not shown) in JEG-3 cells, T3 caused an approximately 7-fold decrease in CAT expression in the JEG-3 cells but a 3-fold increase in COS-7 cells (Table 1Go) as well as in HEK-293 and HepG2 cells (data not shown). Taken together, these results indicated that there were specific sequences between approximately –3.4 and –3.3 kb that confer T3-dependent negative regulation to either the hdio1 or rGH promoters in JEG-3, but not in three other cell lines.



View larger version (21K):
[in this window]
[in a new window]
  		Fig. 1. Localization of hdio1 Sequences Conferring a Negative Response to T3-TR in JEG-3 Cells

A, Effect of T3 on the transient expression of hdio1 5'-FR-CAT of various lengths. A mouse TR{alpha}1 (mTR{alpha}1) expressing vector and TKhGH as a transfection efficiency control were cotransfected, and CAT expression was normalized to hGH expression. Results shown are the mean CAT/hGH ± SD ratios of the constructs. B, Effects of the sequences of the hdio1 fragment conferring a negative response to T3 on expression of the heterologous 137-bp rGH promoter in JEG-3 cells. CAT expression was normalized to hGH expression as described. Values are mean ± SD.

 

View this table:
[in this window]
[in a new window]
  		Table 1. Cell Type-Specific Negative Regulation of 5-kb hdio1-CAT by Thyroid Hormone

 
JEG-3 Cell Nuclear Protein Binding Sequences in the hdio1 Promoter
To determine whether this negative regulation could be explained by an hdio1 binding protein in JEG cells, we performed DNAse I footprinting of sequences between –3.4 and –3.3 kb. We also exposed these sequences to either bacterially expressed chicken TR{alpha} and human RXR{alpha} (cTR{alpha}/hRXR{alpha}) or a nuclear extract (NE) of JEG cells transiently expressing TR exposed or not exposed to T3 (Fig. 2Go). All three complexes protected sequences between –3367 and –3390 with no effect of the coexpression of TR on the JEG cell extract pattern with or without T3 (Fig. 2Go). DNAse I hypersensitivity sites were present at –3369 with the NEs and about 3368–67 with TR-RXR, suggesting a protein-DNA interaction and increased DNAse accessibility at this position. Protection by TR-RXR and NE plus or minus T3 was protein concentration dependent, and there was no obvious effect of T3 on the NE binding.



View larger version (85K):
[in this window]
[in a new window]
  		Fig. 2. Identification of JEG Cell Nuclear Protein-Binding Sites in the Functionally Active Fragment of hdio1 5' FR Using DNAse I Protection

A, A 32P-labeled 54-bp hdio1 fragment (–3431 to –3378) was incubated with various concentrations of bacterially expressed cTR{alpha}1-hRXR{alpha} or JEG cell NE isolated from JEG cells transfected with mTR{alpha}1 cultured in the presence or absence of T3. Unprotected DNA was partially digested by DNAse I (see Materials and Methods) and compared with naked DNA under the same conditions (lane 1). The symbols indicate the quantity of protein added.

 
These results indicated that there were binding sites for a nuclear protein and/or TR-RXR complexes located in this functionally active region of the hdio1 promoter. This was confirmed by EMSAs using the 33-bp fragment corresponding to the footprinted region (–3393 to –3359) and either bacterially expressed TR-RXR or NEs from cells transfected with or without rat TRß1 (Fig. 3Go). The mobility of the TR{alpha}-RXR{alpha}-bound DNA was slightly greater than that of NEs from JEG cells expressing or not expressing TR. Neither the TRß1 cotransfection nor exposure to 50 nm T3 changed the mobility or amount of the NE DNA complex. Because the stimulation of hdio1 appeared to be JEG-3 cell specific, we compared JEG NE with that from COS-7, HeLa, and HEK-293 cells. Little, if any, retardation of the hdio1 sequences (the slowest migrating band) occurred with these nuclear proteins (Fig. 3BGo). Together with the transient expression studies and the footprinting results, these data suggested that a JEG cell nuclear protein conferred the apo-TR and T3-TR effects on hdio1 gene expression in these cells.



View larger version (67K):
[in this window]
[in a new window]
  		Fig. 3. EMSA Using the 33-bp Footprinted DNA Fragment

A, Interaction with bacterially expressed cTR{alpha}-hRXR{alpha} (lane 1) or NE isolated from JEG cells transfected (lanes 2 and 3) or not transfected (lanes 4 and 5) with hTRß1 expression vector in the presence and absence of 50 nM T3. B, Cell type-specific expression of the JEG DNA-binding protein. NEs from JEG-3, COS-7, HeLa, and HEK293 cells were incubated with a 32P-labeled 33-bp fragment. A 1-ng probe was reacted with 0.16/0.05 µg of bacterially expressed cTR{alpha}-hRXR{alpha} or 1.2 µg of NE protein for EMSA.

 
The Nuclear Protein in JEG-3 Cells Is an Enhancer of hdio1 Expression
Given the specific binding of the functionally active hdio1 sequences by JEG nuclear protein, we compared the effects of this 33-bp footprinted fragment on a heterologous rGH CAT promoter in the absence and presence of TR and T3. Expression of the native rGH promoter in JEG-3 cells was more than 40-fold enhanced by three copies of the 33-bp hdio1 gene sequences (Fig. 4Go). That expression was doubled by coexpression of TR{alpha} but transient expression in the presence of TR was reduced about 3-fold by T3. In COS-7 cells, there was no effect of the dio1 sequences on expression of the rGH-CAT construct. TR and T3 were also without effect. These results establish that the 33-bp fragment is a binding site for a potent transcriptional enhancer protein present in JEG cell nuclei. For convenience, we will refer to this cell type-specific protein as the JEG transcription factor (JTF) and the sequences to which it binds in the hdio1 promoter as the JTF-response element or JRE. These results also suggested that the APO-TR and T3-TR effects were not due to direct interaction of TR-RXR with the JRE because they do not occur in COS cells.



View larger version (18K):
[in this window]
[in a new window]
  		Fig. 4. The JEG Cells Nuclear Protein Acts as a Transcriptional Enhancer by Interacting with Sequences from the hdio1 Gene

5'-FR 3X hdio1-rGHCAT and rGHCAT plasmids (33 bp) were transfected with or without mTR{alpha}1 expression plasmid in the absence or presence of 50 nM T3 into JEG-3 and COS-7 cells (see scale difference). CAT expression was normalized to hGH expression as described.

