| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ARTICLE |
Department of Biochemistry, Chang Gung University School of Medicine, Kweisan Taoyuan 333, Taiwan, Republic of China
Address all correspondence and requests for reprints to: Kwang Huei Lin, Ph.D., Department of Biochemistry, Chang Gung University School of Medicine, 259 Wen-hwa First Road, Kweisan Taoyuan 333, Taiwan, Republic of China. E-mail: khlin{at}mail.cgu.edu.tw.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
After binding to DNA, either alone or together with retinoic X receptor (RXR), TRs can recruit transcriptional cofactors (corepressors and coactivators) to the promoter to facilitate transcriptional regulation of the target genes. In the absence of a cognate ligand, corepressors, such as silence mediator of RXR and TR, nuclear receptor corepressor, histone deacetylase and mSin3, are recruited to the promoter by TRs to facilitate gene repression. Alternatively, coactivators, such as steroid receptor coactivator (SRC), cAMP response element-binding protein-binding protein, and p300 and cAMP response element-binding protein-binding protein-associated factor, are tethered to transcriptional complexes to coactivate target gene expression (4, 5, 6, 7, 8).
Down-regulation of receptors after ligand binding is a mechanism conserved among nuclear receptors and other members of this superfamily to modulate the cellular hormone response. This mechanism has been shown to function transcriptionally and posttranslationally, and thus receptor signals are reduced at the mRNA and protein levels. Previous work has shown that the expression levels of most TRs in cells, except for TR
2 that lacks hormone-binding ability, can be reduced by treatment with the thyroid hormone T3 (9). The transcriptional mechanisms involved in turning off their expression have remained elusive. However, recent findings have revealed that the phosphorylation status of TRs plays a pivotal role in deciding their protein stability. Inhibition of protein phosphatase 1 and 2A with okadeic acid (OA) not only significantly potentiates the transcriptional activation ability of TRs, but also increases their protein stability (10, 11). Participation of ubiquitin proteasome in TR degradation has also been suggested. Although several candidate kinases have been proposed, to date the kinases that directly phosphorylate TR in the nucleus in vivo have not been identified.
Among signaling pathways identified in mammalian cells, those involving MAPKs are major mediators in the response to extracellular stimuli. This mediation is carried out through their ability to couple signal transduction and activation of transcription factors, thus triggering specific gene expression programs (12). Three MAPK pathways, including the extracellular signal-regulated kinases (ERK1 and -2), the Jun-N terminal kinases (JNK1, -2, and -3), and the p38 isoforms (, ß, 
, and 
) have been characterized (13). All MAPKs are highly conserved serine/threonine kinases, activated through phosphorylation of a T-X-Y motif. This activation is performed by upstream MAPK kinases, a family of conserved dual specificity kinases that are, in turn, activated by MAPK kinase (MKK) kinases. In general, each group of MAPKs is activated by two homologous MKKs. MEK1 and -2 activate the ERKs, JNK kinases 1 and 2 (JNKK1 and -2 or MKK4 and -7) activate the JNKs, and MKK3 and -6 activate the p38s (14, 15).
The activities of transcription cofactors, such as p300 and silence mediator of RXR and TR, are also modulated by MAPKs through posttranslational modification. Their interaction with transcription factors, such as nuclear hormone receptors, is also regulated by direct participation of MAPKs (16, 17, 18, 19). Recently, a stress-induced yeast MAPK, Hog1, was found to be an integral partner of the transcription activation complexes (20). This discovery has dramatically changed the traditional view that MAPKs regulate the activities of transcription factors by posttranslational modification before or after they engage in transcriptional activation. This finding suggests that MAPKs may participate in transcriptional regulation more actively than previously thought by directly participating in the activation.
In this study we found that MAPK pathways were involved in the stabilization of TR isoform proteins in both the presence and absence of ligand. Cotransfection of MAPK activators with TR potentiates its activity and increases its protein stability. MAPK inhibitors reduced the transcriptional activity of TR and its protein stability when used to treat CV-1 cells transiently transfected with TR expression vectors. This was replicated in HepG2 cells stably transfected with TR expression vectors. Our results suggest that MAPKs are among the major kinases engaged in the nuclear phosphorylation of TR and consequently play a critical role in its transcriptional activities and protein stability.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
1 and TRß1 cDNAs were subcloned from a parental PCLC vector into HindIII and NotI/XhoI sites of pCDNA3.0 (Invitrogen, San Diego, CA), respectively, to make a mammalian expression vector driven by the cytomegalovirus (CMV) promoter. Galactosidase (GAL)-TR1 and -ß1 fusion proteins were created by inserting cDNA into the HindIII and inactivated EcoRI sites of pM vector (CLONTECH Laboratories, Inc., Palo Alto, CA), respectively. Deletion mutants of GAL-TRß1 fusion proteins were created either by internal restriction site deletion (SacI or SmaI) or by subcloning from previous constructs (JL05, JL06, and JL08) (21). Expression vectors for MKK3 and MKK6 were gifts from Dr. Jiahuai Han (The Scripps Clinic, La Jolla, CA). Expression vectors for MAPKs were provided by Dr. Roger Davis (University of Massachusetts Medical School, Boston, MA). Enhanced green fluorescent protein-TR1 was created by inserting cDNA into the HindIII site of an enhanced green fluorescence protein promoter vector (CLONTECH Laboratories, Inc.). Reporter constructs, in which luciferase expression is driven by either TRE- or GAL-DNA-binding domain (DBD)-binding sites, have been described previously (22). pcDNA 3.1 GRIP-1 (glucocorticoid receptor interacting protein-1) was provided by Dr. George Muscat (University of Queensland, Brisbane, Australia). VP16 (herpes simplex virus protein 16) and pM-GRIP-1 were gifts from Dr. M. Stallcup (University of Southern California, Los Angeles, CA).
