If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Regulation of the Phosphatidylinositol 3-Kinase, Akt/Protein Kinase B, FRAP/Mammalian Target of Rapamycin, and Ribosomal S6 Kinase 1 Signaling Pathways by Thyroid-stimulating Hormone (TSH) and Stimulating type TSH Receptor Antibodies in the Thyroid Gland*
To whom correspondence should be addressed: Laboratory of Endocrine Cell Biology, Dept. of Internal Medicine, Chungnam National University School of Medicine, 640 Daesadong Chungku, Taejon 301-040, Korea. Tel.: 82-42-220-7161; Fax: 82-42-257-5753
* This work was supported by National Research Laboratory Program Grant M1-0104-00-0014 and KOSEF Research Grant 2000-2-20500-007-3 from the Ministry of Science and Technology, Korea. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Thyroid-stimulating hormone (TSH) regulates the growth and differentiation of thyrocytes by activating the TSH receptor (TSHR). This study investigated the roles of the phosphatidylinositol 3-kinase (PI3K), PDK1, FRAP/mammalian target of rapamycin, and ribosomal S6 kinase 1 (S6K1) signaling mechanism by which TSH and the stimulating type TSHR antibodies regulate thyrocyte proliferation and the follicle activities in vitro and in vivo. The TSHR immunoprecipitates exhibited PI3K activity, which was higher in the cells treated with either TSH or 8-bromo-cAMP. TSH and cAMP increased the tyrosine phosphorylation of TSHR and the association between TSHR and the p85α regulatory subunit of PI3K. TSH induced a redistribution of PDK1 from the cytoplasm to the plasma membrane in the cells in a PI3K- and protein kinase A-dependent manner. TSH induced the PDK1-dependent phosphorylation of S6K1 but did not induce Akt/protein kinase B phosphorylation. The TSH-induced S6K1 phosphorylation was inhibited by a dominant negative p85α regulatory subunit or by the PI3K inhibitors wortmannin and LY294002. Rapamycin inhibited the phosphorylation of S6K1 in the cells treated with either TSH or 8-bromo-cAMP. The stimulating type TSHR antibodies from patients with Graves disease also induced S6K1 activation, whereas the blocking type TSHR antibodies from patients with primary myxedema inhibited TSH- but not the insulin-induced phosphorylation of S6K1. In addition, rapamycin treatment in vivo inhibited the TSH-stimulated thyroid follicle hyperplasia and follicle activity. These findings suggest an interaction between TSHR and PI3K, which is stimulated by TSH and cAMP and might involve the downstream S6K1 but not Akt/protein kinase B. This pathway may play a role in the TSH/stimulating type TSH receptor antibody-mediated thyrocyte proliferation in vitro and in the response to TSH in vivo.
The pituitary glycoprotein hormones ACTH, follicle-stimulating hormone, luteinizing hormone, and TSH
The abbreviations used are: TSH, thyroid-stimulating hormone; CHO, Chinese hamster ovary; CREB, cAMP-response element-binding protein; 4EBP, eIF-4E-binding protein; MMI, methimazole; mTOR, mammalian target of rapamycin; PDK1, 3′-phosphoinositide-dependent kinase; PI3K, phosphatidylinositol 3-kinase; PKA, protein kinase A; PKB, protein kinase B; S6K1, ribosomal S6 kinase 1; TSAb, stimulating type TSH receptor antibody; TSBAb, blocking type TSH receptor antibody; 8-Br-cAMP, 8-bromo-cAMP; TPCK, N-α-tosyl-l-phenylalanyl chloromethyl ketone; MAP, mitogen-activated protein; HA, hemagglutinin; GFP, green fluorescent protein; PtdIns, phosphatidylinositol.
1The abbreviations used are: TSH, thyroid-stimulating hormone; CHO, Chinese hamster ovary; CREB, cAMP-response element-binding protein; 4EBP, eIF-4E-binding protein; MMI, methimazole; mTOR, mammalian target of rapamycin; PDK1, 3′-phosphoinositide-dependent kinase; PI3K, phosphatidylinositol 3-kinase; PKA, protein kinase A; PKB, protein kinase B; S6K1, ribosomal S6 kinase 1; TSAb, stimulating type TSH receptor antibody; TSBAb, blocking type TSH receptor antibody; 8-Br-cAMP, 8-bromo-cAMP; TPCK, N-α-tosyl-l-phenylalanyl chloromethyl ketone; MAP, mitogen-activated protein; HA, hemagglutinin; GFP, green fluorescent protein; PtdIns, phosphatidylinositol.