 
Specific Identification of Nuclear Protein or TR-RXR Binding Sites Using Dimethyl Sulfide (DMS) Protection and Mutational Analyses
The EMSA technique was combined with DMS methylation by DMS to identify specific G residues in the JRE contacted by either JTF or TR-RXR. JTF produced a specific protection of a site between –3368 and –3381, best seen on the lower strand of the 33-bp fragment (Fig. 5Go). The G-rich region on the lower strand between –3377 to –3381 was also protected by TR-RXR. However, the 3'-portions of the contact sites of JTF and TR-RXR were distinct. The protection pattern using TR-RXR was consistent with its binding, either in an inverted palindromic fashion with the half-sites separated by a single base pair (half-site a + half-site b) or a direct repeat of half-sites separated by 4 bp (the lower strand b+c, Fig. 5BGo). However, none of these half-sites have the YGGTNN sequence characteristic of high-affinity TR-RXR binding sequences.



View larger version (46K):
[in this window]
[in a new window]
  		Fig. 5. DMS in Vitro Footprinting of the 100-bp DNA Fragment

A, 32P-labeled probe methylated by DMS was incubated with cTR{alpha}1-hRXR{alpha} or JEG NE. Bound (B) and free (F) fragments were isolated, and the protected G residues were identified by cleaving methylated G residues with piperidine. Protected G residues are marked with arrowheads. B, Schematic of G residues protected by JEG-NE or cTR{alpha}-hRXR{alpha}. Potential TR binding half-sites are marked with arrows.

 
To confirm the significance of the footprinted G residues, we made specific mutations in the JRE sequences and analyzed their effects on the EMSA (Fig. 6Go, A and B). The studies of JTF employed NEs from JEG cells that were transiently expressing TRß. However, as shown earlier (Fig. 3AGo), TRß1 does not change the electromobility shift by JTF, indicating that JTF is the high-affinity protein responsible for this binding. The M1 mutation caused loss of binding of TR-RXR and JTF. M2 reduced TR-RXR binding but not that of JEG-NE and, as expected, the M3 mutations caused loss of binding to both. The M4, -5, and -6 changes caused decreases in the affinity of the JRE for TR-RXR, but there was still some association. Despite the M4 mutations, JTF still bound to the JRE, but this binding was lost for the M5 mutation and reduced for the single nucleotide mutation in M6. These results confirmed that the "b site" on the lower strand (Fig. 5BGo) was critical for JEG-NE and TR/RXR binding. The footprinted GG residues on the top strand are also critical for JTF binding but less important for that of TR-RXR. The role of the "a" and "c" sites for TR-RXR binding was ambiguous. The results of these studies suggested that the binding of JTF was much stronger and more specific than that of TR-RXR and confirmed that cotransfection of TRß1 did not affect the specificity of this binding.



View larger version (55K):
[in this window]
[in a new window]
  		Fig. 6. Analysis of Mutations Designed to Block JTF or cTR{alpha}-hRXR{alpha} Binding

A, Mutation sites within the JRE sequence. 1–3 bases were mutated as denoted in M1–M6. B, EMSA of WT and mutated 32P-labeled 33-bp fragments with bacterially expressed cTR{alpha}1/hRXR{alpha} or JEG NE (see Materials and Methods). C, Comparison of the binding pattern of NEs from JEG-3 cells and freshly obtained human placenta. Probe (1 ng) was reacted with 0.16/0.05 µg cTR{alpha}1/hRXR{alpha}, 1.2 µg JEG-protein, and placenta.

 
Because JEG-3 cells are a human choriocarcinoma cell line, we compared the binding of NE from human placenta to that obtained with JEG cell NE with both wild-type (WT) JRE and fragments containing M1, -2, and -3 mutations (Fig. 6CGo). The quantity of JTF retarded was much higher for JEG-3 cells than for human placental NE using comparable amounts of protein. However, the pattern of retardation of the WT and mutant JREs was similar with JEG and placental NEs suggesting the presence of a similar nuclear protein presumably in the trophoblasts present in that organ.

TR-TR or TR-RXR Does Not Inhibit JTF Binding to the JRE
Because of the overlap of the binding sequences for JTF and TR-RXR, we examined the possibility of interaction or interference with JTF-JRE binding by TR or TR-RXR and also for binding of TR by JTF (Fig. 7Go). There was potent self-displacement of labeled JRE binding by the unlabeled WT sequences with a near disappearance of labeled JRE binding by JEG NE with as little as a 10-fold increase in unlabeled JRE. The inhibition of binding, however, was not complete, even with a 1000-fold excess of JRE, suggesting either some degree of nonspecific binding or a subset of low-affinity, high-capacity binding sites. Importantly, neither the chicken lysozyme F2 TRE nor the malic enzyme TRE interfered with the JTF-JRE interaction nor did random DNA at a 1000-fold excess (Fig. 7AGo). Furthermore, JEG NE bound only weakly to the F2 chicken lysozyme silencer TRE and did not retard the malic enzyme TRE (Fig. 7BGo). These and the above results suggested that JTF bound sites in JRE which could also bind TR-RXR but that the JTF affinity was much higher.



View larger version (44K):
[in this window]
[in a new window]
  		Fig. 7. The JTF-JRE Interaction Is Not Perturbed by TR-RXR or by TRE Binding

A, The EMSAs were performed with 32P-labeled JRE (1 ng) and JEG NE (0.4 µg). Competition was with various amounts of unlabeled JRE, 27 bp lysozyme silencer F2-TRE (0.1 or 1 µg), and 28 bp rat malic enzyme (ME) TRE (0.1 or 1 µg) or random DNA (RD) (1 µg). B, EMSA using 32P-labeled WT JRE (1 ng), F2-TRE (1 ng), ME-TRE (1 ng), and JEG-NE (0.4 µg). C, Absence of effects of cTR{alpha}-hRXR{alpha} on the interaction of JTF (from 0.4 µg JEG-NE) with WT or M2-mutant 32P-labeled JRE.

 
We therefore tested whether TR{alpha}, without or with RXR{alpha}, could perturb or compete with JTF-JRE interactions. Neither low nor high quantities of TR{alpha} interfered with JTF binding to either the WT or the M2-JRE. This mutation was specifically designed to interfere with TR-RXR binding but not with that of JTF (Fig. 7CGo). Despite this, high concentrations of TR led to the retardation of the WT JRE and M2 by putative TR{alpha} homodimers. There was TR-RXR heterodimer binding to the JRE (Fig. 7CGo) with much higher affinity than for TR homodimers. Despite the binding of significant quantities of JRE by either TR homodimer or TR-RXR heterodimer, the JTF-JRE complex was unchanged in amount and position. This indicates that the binding of either TR homodimer or heterodimer is weak and does not interfere with JTF binding.