Western blot
Aliquots of cell lysate were resolved on 10% SDS-PAGE gels and blotted onto nitrocellulose paper (Amersham Pharmacia Biotech, Piscataway, NJ) on a semidry blotter (C.B.S. Scientific, Del Mar, CA) at 300 mA for 30 min. After extensive washing with PBS with 0.5% Tween 20 (PBST), the blot was preincubated with 5% nonfat milk in PBST at 37 C for 1 h, followed by primary antibody in blocking solution at 4 C overnight. The blot was washed with PBST, followed by incubation with secondary antibody conjugated with horseradish peroxidase in blocking solution at room temperature for 1 h. The signal was detected with a chemiluminescence kit (Amersham Pharmacia Biotech) and visualized on x-ray film. Monoclonal antibodies (J52 and C4) against TR were gifts from Dr. Sheue Yann Cheng (National Cancer Institute, Bethesda, MD). The dilutions of the antibodies are indicated in the relevant figure legends.
Transfection
Transient transfections were performed by mixing aliquots of plasmid DNA together in 1x HEPES buffer [20 mM HEPES (pH 7.0),187 mM NaCl, 5 mM KCl, 0.7 mM Na2HPO4, and 5.5 mM dextrose]. Mixtures of liposome (Lipofectamine, Life Technologies, Inc., Gaithersburg, MD) in 1x HEPES buffer were added to the DNA mixture and incubated at room temperature for 5–15 min, allowing the DNA/liposome complex to form. Aliquots of culture medium were then added to each tube and mixed. Medium containing the DNA/liposome complex was transferred to CV-1 cells in triplicate. The transfection was allowed to proceed overnight before the medium was replaced. Cells were harvested and assayed for luciferase activity 16–24 h after transfection as described previously (5).
EMSA
32P-Labeled F2 oligonucleotide corresponding to the TRE in the chicken lysozyme gene, consisting of two inverted repeats of the half-site binding motif (AGGTCA) separated by six nucleotides, was made by recessive end-filling reaction mediated by Klenow fragment (New England Biolabs, Inc., Beverly, MA). The 32P-labeled probe was further purified by gel purification in 15% acrylamide buffered by 1x 90 mM Tris-borate/1 mM EDTA.
Total lysate (5 µl) from CV-1 cells transiently transfected with vectors expressing TRß1, MKK6EE in the absence or presence of T3 (100 nM), MAPK inhibitors, and lactacystin (a proteasome inhibitor) was used to bind F2 probe in 20 µl binding buffer [25 mM HEPES (pH 7.4), 5 mM MgCl2, 4 mM EDTA, 2 mM dithiothreitol (DTT), 110 mM NaCl, 5 µg/ml BSA, and 0.8% Ficoll] at room temperature for 30 min. Protein and DNA complexes were resolved on a 5% acrylamide low ionic gel containing 5% glycerol at 4 C for at least 3.5 h. Signals on the gels were viewed by autoradiography. J52 antibody (0.1 µg) against TRß1 was included in the binding reaction to verify the TR-containing bands on the gel. Heterodimerization with RXR was tested with the addition of 2 µl in vitro translated RXR. Similar results were observed in at least three independent experiments. No additional hormone, MAPK inhibitor, or lactacystin was included in the EMSA assay.
Glutathione-S-transferase (GST) pulldown
GST and GST fusion proteins were expressed in Escherichia coli strain BL21 and purified using glutathione-agarose (Amersham Pharmacia Biotech) affinity chromatography. The GST fusion proteins were analyzed on 10% SDS-PAGE gels to examine their integrity and to normalize the protein levels. Proteins labeled with [35S]methionine/cystein were produced using the transcription/translation (TNT)-coupled system (Promega Corp., Madison, WI). In vitro binding assays were performed with 3–6 µl 35S-labeled proteins and 200–300 ng GST fusion proteins adsorbed on the glutathione-agarose beads in 300 µl HEMG buffer [40 mM HEPES (pH 7.8), 0.2 mM EDTA, 100 mM KCl, 0.1% Nonidet P-40, 1.5 mM DTT, and 10% glycerol] freshly supplemented with 1% BSA and protease inhibitors. The reaction was allowed to proceed for 2 h at room temperature or overnight at 4 C with gentle rocking. The affinity beads were collected by centrifugation and washed with 1 ml HEMG buffer. The wash step was repeated three times, and the beads were dried before the addition of 2x Laemmli sample buffer to elute bound proteins. Eluted proteins were resolved on 10% SDS-PAGE gels, and signals were visualized with autoradiography.