control the function of specific target cells in the adrenal gland, gonads, and thyroid. All of these hormones are not only important for hormone production but also for maintaining the glandular weight in their target gland. These hormones bind to ligand-specific cell-surface G-protein-coupled receptors and activate adenylyl cyclase to produce cAMP. These glycoprotein hormone receptors also activate the PI3K-dependent signaling pathways (
). TSHR mediates these activities by activating the diverse signaling pathways including the PI3K pathway. The signaling components downstream of TSHR (Gβγ, cAMP, and PKA) may overlap with the downstream components of the PI3K signaling pathway. The Gβγ subunit of heterotrimeric G-proteins specifically activates PI3Kγ in the myeloid-derived cells (
). Both TSH and cAMP induce Ser-473 phosphorylation in Akt/PKB, a major phosphorylation site for the regulation of Akt/PKB by growth factors. TSH- and cAMP-induced phosphorylation of Ser-473 is PI3K-dependent (i.e. wortmannin-sensitive) in rat WRT thyroid cells (
This study examines the interactions between the TSHR and the regulatory p85α subunit of PI3K in the presence and absence of TSH and cAMP. TSH and 8-Br-cAMP stimulate the interaction between TSHR and PI3K, which leads to a PI3K- and PKA-dependent translocation of PDK1. PDK1 phosphorylation of Akt/PKB and S6K1 appears to be differentially stimulated by TSH and insulin. TSH stimulates S6K1 through the PI3K-, PDK1-, and PKA-dependent but Akt/PKB-independent pathways. The FRAP/mTOR inhibitor rapamycin inhibits the TSH-stimulated phosphorylation of S6K1 and inhibits cell cycle progression in the FRTL-5 thyroid cells. Rapamycin also modulates the thyroid follicle activity, which is induced by elevated endogenous TSH levels in vivo. This study suggests that S6K1 plays an important role in TSHR-activated PI3K signaling, which modulates the thyrocyte proliferation and thyroid follicle activity.
Materials—The media and cell culture reagents and materials were purchased from Invitrogen, Sigma, Fisher, Corning Glass, and Hyclone Laboratories, Inc. (Logan, UT). Wortmannin and H89 were from Calbiochem. LY294002 and PD98059 were obtained from Sigma and New England Biolabs (Beverly, MA). Antibodies for cyclin D1 (sc-8396), p85a (sc-423), and p110γ (H-119) were from Santa Cruz Biotechnology (Santa Cruz, CA). Antibodies for HA (2363), Akt (9272), phospho-Akt-Ser-473, and Thr-308 (9271 and 9275); phospho-4EBP1-Ser-65 (9451) and 4EBP1 (9452); Myc (2276), S6 ribosomal protein (2212), and phospho-S6 ribosomal protein-Ser-235/236 (2211); and p44/p42 MAP kinase (9102) and phospho-p44/42 MAP kinase-Thr-202/Tyr-204 (9101) were from Cell Signaling Technology (Beverly, MA). Mouse anti-human TSHR (MCA1571) was from Serotec (Kidlington, Oxford, UK), and anti-FLAG antibody (200472) was from Stratagene (La Jolla, CA). All other materials, including 8-bromo-cAMP (8-Br-cAMP and, N-α-tosyl-l-phenylalanyl chloromethyl ketone (TPCK), were purchased from Sigma.
Preparation of IgG—The control sera were obtained from 5 normal individuals who had no history or clinical or chemical evidence (abnormal thyroid hormone and TSH levels) of thyroid disease. The diagnosis of the 10 patients with Graves disease was based on conventional clinical and laboratory criteria, including elevated serum thyroid hormone levels, undetectable TSH by a sensitive radioimmunoassay, testing positive for TBII (TSH-binding inhibitory immunoglobulins), and a diffuse goiter with increased 99mTc uptake at scintiscan. IgGs were extracted by affinity chromatography using protein A-Sepharose CL-4B columns; IgG was lyophilized and stored at–20 °C until assay. The IgGs were extracted by affinity chromatography from normal pooled sera (NP), as well as the sera from the patients with primary myxedema (PM) who tested positive to the blocking TSHR antibody test.
Cell Culture and Gene Transfection—A fresh subclone (F1) of FRTL-5 was used in the rat thyroid cells (Interthyr Research Foundation, Baltimore, MD). After 6 days in medium with no TSH, the addition of 1 milliunit/ml crude bovine TSH (Sigma) stimulated thymidine incorporation into DNA by at least 10-fold. The doubling time of the cells with TSH was 36 ± 6 h without TSH, and they did not proliferate. The cells used were diploid and between their 5th and 20th passage. The cells were grown in 6H medium consisting of Coon's modified F-12 supplemented with 5% calf serum, 1 mm nonessential amino acids, and a mixture of six hormones as follows: bovine TSH (1 milliunit/ml), insulin (10 μg/ml), cortisol (0.4 ng/ml), transferrin (5 μg/ml), glycyl-l-histidyl- l-lysine acetate (10 ng/ml), and somatostatin (10 ng/ml). Fresh medium was added to all cells every 2 or 3 days, and the cells were passaged every 7–10 days. In individual experiments, the cells were shifted to 3H (which is devoid of insulin, TSH, and somatostatin), 4H (which is devoid of insulin and TSH), or 5H medium (which is devoid of TSH) with or without 5% calf serum before TSH, forskolin (Sigma), or the other agents were added. Chinese hamster ovary (CHO) cells were maintained at 37 °C and 5% CO2 in Ham's F-12 medium containing 10% fetal bovine serum, 100 units/ml penicillin, and 100 μg/ml streptomycin. Expression plasmids carrying hTSHR, p85α, and PDK1 cDNAs were introduced into the FRTL-5 and CHO cells using the LipofectAMINE Plus reagents according to the manufacturer's instructions (Invitrogen).