The Functional Activity of JRE Correlates with JTF Binding
As previously demonstrated, the 33-bp hdio1 fragment conferred an increase in promoter expression to the basal rGH promoter (Fig. 4Go). Accordingly, we tested the relevance of the JTF-JRE physical interaction for its enhancer activity by inserting three copies of WT or mutated JREs 5' to the rGH-CAT construct (Fig. 8Go). The WT, M2, and M4 JREs, each of which binds JTF, markedly increase expression from the rGH promoter in the absence of TR. This expression was increased by TR, and, in addition, T3 decreased rGH-CAT expression to less than its basal level. Neither the M1, M3, or M5 JRE constructs show any enhancement of basal 137rGH CAT expression in the absence or presence of TR. The effect of M6, a JRE that contains a single G nucleotide mutation which decreases JTF binding, is markedly reduced. The loss of enhancer function of the mutated JRE fragments correlates precisely with the loss of JTF binding but not with their RXR-TR binding affinity (Fig. 6Go). The M2 JRE, which has a reduced TR-RXR affinity, confers a WT increase in rGH-CAT activity with APO-TR, which was markedly reduced by T3. This suggests that despite the binding of TR-RXR to the JRE, such an interaction is unlikely to explain its in vivo effects.



View larger version (12K):
[in this window]
[in a new window]
  		Fig. 8. Functional Effects Conferred by WT and Mutant JRE from hdio1 on the 137-bp rGHCAT Vector in JEG-3 Cells

Comparison of the expression of rGHCAT containing three copies of WT or mutant 33-bp JRE fragments inserted 5' to the rGH promoter-CAT plasmid. A, Transient transfection assays with WT and mutations M1–6 (see Fig. 6Go). TR and hGH expression vectors were cotransfected with or without 50 nM T3 or a mouse TR{alpha}1 expression plasmid. CAT expression was normalized to hGH expression as described. B, Disruption of TR{alpha}1-TRE binding by introduction of mutations in the P box GS71 contains mutations converting EG at positions 71 and 72 to GS. Empty vector, mTR{alpha}1, mutant GS71, or mTR{alpha}1+NCoR expression vectors were cotransfected with WT 33-bp JRE-rGH-CAT plasmid into JEG-3 cells.

 
That TR-DNA interactions are required for the APO-TR effects is suggested in Fig. 8BGo. Coexpression of mTR{alpha}1-GS71, which contains mutations (EG to GS) in the P box of the first zinc finger that eliminate TRE binding and T3-dependent positive TRE regulation (Ref.19 and data not shown), does not confer increased expression to the WT JRE (Fig. 8BGo). We also obtained similar results using this same mutation in TRß2 (GS125) kindly provided by Dr. F. Wondisford (19).

The effect of WT APO-TR to increase transient expression of negatively regulated genes has been attributed to HDAC binding to TR-DNA-corepressor complexes. This effect was increased by expression of NCoR (17, 27). We found no effect of transient expression of NCoR on the APO-TR effect (Fig. 8BGo), suggesting that either the NCoR concentration was already maximal or that another mechanism was operative.

Isolation of JTF Candidate Proteins Using DNA-Affinity Chromatography
Using the high-affinity DNA-binding characteristics of JTF, we performed JRE-affinity chromatography to purify JTF from JEG-3 cells. DNA affinity columns were generated by coupling three copies of WT or M1 mutant (M1) JREs to Sepharose or magnetic beads as described in Materials and Methods. In the first step, JEG NE was subjected to two-step affinity chromatography purification procedure using M1-followed by WT-JRE-Sepharose columns. We found that non-JTF DNA-binding proteins were eluted from the M1-JRE-Sepharose column at lower ionic strength. Eluates containing JTF activity were pooled and repurified using the same approach. A pool of JTF containing eluates from the second WT-JRE-magnetic bead affinity chromatography was fractionated to more than 100 kDa, 30 to approximately 100 kDa, and less than 30-kDa proteins by ultrafiltration through Centricon YM30 and 100 filters (Fig. 9AGo). EMSA with JRE showed virtually all activity was in the 30 to approximately 100-kDa fraction (Fig. 9AGo). Finally, the JTF-containing fraction was purified by Mono-S anion-exchange chromatography using 0.1–1 M NaCl in hypotonic buffer. The fractions were analyzed by EMSA and SDS-PAGE (Fig. 9Go, B and C). Silver staining showed the enrichment of five proteins in fractions containing JTF activity. When these bands were sequenced by mass spectrometry, they were identified as Ku86 (86 kDa), Ku70 (70 kDa), Ku86 variant (69 kDa), and Histone1B (two bands at 30 to ~40 kDa). These proteins are ubiquitously expressed and bind to double-stranded DNA termini. We obtained Ku70 and Ku80 expression vectors and transfected COS-7 cells together with WT JRE-rGHCAT. These proteins had no effect on expression of these constructs, indicating they are not JTF.



View larger version (85K):
[in this window]
[in a new window]
  		Fig. 9. Purification of JTF from JEG-3 NE

JTF was partially purified with JEG NE obtained from 109 cells through a two-step affinity chromatography using M1 mutant and WT JRE-Sepharose and magnetic beads as described in Materials and Methods. A, The partially purified JTF was fractionated into three molecular sizes, which were more than 100 kDa, 30–100 kDa, and less than 30-kDa proteins as described in Materials and Methods. B, The 30- to 100-kDa fraction containing most of the JTF was further resolved by Mono-S anion-exchange chromatography. LD, Loading; FT, flow through; E1–9, eluted fractions during 0.3 M-1 M NaCl gradient. JTF-enriched fractions were identified by EMSA. C, Proteins of the eluted fractions were visualized by silver staining after SDS-PAGE.

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
MATERIALS AND METHODS
 REFERENCES
 
The present studies reveal a complex mechanism for the surprising observation that a gene which is positively regulated in vivo in liver cells can be negatively regulated in a cell type-specific fashion in vitro. The results satisfy rigorous criteria for the presence of a "negative" TRE in the hdio1 5'-FR. There was both stimulation of expression by APO-TR of JRE-containing sequences and a 3- to 6-fold down-regulation of expression when T3 is added (Table 1Go). The negative regulation can be transferred to a heterologous (rGH) promoter by the JRE sequences (Figs. 1Go, 4Go, and 8Go). Nonetheless, this negative regulation does not occur in COS-7 cells in which there is 3-fold increase in hdio1 expression with T3 as was initially demonstrated with a 2.5-kb 5'-FR construct in the original studies of this promoter in HEK-293 cells (Table 1Go) (2). This indicates the negative regulation is cell type specific. The prominent effects of APO-TR and T3 on the 5-kb dio1 construct led to an initial hypothesis that TR bound to hdio1 sequences at approximately –3.4, perhaps in concert with a protein expressed in JEG-3, but not in COS-7, cells. Further studies, however, showed a different explanation was more likely. Footprinting with TR-RXR and NEs from JEG cells confirmed colocalization of binding sites for TR-RXR and a JEG nuclear protein that was not expressed in COS-7, HEK-293, or HeLa cells (Fig. 3Go). The region footprinted in the functionally active JRE sequences was GC rich (Fig. 2Go), and DMS footprinting showed that the binding sites for TR-RXR and JTF had some overlapping and some distinct G residue contact sites (Fig. 4Go).