Kinase assay
The protein kinase assay was performed following the protocol described by Raingeaud et al. (23). Either in vitro expressed protein from a TNT-coupled system or recombinant protein expressed in E. coli was used in this assay. 35S-Labeled p38 was kinased in the presence and absence of MKK6EE in 40 µl kinase buffer [25 mM HEPES (pH 7.4), 25 mM ß-glycerophosphate, 25 mM MgCl2, 0.5 mM DTT, and 0.1 mM sodium orthovanadate] supplemented with 50 µM ATP at room temperature for 30–60 min. The products from kinase reaction were directly used in GST pulldown reactions.
Expression and purification of recombinant TR protein were performed as described previously (11). Activated p38 proteins were isolated from confluent CV-1 cells by immunoprecipitation using a polyclonal antibody against p38 (Cell Signaling Technology, Beverly, MA) in 500 µl buffer A [20 mM Tris (pH 7.5), 10% glycerol, 1% Triton X-100, 137 mM NaCl, 2 mM EDTA, 25 mM ß-glycerophosphate, 0.5 mM DTT, and 0.1 mM sodium orthovanadate] as described previously (23). Recombinant TR protein and p38 immunoprecipitation complexes were incubated in the presence of 20 µM ATP and 10 µCi [-32P]ATP in 40 µl kinase buffer at room temperature for 40 min. Then, the p38 immunocomplex was removed from the reaction mixture by centrifugation, and the supernatant was used in an immunoprecipitation assay to isolate recombinant TR from the reaction mixture by using TRß1-specific monoclonal antibody J52. The TRß1 immunocomplex was run on 10% SDS-PAGE gels, and the signal was visualized by autoradiography.
In vivo phosphorus labeling and immunoprecipitation
HepG2 cells stably transfected with wild-type TR1 was transfected with 5 µg MKK6EE, active MEKK-1, or empty vector overnight. Then the cells were washed with Tris-buffered saline three times, followed by two washes with phosphorus-free DMEM to remove free phosphorus. The cells were then phosphorus-starved for 3 h before the addition of 1 mCi [32P]orthophosphoric acid/ml medium. Labeling was allowed to proceed for 3 h, followed by extensive washing with ice-cold PBS. Then the cells were allowed to swell in low salt buffer [10 mM Tris (pH 7.5), 2 mM MgCl2, 5 mM sodium vanadate, 25 mM ß-glycerophosphate, and 10 mM sodium pyrophosphate] supplemented with protease inhibitors for 10 min on ice before scraping was performed to lift the cells. After breaking the cell membrane with freeze-thaw cycles, the nuclei were isolated with centrifugation at 3,000 x g for 10 min. Pelleted nuclei were extracted with 500 µl extraction buffer [20 mM Tris (pH 7.5), 2 mM MgCl2, 2 mM EDTA, 400 mM NaCl, 5 mM sodium vanadate, 25 mM ß-glycerophosphate, and 10 mM sodium pyrophosphate] freshly supplemented with protease inhibitors twice and cleared with centrifugation at 15,000 x g for 30 min. Nuclear protein was incubated with 200 µl IgGsorb (The Enzyme Center, Malden, MA) for 30 min and cleared with centrifugation before being incubated with 2 µg TR-specific antibody, C4, or control IgG (MOPC 21) overnight with gentle shaking. Protein A/G agarose was added, and incubation proceeded for another 2 h. Precipitates were washed three times with extraction buffer and eluted with 40 µl 2x sodium dodecyl sulfate sample buffer. Eluted proteins were run on 10% sodium dodecyl sulfate gel, and phosphorylated signals were viewed with autoradiography.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
, A and B). Wild-type TR1, transiently transfected into CV-1 cells, was degraded after treatment with T3 (Fig. 1A, lanes 1 and 2). In contrast, the expression levels of mutant TR1, which is defective in ligand binding (24), were only marginally affected by the T3 treatment (Fig. 1A, lanes 3 and 4). Equal amounts of sample were used in the experiment, as demonstrated by the tubulin signal on the same blot. This suggests that ligand-induced TR degradation is mediated by direct physical binding between TR and T3, rather than through an alternative indirect pathway.
|
1 (24). The TRs were strongly and constitutively expressed in these stable lines, with expression driven by the potent CMV promoter. The activity of the CMV promoter is not disrupted by hormone treatment. The expression of TR was not detected in either the parental HepG2 cells or the Neo control cell line transfected with the neomycin-resistant vector alone (Fig. 1B
, lanes 1–4). In the wild-type TR1 line, TR was highly expressed in cells grown in T3-depleted medium (Fig. 1B, lane 5). However, the amount of TR1 in the cells was significantly reduced after the addition of T3 (Fig. 1B, lane 6 vs. lane 5). In the mutant TR1 (J7-TR1) stable lines, the addition of T3 had no degradative effect on expression levels (Fig. 1B, lanes 7–12). Equal loading was demonstrated by uniform tubulin expression. These observations are consistent with the results from the transient transfection experiments and further confirm that a physical interaction between TR and T3 is necessary for ligand-induced TR degradation.