PI3K Assay—The cell extracts obtained from the FRTL-5 cells were immunoprecipitated with anti-TSHR monoclonal antibodies (clone 4C1, Serotec, Oxford, UK) and the control IgG antibodies. The samples were washed twice with 1% Nonidet P-40 and 1 mm sodium orthovanadate in phosphate-buffered saline, twice with washing buffer consisting of 100 mm Tris-HCl (pH 7.5), 500 mm LiCl, and 1 mm sodium orthovanadate, and twice with ST with 150 mm NaCl and 50 mm Tris-HCl (pH 7.2). The samples were resuspended in a PI kinase buffer containing 20 mm Hepes (pH 7.2), 100 mm NaCl, 10 μg/ml leupeptin, and 10 μg/ml pepstatin. A phosphoinositide/EGTA solution consisting of 1 mg/ml phosphoinositide and 2.5 mm EGTA was then added, and the samples were incubated at room temperature for 10 min. A solution containing 20 mm Hepes (pH 7.4), 5 mm MnCl2, 10 μm ATP, and 20 μCi of [γ-32P]ATP was added, and the samples were incubated at 30 °C for 20 min. The reactions were quenched by the addition of 1 m HCl, and the phospholipids were extracted using CHCl3. The dried samples were separated by TLC. The phosphorylated lipids were visualized by autoradiography and quantified using a PhosphorImager (BAS1500, Fuji).
Immunoprecipitation and Western Blot Analysis—The following immunoprecipitation procedures were carried out at 4 °C. The cells grown on the 100-mm dishes were washed with phosphate-buffered saline twice prior to lysis. The RIPA buffer containing the protease inhibitors (20 μg/ml leupeptin, 10 μg/ml pepstatin A, 10 μg/ml chymostatin, 2 μg/ml aprotinin, 1 mm phenylmethylsulfonyl fluoride) were added for cell lysis. The cell lysate was collected, triturated, and centrifuged at 1000 × g for 10 min. To preclear the cell lysate, the supernatant was mixed with 20 μl of protein A/G beads (Santa Cruz Biotechnology), incubated for 30 min with rocking, and were centrifuged for 15 min at 1000 × g. Precleared samples were incubated with the primary antibodies for 2 h with rocking, and protein A/G beads were then added, incubated overnight, and centrifuged at 1000 × g. The immunoprecipitates were collected and washed three times with RIPA buffer.
For Western blot analysis, the cells were lysed at 4 °C in a mixture of 10 mm KPO4, 1 mm EDTA, 5 mm EGTA, 10 mm MgCl2, 50 mm β-glycerophosphate, 2 mm dithiothreitol, 1% Nonidet P-40, 1 mm Pefabloc (Roche Applied Science), and 10 μg each of aprotinin and leupeptin/ml. The total protein lysates were denatured by boiling in a Laemmli sample buffer, resolved on 7.5–15% SDS-PAGE, and transferred to polyvinylidene fluoride membranes. The membranes were blocked in phosphate-buffered saline containing 5% (w/v) milk and 0.1% Tween and then incubated for 2 h with the polyclonal antibodies against p70 S6K (supplied by Dr. J. Blenis, Harvard University, Boston).
[3H]Thymidine Incorporation Assay—Confluent FRTL-5 thyroid cells in 100-mm dishes were detached by trypsinization, resuspended in 6H growth medium, seeded at a density of 3 × 104 cells/well in 24-well plates, and incubated for 2–3 days until 80% confluent. The medium was changed to 5H medium and incubated for an additional 7 days. TSH and/or wortmannin, LY294002, H89, and rapamycin were added to the quiescent cells, which were then incubated for 24 h, followed by the addition of 2 μCi/ml [3H]thymidine (PerkinElmer Life Sciences) to pulse the cells for an additional 12 h. The experimental samples were prepared in triplicate. The cells were washed four times with ice-cold phosphate-buffered saline, precipitated twice with ice-cold 10% trichloroacetic acid (30 min each time on ice), briefly washed once with ice-cold ethanol, lysed with 0.2 n NaOH in 0.5% SDS, and incubated at 37 °C for at least 30 min. The level of radioactivity was determined by liquid scintillation spectrometry (Beckman Instruments). The results were measured as the number of counts/min in each well. Each experimental data point represents triplicate wells from at least four different experiments.
Flow Cytometry—The samples were prepared for flow cytometry essentially as described previously (
). Briefly, the cells were washed with 1× phosphate-buffered saline, pH 7.4, and then fixed with ice-cold 70% ethanol. The samples were washed with 1× phosphate-buffered saline and stained with propidium iodide 60 μg/ml (Sigma) containing 100 μg/ml RNase (Sigma) for 30 min at 37 °C. Cell cycle analysis was performed using a BD Biosciences fluorescence-activated cell analyzer and Cell Quest version 1.2 software (BD Biosciences). At least 10,000 cells were analyzed per sample. The cell cycle distribution was quantified using the ModFit LT version 1.01 software (Verity Software House Inc., Topsham, ME).