Given the direct and specific binding of a JEG-3 nuclear protein, we compared effects of the sequences protected by the JEG protein (–3393 to –3359, Figs. 3Go and 4Go) on expression of rGH CAT in JEG-3 and COS-7 cells in the absence and presence of cotransfected TR. Remarkably, these sequences caused a more than 40-fold increase in expression in JEG-3 cells but had no effect in COS-7 cells (Fig. 4Go). This indicated that the JEG nuclear protein acted as a potent, cell type-specific enhancer (JTF) when it bound to the JRE sequences. Of further interest was the observation that the effect of the APO-TR to further increase JTF-rGHCAT expression and the effect of T3 to reduce that expression below the levels present in the absence of TR occurred only in JEG-3 cells (Fig. 4Go).

Because of this important result, we directed our attention toward understanding how APO-TR could synergize with JTF to enhance hdio1 expression and why T3 inhibited its action. Despite the fact that footprinting and EMSA results showed that TR-TR or TR-RXR could bind to the JRE in vitro (Figs. 2–4GoGoGo), coexpression of TR with or without T3 does not change the EMSA pattern of JRE binding (Fig. 3Go). Furthermore, the interaction of labeled JRE with JTF is readily blocked by unlabeled JRE but not by the TREs such as the F2 or malic enzyme elements (Fig. 7Go, A and B). This indicated that JTF does not bind to authentic TREs with high affinity. Furthermore, the binding of high concentrations of TR or RXR-TR to the JRE did not influence the JTF-JRE EMSA either qualitatively or quantitatively (Fig. 7CGo). These results indicated that JTF had a much higher affinity for JRE than TR homo- or heterodimer complexes and that the TR-RXR footprint (Fig. 2Go) probably resulted from the use of high concentrations of bacterially expressed TR and RXR in these experiments. Further studies confirmed that WT, M2, and M4 JRE sequences, all of which bind JTF with high affinity (Fig. 6Go), confer the same TR-dependent enhancement in the absence of T3 and suppression of heterologous promoter expression in the presence of T3 even though the affinity of TR-RXR for M2 and M4 is significantly less than for the intact JRE. Further critical evidence that TR-RXR does not directly affect JRE-driven promoters is that TR, with or without T3, has no effect on expression of the 3XJRE rGH CAT in COS-7 cells (Fig. 4Go). If TR did bind to the JRE, we would have anticipated repression with APO-TR and stimulation by T3 analogous to the effects of TR-T3 on the 5-kb hdio1 5'FR-CAT construct (Table 1Go). This is further evidence that the TR-RXR footprint reflects an in vitro phenomenon that does not occur in vivo. It is compatible with the sequences of the RXR-TR half-sites of the JRE that are not optimal for RXR-TR binding.

Because these in vivo results provided no evidence of direct or indirect TR-JRE interaction to explain the APO-TR effect on a promoter containing JRE, we asked whether DNA binding was required for the effect of TR. Surprisingly, coexpression of either a non-DNA-binding mutant TR{alpha} or TRß2 eliminated both the APO-TR effect and the negative regulation by T3 in JEG cells as was shown previously for the TRH gene (20). It is analogous to more recent results with the TSHß and {alpha} subunit negative regulation that is eliminated both in transient expression assays and that of TSHß in vivo with the same mutation knocked into the TRß1 gene (19). These mutated receptors do not confer T3 regulation to genes containing positive TREs (19).

A recent report shows that a similar E to G mutation at position 71 of TR{alpha}1 receptor does not abrogate the increased expression of the ß-APP gene but does eliminate the repression of that gene by T3 (25). These authors proposed that the positive effects on transcription of the APO-TR on the negatively regulated APP gene promoter occurred via partitioning corepressors and associated HDACs to non-DNA-bound RXR-TR complex, thus permitting its indirect activation. Coexpression of NCoR was reported to enhance this effect consistent with this hypothesis. This is a variation of the squelching model hypothesized for negative T3 regulation by Tagami et al. (17, 27). However, because DNA binding is required for repression of the APP gene, it implies the necessity for TR-DNA contact for the suppressive effect of T3. In the case of the APP gene, the authors postulate the necessity of weak DNA binding of RXR-TR, which interferes with Sp1 binding to sequences in close proximity to a negative TRE. A similar interference by TR with Sp1 binding has been proposed as the explanation for T3-TR suppression of the human epidermal growth factor receptor expression and for attenuation of the positive T3 response to TRE2 of the hdio1 promoter (3, 26).

In addition to Sp1 and JTF, there are a number of examples in which endogenous transcription factors are required for T3-dependent negative regulation of gene expression. These include CTCF, the CCTC-binding factor, which is involved in negative regulation of the rat F1 sequence of the S-2.4 lysozyme transcriptional silencer, the rat TRE-containing genomic DNA element 144, the human c-myc gene enhancer, and perhaps even the APP gene promoter mentioned earlier (23, 29, 30). In these examples, TR-binding sequences are located near, although not adjacent to, CTCF binding sites, and there is no evidence of direct binding of TR or RXR to CTCF such as occurs with T3-TR interference with AP1 activation (29, 31). However, many TRE half-sites in these genes are the high-affinity PuGGTNN sequences, unlike the JRE which has low affinity for RXR-TR (Fig. 7Go) (23). In some of these examples, APO-TR alone as well as TR-T3 may repress gene function rather than increase its expression as does the JTF-JRE-APO-TR combination (22) or the negative TRE in the TSHß and TRH genes (18, 19, 22, 23).

The mechanism for negative regulation by T3 in the present study is not yet clear. The EMSA studies show that TR-RXR can only bind to the JRE when present at high concentrations (Fig. 7CGo), and the transient expression studies show no effect of mutations that decrease RXR-TR affinity on the APO-TR or T3 effect as long as JTF binding is preserved (Figs. 6Go and 8Go). Furthermore, there is no effect of transient TR expression on the EMSA in the JEG-NE (Fig. 3Go) or on the 3XJRE-rGHCAT expression in COS-7 cells. The squelching model does not explain these observations because elimination of the DNA binding blocks TR effects. These P box substitutions have been shown to preserve corepressor binding and transactivation properties of the TRß2 (19). This is supported by the absence of an effect of coexpressed NCoR on the APO-TR transcriptional enhancement. The enhancement of TR-JRE binding by an identified protein expressed in JEG, but not COS cells, cannot be ruled out although we have no direct evidence of this. The fact that RXR{alpha} is more highly expressed in JEG than in COS cells, and that RXR-TR binds to JRE with high affinity (Fig. 7CGo) could contribute to such a possibility, especially in the presence of T3, which enhances the stability of RXR-TR complexes (32, 33, 34). Because the TR-binding sequences overlap those for JTF (Fig. 5Go) and even a high concentration of RXR-TR does not interfere with JTF binding (Fig. 7CGo), it seems unlikely that JTF-JRE interaction could facilitate RXR-TR binding to JRE. Another possibility is that RXR-TR complexes are binding with high affinity to unidentified TREs in the JEG cell genome, thus partitioning corepressors or coactivators away from the JTF-JRE complexes in the absence and presence of T3, respectively. This would imply that DNA-bound RXR-TR complexes are much more effective than the soluble RXR-TR complexes thought to be involved in the squelching model (14, 17). Many of these possible explanations can be better addressed once JTF is cloned.