Previous studies have reported that posttranslational phosphorylation of TR plays a role in increasing protein stability after the binding of T3 and subsequently increasing its transcriptional activity (11). To test this hypothesis in our system, we treated the cells with OA, a protein phosphatase 1 and 2A inhibitor, to examine its effect on the protein-induced TR degradation. As shown in Fig. 1C, T3 induced dose-dependent degradation of TR. Concomitant treatment with OA antagonized the T3-induced TR degradation observed both at low and high doses of T3 (Fig. 2A, lane 4 vs. lane 2). This effect was not isoform specific, as degradation of both TR1 and TRß1 was repressed by OA treatment. This observation suggests that the phosphorylation state of TR plays an important role in the regulation of protein levels in the cell. Consequently, any signaling pathway that changes the phosphorylation status of TR might have a significant influence on TR levels. Thus, we proceeded to study the mechanism of this ligand-induced TR degradation by employing treatments that alter the phosphorylation status of TR.
|
MAPKs are implicated in the phosphorylation of TR
Specific inhibitors of these MAPK pathways have been discovered or developed and are powerful tools for studying the functional roles of their target pathways. By treating TR-transfected cells with these inhibitors, we investigated whether these pathways play differential roles in ligand-induced TR degradation. As shown in Fig. 2A, treatment of transiently transfected CV-1 cells, with MAPK inhibitors (SB202190 for p38s, SB203580 for JNKs, or PD98059 for ERKs), significantly reduced the amount of TR protein in the cells (lanes 5–10) compared with the dimethyloxide-treated control (Fig. 2A, lanes 1 and 2). This observation suggests that MAPKs may participate in increasing the protein stability of TR by actively phosphorylating it. No significant difference was observed between the alternative inhibitor treatments.
As OA treatment can inhibit ligand-induced TR degradation (Fig. 1C, lanes 4–6 and 9–11), and MAPKs participate in the phosphorylation of TR, it is of interest to know whether OA and MAPKs target the same residues on TR. When cells were treated with OA and MAPK inhibitors simultaneously, the antidegradation effect of OA and the reduction of protein expression due to MAPK inhibitors were both compromised (Fig. 2A, lanes 11–16). This result raises the possibility that both OA and MAPKs target the same residues on TR that are critical for the stability of TR in the cells. Endogenous ß-actin was used as the protein input control.
The effects of these treatments on the trans-activation activities of TR on a TRE-driven luciferase reporter construct were also investigated (Fig. 2B). As expected, OA treatment significantly (43- vs. 24-fold) potentiated TR trans-activation (Fig. 2B, lane 4 vs. lane 2). In contrast, treatment with MAPK inhibitors repressed TR-mediated trans-activation activities (Fig. 2B, lanes 5–10). The effects of the MAPK inhibitors and OA treatment were compromised when cells were subjected to both treatments simultaneously (Fig. 2B, lanes 11–16). The effects of these treatments on TR trans-activational activity perfectly correspond to the amount of TR protein in the cells under the same conditions. Both the Western blot and the luciferase assay results suggest that TR is targeted by both OA and MAPKs.
Activated MAPKs stabilize ligand-activated TRs and their activity
To further demonstrate that MAPKs are involved in phosphorylating TR and antagonizing ligand-induced degradation, endogenous MAPKs were activated by the introduction of MAPK upstream activators. MKK6 belongs to a family of kinases that activate the p38 family of MAPK by phosphorylation. Overexpression of MKK6 in CV-1 cells repressed the T3-induced TR degradation to a similar degree as that produced by OA treatment (Fig. 3A). A constitutively active mutant form of MKK6, MKK6EE, was also used in this assay. The effect of MKK6EE overexpression was about 1.5- to 2-fold that of wild-type MKK6 (Fig. 3B, lanes 10–12 vs. lanes 7–9). Both TRß1 and TR1 were targeted by MKK6EE (Figs. 3, A and B, and data not shown). This result clearly demonstrates that activation of MAPK could antagonize the TR degradation induced by T3 and suggests that TR might be a MAPK substrate in vivo.
|
). This suggests that endogenous p38 is involved in TR-mediated trans-activation. Cotransfection of MKK6EE potentiated TR-mediated transcriptional activation about 2- to 4-fold in the presence of T3 (10 and 100 nM). This activation of TR activity was abrogated by the addition of SB202190 to the assay (Fig. 4, A and B). Other inhibitors not specific to the p38 pathway, such as PD98059 and SB203580, had less effect on this MKK6EE-mediated potentiation (data not shown). These observations indicate that MAPK is implicated in TR-mediated downstream gene activation.