Confocal Microscopy—The FRTL-5 thyroid cells were grown on coverslips and transfected with pEGFP-PDK-1 and pCDNA3-PDK-1-myc using the LipofectAMINE method (Invitrogen). Quiescent cells stimulated with TSH were washed three times with cold phosphate-buffered saline and fixed in 3.7% formaldehyde for 40 min. Fixed cells were mounted on glass slides with phosphate-buffered saline and observed using laser-scanning confocal microscopy (Leica TCS SP2).
Protein Kinase Assays—The FRTL-5 thyroid cells were transiently transfected with HA-tagged p70 S6K1 and immunoprecipitated with an anti-HA monoclonal antibody coupled to protein A/G-agarose (Santa Cruz Biotechnology). The samples were washed twice with Buffer A (containing 20 mm Tris-HCl (pH 7.5), 0.1% Nonidet P-40, 1 mm EDTA, 5 mm EGTA, 10 mm MgCl2, 50 mm β-glycerophosphate, 1 mm sodium orthovanadate, 2 mm dithiothreitol, 40 μg/ml phenylmethylsulfonyl fluoride, and 10 μg/ml leupeptin) and then twice with Buffer B containing 500 mm NaCl. Finally the immunocomplexes were washed, in succession, with Buffer C (containing 20 mm Hepes (pH 7.2), 10 mm MgCl2, 0.1 mg/ml bovine serum albumin, and 3 mm β-mercaptoethanol), Buffer D (containing 20 mm Hepes (pH 7.2), 10 mm MgCl2,10mm MnCl2,1mm dithiothreitol, and 0.2 mm EGTA), and Buffer E (containing 50 mm Tris (pH 7.5), 10 mm NaCl, 1 mm dithiothreitol, and 10% glycerol). The S6 phosphotransferase activities were assayed in a reaction mixture consisting of 1× Buffer C, 1 μg of S6 peptide, 20 μm ATP, and 5 μCi of [γ-32P]ATP (specific activity: 3000 Ci/mmol; PerkinElmer Life Sciences) at 30 °C for 20 min. The samples were subjected to liquid scintillation counting (Hewlett-Packard).
Preparation of Thyroid Gland and Immunohistochemistry—Male Sprague-Dawley rats (120–130 g) were fed with water containing 0.025% MMI for 2 weeks. Rapamycin (Calbiochem) was delivered once daily by an intraperitoneal injection at a dose of 1.5 mg/kg dissolved in 2% carboxymethylcellulose for 1 week before histologic examination. Tissue samples of the rat thyroid gland were fixed in 10% buffered formalin, processed routinely, and embedded in paraffin. Three-micrometer-thick sections were cut from the paraffin blocks and stained with hematoxylin and eosin (H&E). The number of follicles in 2 medium power fields (×200) was counted. Further sections were used for immunohistochemistry. All immunostaining steps were carried out at room temperature. After deparaffinization and antigen retrieval by autoclaving in 10 mm sodium citrate buffer (pH 6.0) at full power for 10 min, the tissue sections were treated with blocking rabbit serum for 15 min. The primary antibody, polyclonal rabbit anti-phospho-S6 ribosomal protein (Cell Signaling Technology), was diluted (1:200) with a background-reducing diluent (Dako, Carpinteria, CA) and incubated for 60 min. The sections were then incubated with a rabbit EnVision-HRP detection system reagent (Dako, Carpinteria, CA) for 30 min. The sections were then sequentially incubated with 3,3-diaminobenzidine (Dako) chromogen for 5 min, counterstained with Meyer's hematoxylin, and mounted. Careful rinses with several changes of Tris-buffered saline, 0.3% Tween buffer were performed between each step. A negative control that excluded the primary antibody was used. Cells with cytoplasmic granular staining were considered positive.
Other Assays—The protein concentration was determined by the Bradford method (Bio-Rad) using recrystallized bovine serum albumin as a standard. The sera were stored–70 °C until IgG preparation. The IgGs were extracted by affinity chromatography using protein A-Sepharose CL-4B columns (Amersham Biosciences) followed by dialysis. The purity of the IgG preparation was confirmed by the documentation of undetectable TSH levels with immunoradiometric assay.
Statistical Analysis—All experiments were repeated at least three times with different cells. The values are the mean ± S.E. of these experiments. Statistical significance was determined by a two-way analysis of variance.
Association of PI3K Activities with TSHR—TSHR was immunoprecipitated from extracts from FRTL-5 thyroid cells exposed to TSH or 8-Br-cAMP using monoclonal anti-TSHR antibodies, and the immunoprecipitate was tested for PI3K activity. The phosphatidylinositol phosphotransferase activity was detected in the immunoprecipitate, which was inhibited by the PI3K inhibitors wortmannin (100 nm) or LY294002 (0.5 μm) (Fig. 1A). The TSHR-associated PI3K activity was barely detectable in the untreated FRTL-5 thyroid cells (Fig. 1, A, lane 1, and B, lane 1). Both TSH and 8-Br-cAMP stimulated the TSHR-associated PI3K activity (Fig. 1B, lanes 2 and 4, respectively). H89 did not inhibit the PI3K kinase activity in vitro (data not shown), but the H89 treated cells had a lower level of TSHR-associated PI3K activity in the TSH- or 8-Br-cAMP-treated cells (Fig. 1B, lanes 3 and 5).