To date, our attempts to identify JTF have been unsuccessful. Our initial approach was to take advantage of DNA-affinity columns using inactive JRE sequences such as M1 coupled to Sepharose to clear the NE of irrelevant DNA binding proteins. This was followed by chromatography with WT JRE-Sepharose conjugates to isolate JTF. We were encouraged by the results using this approach because the elution of JRE binding activity occurred at a significantly higher molarity than did that to the M1-Sepharose conjugates. This allowed partial purification and determination of an approximate molecular mass range of 30–100 kDa for JTF protein (Fig. 9Go). However, the purified active fraction still contained four to five visible bands on silver-stained gels. Mass spectrometry of these proteins showed they were Ku proteins and one histone subunit (Fig. 9Go). Both direct testing of a Ku-expressing constructs as well as the fact that the Ku proteins are ubiquitously expressed DNA-binding proteins indicate that these are not JTF (35, 36, 37, 38, 39). A different approach will be required to identify this protein despite its high DNA binding affinity.

The negative regulation of the hdio1 gene by T3-TR and the presence of JTF-like activity in placenta (Fig. 6Go), suggests that D1 activity might be expressed in the cyto- or syncytiotrophoblasts of the human placenta. If thyroid hormone levels were reduced, this could lead to a paradoxical increase in expression of D1. Under normal circumstances, D1 is not expressed in human placenta even though there is high expression of type 3 deiodinase and low expression of the type 2 deiodinase (40, 41), indicating that all of the elements required for selenoprotein synthesis are present in these cells.

A computer search showed that exact copies of the core JRE sequences –3382 CCCCCTCAGGCGC –3369 are present in 22 mammalian and 27 other eukaryotic genes. In addition, the same sequences were identified in human DNA in several chromosomes (nos. 1, 6, 9, 11, 13, 19, and 22). This suggests that JTF could be present and regulate JRE-containing genes in other cells. Confirmation of this will require successful cloning of this protein.

In summary, we have demonstrated promoter-independent, TR-T3-dependent negative regulation of a specific gene in which the negative TRE may not bind TR in vivo but can in vitro. The negative effect requires an as yet unidentified JEG cell transcription factor, JTF. The fact that TR-DNA binding is required for the effect but that the JRE does not interact with TR in other cells suggests that binding of transiently expressed APO-TRs to other unidentified TREs in these cells could be important given the high potency of JTF as a transcription factor. There are a number of other possibilities for this effect, which will require further studies once the identity of JTF is known. Whatever the mechanism, the present results indicate that it does not fit any of the currently proposed models for T3-dependent negative regulation.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
REFERENCES
 
Cell Culture
JEG-3, COS-7, HeLa, and HEK-293 cells were maintained in Ham’s F10 medium containing 14 mM sodium bicarbonate (pH 7.4) and 2 mM L-glutamine supplemented with 10% fetal bovine serum (FBS, Life Technologies, Gaithersburg, MD) until cells were grown to about 70–80% confluence (42). For the stimulation with T3, the culture medium was removed from cells, the cells were rinsed with PBS once, and Ham’s F10 medium containing charcoal-stripped 10% FBS was added. The medium was changed after 1 d, and the incubation continued for another 2 d. T3 (50 nM) was diluted in Ham’s F10 medium+charcoal-stripped 10% FBS and added to cells for the indicated times.

Plasmid Constructions and Mutagenesis
Standard techniques were used for all plasmid constructions (43). 5'FR genomic fragments (5, 3.7, 3.5, 3.3, 2.5 and 0.8 kb) were amplified by PCR and cloned into BamHI site of plasmid pOCAT2, a promoter insertion chloramphenicol acetyltransferase (CAT) expression vector (44). These were designated –5 kb, –3.7 kb, –3.5 kb, –3.3 kb, –2.5 kb, and –0.8 kb 5' FR hdio1-CAT. To insert hdio1 fragment to the heterologous promoter-CAT, 137 bp basal rGH promoter-pOCAT plasmid, PCR-amplified 200-bp, 100-bp, and 33-bp DNA fragments from –3.5 kb 5'FR hdio1-CAT were cloned into the BamHI site of the 137-bp rGH promoter-pOCAT plasmid (137 bp rGH-CAT) (45), which were designated 200 bp, 100 bp, and 33 bp hdio1-137 bp rGH-CAT. Point mutations of core sequences in the 33 bp JRE were generated by synthesizing top-strand and bottom-strand oligos and hybridizing two oligos to make double-stranded fragments. Three copies of 33-bp WT or mutated sequences (M1–M6) were inserted into 137-bp rGHCAT. To generate a mTRa1 mutant, GS71, two bases at 614 (A to G) converting amino acid 71 from E to G and at 616 converting amino acid from G to S from TSS of mTRa1 cDNA-CDM expression vector were mutated by using QuikChange Site-Directed Mutagenesis kit (Stratagene, La Jolla, CA) according to manufacturer’s protocol. All constructs were confirmed by sequencing and restriction digestion mapping. GS125 containing the same mutation in hTRb2 was kindly provided in a pSG5 expression vector by Dr. F. Wondisford (19).

Transient Transfections and CAT Assay
Transfections were performed as previously described (42, 46) with FuGENE 6 Transfection Reagent (Roche Clinical Laboratories, Indianapolis, IN) in JEG-3, COS-7, HeLa, and HEK293 cells. Each plate was transfected with 2 µg CAT reporter plasmids and 0.5 µg TKGH, which constitutively expresses human GH (hGH) as an transfection efficiency control. CDM8 vector (0.5 µg) expressing mouse TRa1 (WT or mutant) or a pSG5 vector expressing hTRß2 was cotransfected. As a control, CDM8 and pSG5 empty vectors were used. After transfection, cells were incubated in 60-mm culture dishes in the absence of thyroid hormone for 2 d and then exposed to 50 nm T3 for 24 h at 37 C. CAT activity was determined by a phase extraction procedure with minor modifications as described previously (47). CAT expression was normalized to hGH expression in the absence and presence of T3 to calculate T3 responsiveness. Transfection data are the means ± SD of a minimum of triplicate samples.