|
In the absence of T3, DNA bound by TR showed transcriptional repression of more than 50% and no interaction between TR and VP16-GRIP-1 was detected (Fig. 5). Activation of p38 MAPK by cotransfection of MKK6EE had no effect on either TR trans-activational activity or the interaction between TR and VP16-GRIP-1 (Fig. 5). This suggests that activation of MAPK does not potentiate unliganded TR or increase its repression on gene transcription.
|
). Figure 6A represents the protein chimeras created by the in-frame fusion of the GAL-DBD with either wild-type TRß1 or its deletion mutants. Deletion of the C terminus, harboring the ligand-binding domain, abrogated ligand-inducible TR activities (mutants AB, ABC, and 
C). Activation of MAPK had no effect on the activity of these mutants (Fig. 6B, lanes 5–10). The activity of the mutant with only part of the E domain (mutant C) was not inducible by the addition of T3 or potentiated by overexpression of MKK6EE (Fig. 6B, lanes 11 and 12). However, in mutants possessing an intact hormone-binding domain with preserved ligand-binding activity, the result is entirely different. JL-08 is an AB domain deletion mutant, possessing activity that is considerably higher than that of the wild-type (82- vs. 27-fold; Fig. 6B, lanes 13 and 14). Additionally, cotransfection of MKK6EE potentiated its activity (144- vs. 82-fold). Further deletion of the N terminus created a mutant, JL05, possessing an even higher activity than that of JL-08, also potentiated by cotransfection of MKK6EE (Fig. 6B, lanes 15 and 16). A mutant designated JL06, containing only the D and E domains, is also highly active alone and interacted with VP16-GRIP-1 vigorously (Fig. 6B, lanes 17 and 18, and data not shown). Its ligand-induced activity and interaction with VP16-GRIP-1 (Fig. 6 and data not shown) were also potentiated by activation of MAPK. These observations suggest that 1) activated MAPK only targets T3 activated TR, and the activity of unliganded TR is not affected by MAPK; and 2) D and E domains alone are sufficient to mediate MAPK coactivation. This indicates that MAPK targets the D and E domains of TRß1.
|
, lanes 2 and 6). Cotransfection of MKK6EE potentiated GAL-TRß1 activity about 2- to 3-fold (36- vs. 23-fold; Fig. 7, lane 4 vs. lane 2, and Fig. 6B). Synergism between MKK6EE and VP16-GRIP-1 on TR activity was also observed in cells transfected with both constructs in the presence of GAL-TRß1 (309- vs. 108-fold; Fig. 7, lane 8 vs. lane 6). This observation suggests that activation of MAPKs can cooperate with GRIP-1 to potentiate TR transcriptional activity. However, the interaction between TR and GRIP-1 does not seem to be altered by MAPK activation.
|
were used as positive controls for this assay (data not shown). Trace amounts of TR were identified in CV-1 cells (Fig. 8A, lane 1). A high level of TR was identified in transfected cells, as indicated by supershift using the monoclonal antibody J52 against TRß1 (Fig. 8A, lane 3). T3 treatment induced degradation of TR, and reduced binding to DNA (F2 probe) was also observed (Fig. 8A, lane 4). Interestingly, as the T3 concentration was increased from 10 to 100 nM, TR binding to DNA was also increased (Fig. 8A, lane 4 vs. lane 5, and Fig. 8B, lane 2 vs. lane 3). This binding remained lower than that of TR in the absence of T3 (Fig. 8A, lane 2 vs. lane 5). This result suggests that despite higher levels of TR protein in the cells at a low T3 concentration, a greater level of DNA-binding activity occurs in cells with a higher T3 dose. This result also implies that in the absence of T3, most TR molecules have the ability to bind DNA; however, in the presence of T3, the DNA-binding ability of TR is dependent on the concentration of the hormone. At a low T3 concentration, only a small percentage of TR molecules bound to DNA, but the percentage of DNA-binding TR molecules was increased at a high T3 concentration. Further studies are required to reveal the dynamics and significance of the partition between DNA-bound and unbound TR in the presence of T3.
|
, lanes 6–11, and data not shown). Activation of MAPK by cotransfection with MKK6EE increased the DNA-binding activity about 2- to 3-fold to a level similar to that with lactacystin treatment (Fig. 8B, lane 1 vs. lane 7 and lane 3 vs. lane 9). These results suggest that increased intracellular TR due to activation of MAPKs is functional in DNA binding and trans-activation.