The specificity of the interaction between TSHR and PI3K was determined in the following experiment. The extracts were prepared from the wild-type CHO cells or the CHO cells expressing human TSHR (CHO-TSHR) (
), and immunoprecipitation was carried out using anti-TSHR (Fig. 1C, lanes 1–4) or the control IgG (Fig. 1C, lane 5). A low level of PI3K activity was detected in the anti-TSHR immunoprecipitates from the CHO-TSHR cells (Fig. 1C, lane 2). This activity increased in the cells treated with TSH (Fig. 1C, lanes 2 versus 3) and was inhibited by 100 nm wortmannin (Fig. 1C, lane 4). No PI3K activity was detected in the experiments using the wild-type CHO cells (Fig. 1C, lane 1) or the control IgG (Fig. 1C, lane 5). These findings suggest that TSHR interacts specifically with PI3K and that TSH and cAMP stimulate this interaction.
TSHR Interacts with the p85α Regulatory Subunit of PI3K— Western blot analysis was carried out to identify the specific regulatory and catalytic subunits of class I PI3K in FRTL-5 thyroid cells. The p85α regulatory subunit of class I PI3K (p85α) was expressed strongly in the cells treated with or without TSH, 8-Br-cAMP, and insulin (Fig. 2A). In contrast, the p110γ catalytic subunit of class I PI3K, which is activated by the Gβγ subunits of the heterodimeric G-proteins, was not expressed at a detectable level in these cells (Fig. 2A).
The following experiments examined the association between TSHR and the regulatory subunit of PI3K. Human TSHR and HA-tagged p85α were co-expressed in CHO cells. TSHR was immunoprecipitated, and the immune complexes were analyzed with anti-HA antibodies (Fig. 2B). The TSHR immunoprecipitates from the CHO-TSHR cells included p85α (Fig. 2B, lane 4). TSHR immunoprecipitates from FRTL-5 cells treated with TSH included p85α (Fig. 2C, lanes 2 and 3). The cell extracts were also immunoprecipitated with control IgG; the results show that the interaction between TSHR and HA-p85α is specific (Fig. 2C, lanes C versus T).
The immune complexes precipitated with the anti-TSHR antibodies were also probed with anti-phosphotyrosine antibodies (4G10). In untreated cells, the TSHR had a low level of tyrosine phosphorylation. In contrast, the TSHR from the cells treated with TSH or 8-Br-cAMP had a significantly higher phosphotyrosine content (Fig. 2D, lanes 1–4). Endogenous p85α was detected in the TSHR immunoprecipitates from the TSH- or 8-Br-cAMP treated FRTL-5 thyroid cells.
These findings suggest the following. 1) The TSH receptor is specifically associated with the p85α regulatory subunit of PI3K in the thyroid cells, and the TSHR-transfected CHO cells. 2) The association between the endogenous p85α and TSHR depends on the TSH/8-Br-cAMP treatment and on tyrosine phosphorylation status of TSHR.
PDK1 Subcellular Localization and Tyrosine Phosphorylation—When PI3K is activated by growth factors, several events follow. These include the increased synthesis of PtdIns-3,4,5-P3 and PtdIns-3,4-P2, activation of PDK1, and translocation of PDK1 and Akt/PKB to the plasma membrane (
). Therefore, the localization of PDK1 was examined in the TSH-treated cells expressing GFP-tagged PDK1. In the cells treated with TSH or 8-Br-cAMP, PDK1 rapidly redistributes to the plasma membrane and the perinuclear region. This redistribution was inhibited by wortmannin, LY294002, and H89 (Fig. 3A). In the control cells, GFP fluorescence was evenly distributed within the cytoplasm.
Because tyrosine phosphorylation is associated with the activation of PDK1 (
), the TSH-treated cells were examined for the phosphorylation state of PDK1. The Myc-tagged PDK1 and TSHR were expressed in CHO cells, and tyrosine phosphorylation was analyzed using anti-phosphotyrosine antibodies (4G10). Myc-PDK1 was constitutively tyrosine-phosphorylated (Fig. 3B) in the TSH or 8-Br-cAMP treated or untreated cells. Therefore, in the CHO cells, the tyrosine phosphorylation of PDK1 was not correlated with the cellular localization or TSH treatment.
Akt/PKB and S6K1 Activities in TSH-treated Cells—Akt/PKB and S6K1 are substrates of PDK1, which are important downstream effectors of PI3K/PDK1 signaling. Therefore, Akt/PKB and S6K1 phosphorylation were examined in the TSH-treated thyroid cells. Previous studies show that Akt/PKB is phosphorylated on Thr-308 and Ser-473 by PDK1 and PDK2, respectively, in insulin-treated cells. Insulin treatment had similar effects on the thyroid cells (Fig. 4B and data not shown). However, TSH treatment did not induce Akt/PKB Thr-308 and Ser-473 phosphorylation in the thyroid cells (Fig. 4A). This result was confirmed by comparing the phosphorylation of 4EBP-1/PHAS-1, which is mediated by Akt/PKB, in the insulin- and TSH-treated thyroid cells. Insulin rapidly stimulated the phosphorylation of S65 of 4EBP-1/PHAS-1 but TSH did not (Fig. 4, D and C). Rapamycin, an inhibitor of FRAP/mTOR, inhibited the phosphorylation of 4EBP-1/PHAS-1 in the insulin-treated thyroid cells (Fig. 4D).