Isolation of NE
NE was isolated according to a standard protocol (43). Briefly, JEG-3, COS-7, HeLa, and 293-HEK cells were cultured in twenty 100-mm dishes in the presence or absence of 50 nM T3. Cells were harvested into 50-ml Conical tubes, rinsed with cold PBS and cold hypotonic solution (10 mM HEPES, pH 7.9; 1.5 mM MgCl2; 10 mM KCl) with protease inhibitors. Cells were then dispersed in 3 volumes of hypotonic solution containing protease inhibitors, incubated on ice for 10 min, and then homogenized. Nuclei were isolated by removing the cytoplasmic fraction after centrifugation at 1850 x g for 15 min at 4C. NE was isolated from nuclei in 300 mM KCl by incubating for 30 min at 4 C and centrifuging at 25,000 x g for 30 min. NE from placenta tissue was isolated in the same way except for a minor modification at the homogenization step. In this case, 1 g placenta tissue was minced into small pieces, which were incubated in hypotonic buffer on ice for 10 min and then homogenized for NE preparation.

EMSA
EMSA was performed as described previously (3). Double-strand oligonucleotides of WT and mutant JREs were radiolabeled with [32P]dCTP (New England Nuclear, Boston, MA) by fill-in reaction using Klenow DNA polymerase and gel purified. Chicken TR{alpha}1 and human RXR{alpha} were over-expressed in Escherichia coli and purified as described previously (48). NEs were prepared from JEG-3, COS-7, HeLa, and HEK293 mammalian cells as described above. Radiolabeled probes were reacted with bacterially synthesized cTR{alpha}, cTR{alpha}+hRXR{alpha} or NEs and were resolved on a nondenaturing PAGE.

In Vitro DNAse I Footprinting
The DNAse I digestion reaction was performed according to standard protocol with minor modifications (43). Briefly, 32P-labeled 54-bp DNA fragment (–3405 to ~–3352) was prepared by PCR using 32P-labeled 5'-end primer at –3405 (5'-GAATATATGCTTTTCCTGTTTT) and unlabeled 3'-end primer at –3352 (5'-TGAAAGCTATATGAAAGGGC). Gel-purified probe was reacted with cTR{alpha}, cTR{alpha}+hRXR{alpha} or NEs in EMSA reaction condition as described above, which were then exposed to 20 U of DNAse I (Worthington Biochemical Corp., Freehold, NJ) for 2 min at room temperature. The reaction was stopped by addition of stop solution and DNA precipitated in dry-ice ethanol. Protection sites were analyzed by denatured sequencing gel electrophoresis.

Methylation Interference Assay
DMS methylation interference assay was performed using a standard protocol as described previously (2).

Purification
Three copies of 33-bp WT or mutant (M1) JRE sequences were amplified from WT and M1 JRE-rGHCAT plasmids by PCR using biotinylated specific primers. The amplified DNA were conjugated to streptavidin-Sepharose (Sigma Chemical Co., St. Louis, MO) or streptavidin-magnetic beads (DynAl, Great Neck, NY) according to the manufacturer’s protocol. JEG NEs from 109 JEG-3 cells were diluted to 0.1 M NaCl in hypotonic buffer, which was partially purified by M1 mutant and WT JRE-Sepharose affinity chromatography. M1 mutant JRE-Sepharose column was first used to remove nonspecifically binding proteins during the affinity chromatography. JEG NE was passed through M! JRE-Sepharose column, which were subsequently loaded into WT JRE-Sepharose column. Nonspecific weak binding proteins were eluated by 0.3 M NaCl whereas specific binding proteins by 0.3–1 M NaCl step-gradient. JTF-enriched fractions were identified by EMSA using 32P-labeled JRE. JTF-enriched fractions were pooled and purified again using M1-mutant and WT JRE magnetic bead affinity chromatography. Once more, nontarget DNA-binding proteins in the partially purified eluates were removed by incubating the partially purified proteins with M1 mutant coupled to magnetic beads. Then JTF was bound to WT JRE-magnetic beads by incubating 1 h at 4C with gentle shaking. JTF was again eluted by 0.3–1M NaCl step gradient. JTF fractions were pooled and used to fractionate protein into different sizes, more than100 kDa, 30–100 kDa, and less than 30 kDa by ultrafiltration using Centricon YM-30 and -100 (Millipore Corp., Bedford, MA). The JTF-containing fraction (30–100 kDa) was finally purified by Mono-S anion-exchange chromatography and resolved using 0.1–1 M NaCl salt gradient. The most active fractions were identified by EMSA, and purified proteins were visualized by SDS-PAGE and silver staining.

Miscellaneous
All chemicals were of reagent grade and obtained from commercial sources. All molecular biological manipulations were performed using standard techniques (10). Human tissues were obtained under an Institutional Review Board (IRB)-approved protocol.


   ACKNOWLEDGMENTS
 
We thank Dr. Ann Marie Zavacki for a gift of TR{alpha} and -ß-expressing bacterial clones, Dr. Fredric E. Wondisford for a TRß2-expressing plasmid, Dr. Herbert Samuels for a chick TRa1 bacterial expression, and Dr. Michael G. Rosenfeld for NCoR expression vectors.


   FOOTNOTES
 
This work was supported by National Institutes of Health (NIH) Grant DK 44128, NIH KO1 Grant DK02867, and NIH Training Grant DK07529.

Abbreviations: APP, amyloid precursor protein; CAT, chloramphenicol acetyltransferase; DMS, dimethyl sulfide; FBS, fetal bovine serum; 5'-FR, 5'-flanking region; HDAC, histone deacetylase; hdio1, human type 1 iodothyronine deiodinase gene; HEK, human embryonic kidney; NCoR, nuclear receptor corepressor; NE, nuclear extract; JRE, JTF-responsive element; JTF, JEG cell-specific transcription factor; RXR, retinoid X receptor; TR, thyroid hormone receptor; TRE, thyroid responsive element; TSS, transcription start site; WT, wild-type.