Previous studies have reported that the 26S ubiquitin proteasome complex was involved in ligand-induced nuclear hormone receptor degradation (9). Therefore, it is of interest to compare the effects of activation of MAPKs and inhibition of proteasome complex. As shown in Fig. 8B, the addition of lactacystin, a potent 26S proteasome inhibitor, highly increased the DNA-binding activity about 3-fold in the absence of T3 (Fig. 8B, lane 1 vs. lane 4) and about 2-fold in the presence of T3 (Fig. 8B, lane 2 vs. lane 5). We also observed increased TR protein levels in cells treated with lactacystin (data not shown). The effect of MAPK activation is similar to that of proteasome inhibition (Fig. 8B, lane 1 vs. lane 7 and lane 3 vs. lane 9). However, the effect of MKK6EE cotransfection is not synergistic with that of lactacystin treatment (Fig. 8B, lanes 10–12). Thus, this suggests that these two treatment, activation of MAPKs and inhibition of proteasome, increase intracellular DNA-binding activity by employing the same mechanism, i.e. stabilizing intracellular TR proteins.
The ability of TR to form a heterodimeric complex with RXR was also examined in the presence and absence of T3 by the addition of in vitro translated RXR in the total lysate. In the dimethylsulfoxide control, TR vigorously formed a heterodimeric complex with RXR in the absence of T3. The addition of T3 reduced the heterodimeric complex to about 90% due to reduced TR protein amounts in the cells (Fig. 8B, lane 13 vs. lane 14). Total lysate from cells treated with lactacystin or cotransfected with MKK6EE showed approximately 2- to 3-fold increased heterodimeric complex forming ability with RXR in both the absence and presence of T3 (Fig. 8B, lane 13 vs. lanes 15 and 17), which faithfully reflects the amount of functional TR in the cells, as shown in the absence of RXR. The ability of TR to form higher complex with the SRC-1 receptor interaction domain on DNA in the presence and absence of T3 was also investigated, and similar results were found (data not shown). All of our EMSA observations suggest that both MAPKs and the 26S proteasome complex target TR to antagonize its ligand-induced degradation and thus increase intracellular TR protein amount and DNA-binding activity.
TR proteins are targeted by p38 MAPK in vitro and in vivo
Phosphorylation of TR by p38 kinase requires the interaction and binding of these two proteins. A protein-protein interaction represents the first step before the kinase reaction can be initiated. A GST-TRß1 fusion protein was produced and purified from E. coli (BL21) and used to pull down in vitro translated 35S-labeled p38 kinase that was either activated by MKK6EE or inactivated in the presence or absence of T3. In the absence of T3, the interaction between inert p38 and GST-TRß1 was undetectable (data not shown). The addition of T3 allowed the detection of the interaction by our assay system (Fig. 9A, lane 7), although the interaction was relatively weak. p38 has to be activated by its upstream regulator, MKK3/6, before it is able to exert its kinase activity in vivo. This led to the postulation that p38 may need to be activated before it can interact with TR. Thus, we performed an in vitro kinase reaction by coincubating in vitro translated p38 and MKK6EE proteins in the presence of ATP to activate p38 kinase before the GST pulldown assay was performed. As expected, activated p38 kinase interacted with GST-TRß1 to a greater degree (
3-fold) than inert p38 (Fig. 9A, lane 5 vs. lane 7). We also found that the presence of T3 could potentiate this interaction about 4-fold (Fig. 9A, lane 5 vs. lane 6). Removal of T3 significantly reduced this interaction, despite the fact that the p38 kinase had already been activated (Fig. 9A, lane 6). These observations suggest that p38 kinase can interact with TR through a direct protein-protein interaction, and this interaction depends critically on the presence of T3 and the status of p38 activity.
|
). Surprisingly, RXR was neither pulled down by GST-p38 nor coprecipitated with TR (Fig. 9B). This observation suggests that interaction between TR and MAPKs might occur before TR/RXR heterodimer formation on the DNA.
To investigate TR as a substrate of p38 kinase activity in vitro, we expressed and purified recombinant TRß1 protein in E. coli. Purified TR was stained singly on SDS-PAGE stained with Coomassie Blue (data not shown). Endogenous p38 protein from CV-1 cells was isolated by immunoprecipitation using a polyclonal antibody against p38 as described by Davis’s group (23). Isolated p38 immunocomplex was used as an active enzyme for phosphorylating recombinant TR protein in the presence of 10 µCi [-32P]ATP. After the kinase reaction, the p38 immunocomplex was removed by centrifugation, and the supernatant was used to immunoprecipitate TRß1 to separate TRß1 from other nonspecifically labeled proteins carried over from the p38 immunocomplex. As shown in Fig. 9C, the p38 immunocomplex vigorously phosphorylated recombinant TR protein (Fig. 9C, lanes 2 and 3). Phosphorylated TR had slower mobility than the in vitro translated product (Fig. 9C, lane 1), a phenomenon often observed in proteins regulated by posttranslational phosphorylation. This observation suggests that TR is a candidate substrate for p38 kinase activity in vitro and also implies that TR can be targeted by p38 kinase activity in vivo.