S6K1 has two protein subunits, p70 and p85, both of which are phosphorylated in insulin-treated cells. Fig. 5 shows that TSH induces the rapid phosphorylation of p70 and p85, as detected by the slow-migrating hyperphosphorylated protein species on a Western blot. The appearance of these species correlated with the increased S6 kinase activity in the immunoprecipitates (Fig. 5, A and B). The phosphorylation of p70 and p85 increases with increasing TSH (1 milliunit/ml) or insulin (10 μg/ml) exposure time (Fig. 5, A and C). After 30 min of exposure to TSH, the S6K1 activity reached a maximum, increasing 3.5-fold (Fig. 5B). The insulin-induced phosphorylation of S6K1 (Fig. 5C) reached a maximum 15 min after treatment (Fig. 5D). These experiments suggest that TSH activates S6K1 but not Akt/PKB; in contrast, insulin activates Akt/PKB and S6K1.
Roles of PI3K, PDK1, and PKA in Activating S6K1 in TSH-treated Cells—The above results suggest that TSH may stimulate a response in thyroid cells, which involves TSHR, PI3K, PDK1, and S6K1. The activities upstream of S6K1 were identified using the inhibitors of PI3K (wortmannin and LY294002), PKA (H89), FRAP/mTOR (rapamycin), and PDK1 (TPCK). LY294002 (500 nm) and wortmannin (300 nm) completely inhibited the TSH-induced phosphorylation of S6K1 (Fig. 6, A and B). LY294002 and wortmannin began to inhibit S6K1 phosphorylation at 500 and 100 nm, respectively (Fig. 6, A, lane 3, and B, lane 4). Forskolin (10 μm) or 8-Br-cAMP (1 mm) rapidly induced the phosphorylation of S6K1, and this effect was inhibited by LY294002 (500 nm) and wortmannin (100 nm) or H89 (Fig. 6, D and E). The inhibitory effect of H89 was dose-dependent (Fig. 6F). H89 began to inhibit TSH-induced phosphorylation of S6K1 at 5 μm and completely inhibited S6K1 (and CREB; data not shown) phosphorylation at 50 μm. The PDK-1 inhibitor TPCK also inhibited the TSH-induced phosphorylation of p70S6K1 (Fig. 7, A and B) at 20–50 μm and inhibited the phosphorylation of the ribosomal protein, S6, in a dose-dependent manner (Fig. 7A).
The role of PI3K in TSH-stimulated activation of S6K1 was also examined using the cells expressing a dominant negative form of HA-tagged p85α (cell line L5-Δ iSH2p85 expressing pCMV-ΔiSH2p85-HA). The cells expressing the wild-type p85α (cell line L5-WTp85 expressing pCMV-p85-HA) were used as the control (Fig. 6C) (
). The phosphorylation of S6K1 was measured in the presence and absence of TSH. In the cells expressing the wild-type p85, S6K1 was rapidly phosphorylated in response to TSH, but S6K1 phosphorylation was strongly reduced in the cells expressing the dominant negative p85 (Fig. 6C). Similar results were observed in the cells treated with forskolin or 8-Br-cAMP (data not shown).
Roles of Mitogen-activated Protein Kinase and FRAP/mTOR in Activating S6K1 in TSH-treated Cells—S6K1 is regulated by phosphorylation and has at least eight phosphorylation sites. PDK1, Akt/PKB, NEK6/7, FRAP/mTOR, and MAP kinases act upstream of S6K1. In vitro, S6K1 is phosphorylated by the MAP kinases (
). The thyroid cells were cultured in the presence or absence of TSH, and Erk1 and Erk2 phosphorylation were examined. Insulin (10 μg/ml) induced Erk1 and Erk2 phosphorylation at threonine 202 and tyrosine 204 after 15 min (Fig. 8A). In contrast, TSH did not induce Erk1 and Erk2 phosphorylation until 60 min after treatment (Fig. 8B). In the cells cultured in the absence of TSH, insulin, and serum, S6K1 was not phosphorylated, but Erk1/Erk2 were constitutively phosphorylated (Fig. 8). The TSH-induced S6K1 phosphorylation was not inhibited by a pretreatment with a MEK inhibitor, PD98059 (Fig. 8C, lanes 2 and 4). These results suggest that the MAP kinase pathway does not play a role in the TSH-stimulated phosphorylation of S6K1 in thyroid cells.
Rapamycin completely inhibited the S6K1 phosphorylation in the cells treated with TSH (Fig. 8C), forskolin, or 8-Br-cAMP (Fig. 6, D and E), suggesting a role for FRAP/mTOR in the response to TSH in thyroid cells.