Received for publication June 24, 2004. Accepted for publication August 9, 2004.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 

  1. Bianco AC, Salvatore D, Gereben B, Berry MJ, Larsen PR 2002 Biochemistry, cellular and molecular biology and physiological roles of the iodothyronine selenodeiodinases. Endocr Rev 23:38–89[Abstract/Free Full Text]
  2. Toyoda N, Zavacki AM, Maia AL, Harney JW, Larsen PR 1995 A novel retinoid X receptor-independent thyroid hormone response element is present in the human type 1 deiodinase gene. Mol Cell Biol 15:5100–5112[Abstract]
  3. Zhang C, Kim S, Harney JW, Larsen PR 1998 Further characterization of thyroid hormone response elements in the human type 1 iodothyronine deiodinase gene. Endocrinology 139:1156–1163[Abstract/Free Full Text]
  4. Horlein AJ, Naar AM, Heinzel T, Torchia J, Gloss B, Kurokawa R, Ryan A, Kamei Y, Soderstrom M, Glass CK 1995 Ligand-independent repression by the thyroid hormone receptor mediated by a nuclear receptor co-repressor. Nature 377:397–404[CrossRef][Medline]
  5. Heinzel T, Lavinsky RM, Mullen TM, Soderstrom M, Laherty CD, Torchia J, Yang WM, Brard G, Ngo SD, Davie JR, Seto E, Eisenman RN, Rose DW, Glass CK, Rosenfeld MG 1997 A complex containing N-CoR, mSin3 and histone deacetylase mediates transcriptional repression. Nature 387:43–48[CrossRef][Medline]
  6. Chen JD, Evans RM 1995 A transcriptional co-repressor that interacts with nuclear hormone receptors. Nature 377:454–457[CrossRef][Medline]
  7. Kao HY, Downes M, Ordentlich P, Evans RM 2000 Isolation of a novel histone deacetylase reveals that class I and class II deacetylases promote SMRT-mediated repression. Genes Dev 14:55–66[Abstract/Free Full Text]
  8. Nagy L, Kao HY, Chakravarti D, Lin RJ, Hassig CA, Ayer DA, Schrieber SL, Evans RM 1997 Nuclear receptor repression mediated by a complex containing SMRT, mSin3A, and histone deacetylase. Cell 89:373–380[CrossRef][Medline]
  9. Huang EY, Zhang J, Miska EA, Guenther MG, Kouzarides T, Lazar MA 2000 Nuclear receptor corepressors partner with class II histone deacetylases in a Sin3-independent repression pathway. Genes Dev 14:45–54[Abstract/Free Full Text]
  10. Ogryzko VV, Schiltz RL, Russanova V, Howard BH, Nakatani Y 1996 The transcriptional coactivators p300 and CBP are histone acetyltransferases. Cell 87:953–959[CrossRef][Medline]
  11. Korzus E, Torchia J, Rose DW, Xu L, Kurokawa R, McInerney EM, Mullen TM, Glass CK, Rosenfeld MG 1998 Transcription factor-specific requirements for coactivators and their acetyltransferase functions. Science 279:703–707[Abstract/Free Full Text]
  12. Blanco JC, Minucci S, Lu J, Yang XJ, Walker KK, Chen H, Evans RM, Nakatani Y, Ozato K 1998 The histone acetylase PCAF is a nuclear receptor coactivator. Genes Dev 12:1638–1651[Abstract/Free Full Text]
  13. Hassig CA, Tong JK, Fleischer TC, Owa T, Grable PG, Ayer DE, Schreiber SL 1998 A role for histone deacetylase activity in HDAC1-mediated transcriptional repression. Proc Natl Acad Sci USA 95:3519–3524[Abstract/Free Full Text]
  14. Lazar MA 2003 Thyroid hormone action: a binding contract. J Clin Invest 112:497–499[CrossRef][Medline]
  15. Wondisford FE, Radovick S, Moates JM, Usala SJ, Weintraub BD 1988 Isolation and characterization of the human thyrotropin ß-subunit gene: differences in gene structure and promoter function from murine species. J Biol Chem 263:12538–12542[Abstract/Free Full Text]
  16. Carr FE, Kaseem LL, Wong NCW 1992 Thryoid hormone inhibits thyrotropin gene expression via a position-independent negative L-triiodothyronine-responsive element. J Biol Chem 267:18689–18694[Abstract/Free Full Text]
  17. Tagami T, Park Y, Jameson JL 1999 Mechanisms that mediate negative regulation of the thyroid-stimulating hormone {alpha} gene by the thyroid hormone receptor. J Biol Chem 274:22345–22353[Abstract/Free Full Text]
  18. Shibusawa N, Hashimoto K, Nikrodhanond AA, Liberman MC, Applebury ML, Liao XH, Robbins JT, Refetoff S, Cohen RN, Wondisford FE 2003 Thyroid hormone action in the absence of thyroid hormone receptor DNA-binding in vivo. J Clin Invest 112:588–597[CrossRef][Medline]
  19. Shibusawa N, Hollenberg AN, Wondisford FE 2003 Thyroid hormone receptor DNA binding is required for both positive and negative gene regulation. J Biol Chem 278:732–738[Abstract/Free Full Text]
  20. Takeda T, Nagasawa T, Miyamoto T, Hashizume K, DeGroot LJ 1997 The function of retinoid X receptors on negative thyroid hormone response elements. Mol Cell Endocrinol 128:85–96[CrossRef][Medline]
  21. Sasaki S, Lesoon-Wood LA, Dey A, Myers FA, Thorne AW, Crane-Robinson C, Bonifer C, Filippova GN, Lobanenkov V, Renkawitz R 1999 Ligand-induced recruitment of a histone deacetylase in the negative-feedback regulation of the thyrotropin ß gene. EMBO J 18:5389–5398[CrossRef][Medline]
  22. Lutz M, Baniahmad A, Renkawitz R 2000 Modulation of thyroid hormone receptor silencing function by co-repressors and a synergizing transcription factor. Biochem Soc Trans 28:386–389[Medline]
  23. Lutz M, Burke LJ, LeFevre P, Myers FA, Thorne AW, Crane-Robinson C, Bonifer C, Filippova GN, Lobanenkov V, Renkawitz R 2003 Thyroid hormone-regulated enhancer blocking: cooperation of CTCF and thyroid hormone receptor. EMBO J 22:1579–1587[CrossRef][Medline]
  24. Perez-Juste G, Garcia-Silva S, Aranda A 2000 An element in the region responsible for premature termination of transcription mediates repression of c-myc gene expression by thyroid hormone in neuroblastoma cells. J Biol Chem 275:1307–1314[Abstract/Free Full Text]
  25. Villa A, Santiago J, Belandia B, Pascual A 2004 A response unit in the first exon of the ß-amyloid precursor protein gene containing thyroid hormone receptor and Sp1 binding sites mediates negative regulation by 3,5,3'-triiodothyronine. Mol Endocrinol 18:863–873[Abstract/Free Full Text]