To examine whether activation of the p38 MAPK pathway in the cells affected the phosphorylation state of TR, HepG2 cells stably transfected with wild-type TR1 were metabolically labeled with [32P]orthophosphoric acid, and TR1 protein in the cells was isolated with immunoprecipitation. As expected, the amount of phosphorylated TR1 increased significantly in cells transfected with MKK6EE compared with cells transfected with empty vector (Fig. 9D, lanes 1 and 2). The signal of TR1 was verified using MOPC 21 (a nonspecific monoclonal antibody) as a control (Fig. 9D, lane 4). We also examined the effect of overexpressing constitutively active form of MEKK-1 on TR1 phosphorylation in vivo. Activation of the MEKK-1 downstream pathway increased the phosphorylation state of TR (Fig. 9D, lane 3) about 2-fold, the magnitude of which is similar to its effect on the trans-activational activity of TR on reporter constructs (data not shown).
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MAPKs target the D and E domains of TR
The ligand binding-dependent interaction of TR with coactivators is mediated by the hormone-binding domain in the C terminus of TR (6). Our study shows that TR deletion mutants (JL-05, -06, and -08) harboring the intact ligand-binding domain activated gene transcription vigorously and interacted strongly with VP16-GRIP-1. Their transcriptional activity in either the absence or presence of VP16-GRIP-1 was potentiated by cotransfection of MKK6EE. The mammalian two-hybrid assay suggests that the ligand binding-induced conformational change in TR is a prerequisite for MAPK potentiation. These observations also indicate that the ligand-binding domain might be the most important domain targeted by MAPKs, because the activity of the deletion mutant (JL-06), harboring only the ligand-binding domain and a small fragment of the hinge domain (D and E domains, respectively), can be potentiated by MAPK activation.
Protein kinase A (PKA) is also involved in phosphorylating TRs and altering their DNA binding specificity (27). Alterations in PKA activity have been reported to modify TR activity in the cells. The use of alanine substitution and kinase inhibitors has identified the PKA target residues as serine 16/17 located in the A/B domain (28, 29). However, baculovirus-expressed TR proteins were used in this study, and T3 treatment was performed in vitro, instead of treating the cells. Thus, no ligand-induced degradation was observed in the study (27). The effect of PKA treatment on ligand-induced TR degradation has yet to be determined. Our observations suggest that MAPKs target TRß1 in the D and E domains, instead of the A/B domain. Thus, it is possible that TR and PKA may cooperate with each other to regulate the activity and stability of TR.
MAPKs increase TR DNA-binding ability in the cells
Previous studies have shown that TR1 and v-erbA are targeted by PKA at serine 28/29 and serine 16/17, respectively (28, 29). Treatment with PKA inhibits DNA binding by TR monomers without affecting the DNA binding of TR homo- and heterodimers (27). Thus, phosphorylation of TR by PKA regulates TR DNA recognition. However, a similar result was not observed for MAPKs. In this study we observed that DNA-binding activity was significantly reduced in cells treated with MAPK inhibitors compared with the control cells, and activation of the MAPK pathway by cotransfection of MKK6EE significantly increased the DNA-binding activity. Both effects were probably due to the MAPK-mediated protein stability change as observed in Figs. 1–3. The total lysate used in our EMSA study is derived from cells lysed with lysis buffer containing Triton X-100. It is possible that TR proteins might be denatured to some degree in the lysis buffer, and only strong binding to DNA can be observed. This may explain the fact that no monomer binding to DNA was observed in our EMSA. However, homodimers, heterodimers, and even higher complexes formed by a specific antibody against TRß1 or the SRC/p160 coactivator receptor interaction domain are clearly observed. These complexes, formed on the DNA, suggest that TR protein is still capable of binding to DNA and recruiting essential cofactors to the DNA.
TRs are substrates of p38 kinase activity
Protein-protein interactions have been observed between p38 and most of its substrates and are recognized as a characteristic of these substrates. In this study we found that the interaction between TRß1 and p38 kinase depends critically on the presence of T3 and the activational state of p38 kinase. The dependence of hormone binding is analogous to the interaction between most nuclear hormone receptors and transcriptional coactivators, in which the conformational change induced by ligand binding plays a critical role in creating a suitable conformation for interacting with coactivators and subsequent transcriptional activation. Thus, T3 dependence suggests that the conformational change induced by ligand binding is a prerequisite for the p38 and TR interaction. This statement is also consistent with observations from the mammalian two-hybrid assay, in which activation of p38 has no effect on the activity of unliganded TR, but has a significant effect on that of ligand-bound TR. These two experiments suggest that p38 kinase preferentially targets ligand-bound TR to increase their stability and consequently amplify their transcriptional activity.
Kinase assays in vitro have shown that TR is targeted by p38 kinase activity and also imply that it is a possible candidate substrate of p38 in vivo. The GST pulldown assay suggests that these two proteins interact directly in vitro. Overexpression of MKK6EE in the HepG2 cells stably transfected with wild-type TR1 has clearly demonstrated that activation of p38 MAPK can increase the phosphorylation of TR in the cells. Therefore, the phosphorylation state of TR can be targeted by p38 MAPK either in vitro or in vivo. A previous study by Davis et al. (30) found that T4 can nongenomically activate MAPK, which causes complexing of TR and MAPK and subsequent phosphorylation of TR by MAPK. Their study has lent powerful support to our finding of a ligand-dependent interaction between TR and MAPK. However, the effect of T4-activated MAPK on TR phosphorylation is transient, the consequence of which is not examined in their study, except for increased nuclear accumulation of TR. Our finding has further demonstrated that persistent activation of MAPK can increase the phosphorylation of TRs and consequently increase their protein stability and trans-activational activity.