Regulation of S6K1 by TSH Receptor Autoantibodies—There are two types of TSH receptor autoantibodies: the stimulating (TSAb) antibodies, which are agonists of the TSH receptor and associated with Graves disease, goiter, and hyperthyroidism; and the blocking (TSBAb) antibodies, which are antagonists of the TSH receptor and are associated with primary myxedema, thyroid atrophy, and hypothyroidism. The influence of the stimulating and blocking TSH receptor autoantibodies on S6K1 phosphorylation was examined in thyroid cells. Normal pooled IgG (NP) was obtained from normal adults with no known thyroid disease and were used as the controls. Autoantibodies from patients with Graves disease (GD) significantly enhanced phosphorylation of S6K1, but the control IgG (NP) did not (Fig. 9A). IgG from 10 patients with Graves disease were tested in a radioreceptor assay for TSH-binding inhibitory immunoglobulin (TBII). Seven patients with TBII-positive Graves disease stimulated S6K1 phosphorylation and the ribosomal protein, S6 (Fig. 9A). This activity was completely inhibited by LY294002 and wortmannin but not by PD98059 (Fig. 9B). TSBAb IgG from patients with primary myxedema (PM) did not induce S6K1 phosphorylation (Fig. 9A, lane PM). TSBAb IgG did not inhibit the insulin-induced phosphorylation of S6K1 but did inhibit the TSH/TSAb-induced phosphorylation of S6K1 in thyroid cells (Fig. 9C, lanes 4–6).
Effects of S6K1 Signaling Pathways on Thyrocyte Proliferation—In thyroid cells, the cellular proliferation is regulated by TSH, insulin, and other growth factors (
). The role of signaling kinases in TSH-stimulated proliferation was examined by measuring the level of DNA synthesis in the presence of wortmannin, LY294002, rapamycin, or H89. Thyroid cells were cultured in the absence of TSH for 7 days and then stimulated with TSH in the absence or presence of the kinase inhibitor and [3H]thymidine. The incorporation of [3H]thymidine increased 17-fold in the cells exposed to TSH (Fig. 10A). Rapamycin (20 nm), wortmannin (100 nm), LY294002 (500 nm), and H89 (50 μm) completely inhibited TSH-stimulated DNA synthesis.
TSH regulates the expression of cyclins D and E, which are rate-limiting for entry into the S phase. Cyclin D1 is expressed at a low basal level in the absence of TSH. The expression of cyclin D1 increases ∼12 h after exposure to TSH and decreases to the basal level after 24 h post-exposure (Fig. 10B). Rapamycin inhibits TSH-induced cyclin D1 expression and does not change the basal level of cyclin D1 (Fig. 10C, lanes 1 and 2). This result is consistent with previous observations showing that rapamycin inhibits the G1/S transition and prolongs G1 in several types of cells (
The cell cycle distribution was also examined in TSH/insulin-treated thyroid cells in the presence and absence of rapamycin. The cells maintained in 3H medium with 5% calf serum (3H5%), which lacks insulin, TSH, and somatostatin, were quiescent (G1, 82.9 ± 5%; S, 1.88 ± 0.7%) (Fig. 10D). The TSH/insulin treatment promoted cell proliferation (G1, 49.3 ± 5%; S, 19.5 ± 0.9%) (Fig. 10D). Rapamycin partially inhibited this change in the cell cycle distribution (G1, 72.6 ± 0.3%) (Fig. 10D).
The phosphorylation of S6K1 was also examined in the experiment shown in Fig. 10. TSH stimulated a lower level of S6K1 phosphorylation than the TSH/insulin (Fig. 10D, lane 4), and rapamycin completely inhibited S6K1 phosphorylation in the presence of TSH or TSH/insulin (lanes 3 or 5).
Regulation of Thyroid Follicle Activity by Rapamycin in Vivo—The above results suggest that rapamycin-sensitive S6K1 signaling plays a major role in the TSH-stimulated thyrocyte proliferation. The following experiments test the in vivo role of FRAP/mTOR and S6K1 in the TSH-stimulated thyroid follicles. The in vivo experimental system involved feeding male Sprague-Dawley rats water containing 0.025% MMI; this protocol provides a long term TSH stimulation of the thyroid follicles. In the MMI-treated rats, the thyroid gland showed a prominent dark red swelling after 2 weeks of treatment. The thyroid follicle histology revealed the characteristic findings of long term TSH stimulation. These include the depletion of colloids, small hyperplastic collapsed follicles, hypertrophic columnar lining cells, and crowded hyperchromatic nuclei (Fig. 11A, 2nd panel). The histology was similar in rats receiving intraperitoneal injections of 2% carboxymethylcellulose for 1 week. However, the histology was altered in the thyroid glands of the MMI-treated animals administered 2% carboxymethyl-cellulose containing rapamycin (1.5 mg/kg) once daily. In these animals, colloid-filled follicles were more abundant (Fig. 11A, 4th, and B).
The S6K1 kinase activity was estimated in these rats by measuring the quantity of phosphorylated ribosomal protein S6. In the untreated rats, phosphorylated S6 was not present in the thyroid follicular cells but was abundant in the parafollicular cells (Fig. 11C, left panel). In the MMI-treated rats, a significantly higher percentage of thyroid cells possessed the phosphorylated S6 protein, (Fig. 11C, middle panel), and rapamycin completely inhibited S6 phosphorylation in the thyroid cells (Fig. 11C, right panel). These observations suggest that rapamycin-sensitive S6K1 signaling plays a role in the TSH-stimulated follicles in vivo.