  26. Xu J, Thompson KL, Shephard LB, Hudson LG, Gill GN 1993 T3 receptor suppression of Sp1-dependent transcription from epidermal growth factor receptor promoter via overlapping DNA-binding sites. J Biol Chem 268:16065–16073[Abstract/Free Full Text]
  27. Tagami T, Madison LD, Nagaya T, Jameson JL 1997 Nuclear receptor corepressors activate rather than suppress basal transcription of genes that are negatively regulated by thyroid hormone. Mol Cell Biol 17:2642–2648[Abstract]
  28. Lopez G, Schaufele F, Webb P, Holloway JM, Baxter JD, Kushner PJ 1993 Positive and negative modulation of Jun action by thyroid hormone receptor at a unique AP1 site. Mol Cell Biol 13:3042–3049[Abstract/Free Full Text]
  29. Awad TA, Bigler J, Ulmer JE, Hu YJ, Moore JM, Lutz M, Neiman PE, Collins SJ, Renkawitz R, Lobanenkov VV, Filippova GN 1999 Negative transcriptional regulation mediated by thyroid hormone response element 144 requires binding of the multivalent factor CTCF to a novel target DNA sequence. J Biol Chem 274:27092–27098[Abstract/Free Full Text]
  30. Bigler J, Eisenman RN 1994 Isolation of a thyroid hormone-responsive gene by immunoprecipitation of thyroid hormone receptor-DNA complexes. Mol Cell Biol 14:7621–7632[Abstract/Free Full Text]
  31. Zhang XK, Wills KN, Husmann M, Hermann T, Pfahl M 1991 Novel pathway for thyroid hormone receptor action through interaction with jun and fos oncogene activities. Mol Cell Biol 11:6016–6025[Abstract/Free Full Text]
  32. Hsu J-H, Zavacki AM, Harney JW, Brent GA 1995 Retinoid-X-receptor (RXR) differentially augments thyroid hormone response in cell lines as a function of the response element and endogenous RXR content. Endocrinology 136:421–430[Abstract]
  33. Zhang J, Zamir I, Lazar MA 1997 Differential recognition of liganded and unliganded thyroid hormone receptor by retinoid X receptor regulates transcriptional repression. Mol Cell Biol 17:6887–6897[Abstract]
  34. Collingwood TN, Butler A, Tone Y, Clifton-Bligh RJ, Parker MG, Chatterjee VK 1997 Thyroid hormone-mediated enhancement of heterodimer formation between thyroid hormone receptor ß and retinoid X receptor. J Biol Chem 272:13060–13065[Abstract/Free Full Text]
  35. Chan DW, Ye R, Veillette CJ, Lees-Miller SP 1999 DNA-dependent protein kinase phosphorylation sites in Ku 70/80 heterodimer. Biochemistry 38:1819–1828[CrossRef][Medline]
  36. Mimori T, Hardin JA, Steitz JA 1986 Characterization of the DNA-binding protein antigen Ku recognized by autoantibodies from patients with rheumatic disorders. J Biol Chem 261:2274–2278[Abstract/Free Full Text]
  37. Blier PR, Griffith AJ, Craft J, Hardin JA 1993 Binding of Ku protein to DNA. Measurement of affinity for ends and demonstration of binding to nicks. J Biol Chem 268:7594–7601[Abstract/Free Full Text]
  38. Falzon M, Fewell JW, Kuff EL 1993 EBP-80, a transcription factor closely resembling the human autoantigen Ku, recognizes single- to double-strand transitions in DNA. J Biol Chem 268:10546–10552[Abstract/Free Full Text]
  39. Cary RB, Peterson SR, Wang J, Bear DG, Bradbury EM, Chen DJ 1997 DNA looping by Ku and the DNA-dependent protein kinase. Proc Natl Acad Sci USA 94:4267–4272[Abstract/Free Full Text]
  40. Huang SA, Dorfman DM, Genest DR, Salvatore D, Larsen PR 2003 Type 3 iodothyronine deiodinase is highly expressed in the human uteroplacental unit and in fetal epithelium. J Clin Endocrinol Metab 88:1384–1388[Abstract/Free Full Text]
  41. Kaplan MM, Shaw EA 1984 Type II iodothyronine 5'-deiodination by human and rat placenta in vitro. J Clin Endocrinol Metab 59:253–257[Abstract]
  42. Kim SW, Ahn IM, Larsen PR 1996 In vivo genomic footprinting of thyroid hormone-responsive genes in pituitary tumor cell lines. Mol Cell Biol 16:4465–4477[Abstract]
  43. Ausubel FM, Brent R, Kingston RE, Moore DD, Seidman JG, Smith JA, Struhl K 1992 Ligation-mediated PCR for genomic sequencing and footprinting. In: Current protocols in molecular biology. New York: Wiley and Sons, Inc.; 15.5.1–15.6.10
  44. Prost E, Moore DD 1986 CAT vectors for analysis of eukaryotic promoters and enhancers. Gene 45:107–111[CrossRef][Medline]
  45. Brent GA, Larsen PR, Harney JW, Koenig RJ, Moore DD 1989 Functional characterization of the rat growth hormone promoter elements required for induction by thyroid hormone with and without a co-transfected ß type thyroid hormone receptor. J Biol Chem 264:178–182[Abstract/Free Full Text]
  46. Bartha T, Kim SW, Salvatore D, Gereben B, Tu HM, Harney JW, Rudas P, Larsen PR 2000 Characterization of the 5'-flanking and 5'-untranslated regions of the cyclic adenosine 3',5'-monophosphate-responsive human type 2 iodothyronine deiodinase gene. Endocrinology 141:229–237[Abstract/Free Full Text]
  47. Seed B, Sheen J-Y 1988 A simple phase-extraction assay for chloramphenicol acetyltransferase activity. Gene 67:271–277[CrossRef][Medline]
  48. Zavacki AM, Zhang CY, Harney JW, Larsen PR 1996 Enhancement of thyroid hormone receptor isoform specificity by insertion of a distant half-site into a thyroid hormone response element. Endocrinology 137:1438–1446[Abstract]

NURSA Molecule Pages Link:

Nuclear Receptors:   TRα  |  RXRα
Coregulators:   NCOR
Ligands:   Thyroid hormone



This article has been cited by other articles:


Home page
 J Mol EndocrinolHome page
A. Wulf, M. G Wetzel, M. Kebenko, M. Kroger, A. Harneit, J. Merz, and J. M Weitzel
The role of thyroid hormone receptor DNA binding in negative thyroid hormone-mediated gene transcription
J. Mol. Endocrinol., July 1, 2008; 41(1): 25 - 34.
[Abstract] [Full Text] [PDF]

Home page
 J Mol EndocrinolHome page
M. Nygard, N. Becker, B. Demeneix, K. Pettersson, and M. Bondesson
Thyroid hormone-mediated negative transcriptional regulation of Necdin expression.
J. Mol. Endocrinol., June 1, 2006; 36(3): 517 - 530.