In summary, we found that activation of p38 kinase increased TR protein stability and potentiated its transcriptional activity. Interaction between TR and one of its coactivators, GRIP-1, was also synergistically potentiated by activation of p38 kinase. TR was found to interact with activated p38 kinase in a ligand-dependent manner, and the phosphorylation of TR was significantly increased by persistent activation of p38 MAPK.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Abbreviations: CMV, Cytomegalovirus; DTT, dithiothreitol; ERK, extracellular signal-regulated kinase; GAL-DBD, ß-galactosidase DNA-binding domain; GRIP-1, glucocorticoid receptor interacting protein-1; GST, glutathione-S-transferase; IP, immunoprecipitation; JNK, c-Jun N-terminal kinase; LAP, inverted palindromic repeat thyroid hormone response element; MKK, MAPK kinase; OA, okadeic acid; PBST, PBS with 0.5% Tween 20; PKA, protein kinase A; RXR, retinoid X receptor; SRC, steroid receptor coactivator; TK, thymidine kinase; TR, thyroid hormone receptor; TRE, thyroid hormone response element; VP16, herpes simplex virus protein 16.
Received August 30, 2002.
Accepted for publication January 7, 2003.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
1 and ß1 receptors in a human hepatocellular carcinoma cell line. Endocrinology 137:4073–4081[Abstract]
This article has been cited by other articles:
 |
|
 |
|
  M. R. Bratton, D. E. Frigo, K. A. Vigh-Conrad, D. Fan, S. Wadsworth, J. A. McLachlan, and M. E. Burow Organochlorine-mediated potentiation of the general coactivator p300 through p38 mitogen-activated protein kinase Carcinogenesis, January 1, 2009; 30(1): 106 - 113. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Narayanan, C. C. Coss, M. Yepuru, J. D. Kearbey, D. D. Miller, and J. T. Dalton Steroidal Androgens and Nonsteroidal, Tissue-Selective Androgen Receptor Modulator, S-22, Regulate Androgen Receptor Function through Distinct Genomic and Nongenomic Signaling Pathways Mol. Endocrinol., November 1, 2008; 22(11): 2448 - 2465. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R.-N. Chen, Y.-H. Huang, Y.-C. Lin, C.-T. Yeh, Y. Liang, S.-L. Chen, and K.-H. Lin Thyroid Hormone Promotes Cell Invasion through Activation of Furin Expression in Human Hepatoma Cell Lines Endocrinology, August 1, 2008; 149(8): 3817 - 3831. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Lei, C. N. Mariash, M. Bhargava, E. V. Wattenberg, and D. H. Ingbar T3 increases Na-K-ATPase activity via a MAPK/ERK1/2-dependent pathway in rat adult alveolar epithelial cells Am J Physiol Lung Cell Mol Physiol, April 1, 2008; 294(4): L749 - L754. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 I. Irrcher, D. R. Walkinshaw, T. E. Sheehan, and D. A. Hood Thyroid hormone (T3) rapidly activates p38 and AMPK in skeletal muscle in vivo J Appl Physiol, January 1, 2008; 104(1): 178 - 185. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P.-J. Tai, Y.-H. Huang, C.-H. Shih, R.-N. Chen, C.-D. Chen, W.-J. Chen, C.-S. Wang, and K.-H. Lin Direct Regulation of Androgen Receptor-Associated Protein 70 by Thyroid Hormone and Its Receptors Endocrinology, July 1, 2007; 148(7): 3485 - 3495. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. E. Frigo, A. Basu, E. N. Nierth-Simpson, C. B. Weldon, C. M. Dugan, S. Elliott, B. M. Collins-Burow, V. A. Salvo, Y. Zhu, L. I. Melnik, et al. p38 Mitogen-Activated Protein Kinase Stimulates Estrogen-Mediated Transcription and Proliferation through the Phosphorylation and Potentiation of the p160 Coactivator Glucocorticoid Receptor-Interacting Protein 1 Mol. Endocrinol., May 1, 2006; 20(5): 971 - 983. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M.-B. Poirier, G. Hamann, M.-E. Domingue, M. Roy, T. Bardati, and M.-F. Langlois General Receptor for Phosphoinositides 1, a Novel Repressor of Thyroid Hormone Receptor Action that Prevents Deoxyribonucleic Acid Binding Mol. Endocrinol., August 1, 2005; 19(8): 1991 - 2005. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. Maruvada, N. I. Dmitrieva, J. East-Palmer, and P. M. Yen Cell Cycle-dependent Expression of Thyroid Hormone Receptor-{beta} Is a Mechanism for Variable Hormone Sensitivity Mol. Biol. Cell, April 1, 2004; 15(4): 1895 - 1903. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