The mechanisms of cellular proliferation in the thyroid gland may be relevant to human thyroid diseases including benign thyroid nodules, differentiated thyroid cancer, and Graves disease (
). Thyrocyte proliferation is regulated by TSH, which is one of the most important factors regulating the proliferation and function of thyroid cells, and by other hormones and growth factors that activate the specific signaling pathways (
). TSH activates TSHR, which activates adenylyl cyclase, elevates cAMP, and activates PKA. However, TSH regulates the thyroid function by activating a complex signaling network.
This study presents evidence showing that PI3K activates S6K1 in response to TSH in the proliferating thyroid cells (Fig. 12). TSHR interacts with the p85α regulatory subunit of PI3K and has inducible tyrosine phosphorylation sites (Fig. 2D). It is possible that a phosphotyrosine residue in the intracellular loop or C-terminal tail of TSHR interacts with the SH2 domain of p85α. Tyrosine phosphorylation of TSHR is stimulated by exogenous 8-Br-cAMP and might involve a cAMP-dependent tyrosine kinase (Fig. 2D). The tyrosine kinase inhibitor genistein inhibits the tyrosine phosphorylation of TSHR in the cells treated with TSH/8-Br-cAMP and inhibits the association between PI3K and TSHR (data not shown). These observations suggest that TSH/cAMP stimulates the tyrosine phosphorylation of TSHR, which may stimulate the TSHR-dependent activation of PI3K in thyroid cells. A dominant negative p85α, which lacks the iSH2 region that binds the p110 catalytic domain of PI3K, strongly inhibits the TSH-stimulated phosphorylation of S6K1. This suggests that a class I PI3K may be involved in TSH signaling. To date, there is no evidence showing that the adaptor molecules Shc, Grb2, and Gab2 facilitate the interaction between TSHR and PI3K, although these molecules bridge the interactions between other the receptors and PI3K (
This study suggests that PKA may be required for the TSH/cAMP-stimulated phosphorylation of S6K1 in thyroid cells. LY294002 and wortmannin did not affect the TSH-stimulated phosphorylation of CREB (data not shown), suggesting that PI3K acts downstream of PKA. However, H89 inhibits the 8-Br-cAMP/TSH-induced association between TSHR and PI3K. Therefore, cAMP/PKA may stimulate the formation of an active TSHR·PI3K complex or modify the subunit composition of PI3K. Recent studies show that cAMP/PKA phosphorylates p85α and stabilizes the p85α-p110 complex (
This study suggests that PDK1 plays a role in the TSH-stimulated activation of S6K1 (Figs. 3 and 7). PDK1 is regulated primarily by the substrate conformation, cellular localization, and tyrosine phosphorylation (
). PDK1 redistributes from the cytosol to the plasma membrane and perinuclear region in the TSH-treated cells. This process is inhibited in the presence of the PI3K inhibitors and might involve the PI3K lipids because it requires the pleckstrin homology domain of PDK1. PDK1 phosphorylation may also regulate its activity (
), although TSH did not stimulate the tyrosine phosphorylation of PDK1 in the thyroid cells overexpressing PDK1 (data not shown) or in the CHO cells overexpressing TSHR (Fig. 3B). S6K1 (p70 and p85) and Akt/PKB are well known targets of PDK1 in the growth factor signaling pathways (
). It is not yet understood why TSH does not induce the Thr-308- or Ser-473-phosphorylated forms of Akt/PKB (Fig. 4 and Fig. 12) but does induce S6K1 phosphorylation. Protein-protein interactions or structural conformation may influence this specificity. TSH and insulin also have different effects on these signaling pathways, which may explain why the thyroid cells require both TSH and insulin for cellular proliferation.
4E-BP1 (also known as PHAS-1) normally binds eIF4E, initiating cap-dependent phosphorylation (
). Insulin induces the phosphorylation of Ser-65 of 4E-BP1, and this phosphorylation is inhibited by rapamycin. This suggests that insulin induces FRAP/mTOR, which phosphorylates the Ser-65 of 4E-BP1. In contrast, TSH did not induce the phosphorylation of Ser-65 of 4E-BP1. Akt/PKB may also be required for the phosphorylation and activation of 4E-BP1, and TSH does not activate Akt/PKB, but insulin does. These findings suggest that FRAP/mTOR is a required downstream effector of Akt/PKB, which phosphorylates 4E-BP1 (
) on WRT thyroid cells showed that cAMP stimulates S6K1 through a PKA-dependent, PI3K-independent pathway. This conclusion was based on the observation that cAMP stimulates the phosphorylation of the S6 protein and was repressed by H89 (50 μm) but not by low-dose wortmannin or by a microinjection of the N-terminal SH2 domain of the p85 regulatory subunit of PI3K. Cohen and co-workers (
) reported that H89 (10 μm) is equally effective in inhibiting MSK1, S6K1, ROCK-II, and PKA. Therefore, H89 might inhibit S6K1 directly. However, in this study, H89 inhibited the association between TSHR and PI3K and the activation of PI3K (Fig. 1), suggesting that PKA plays a role in the TSHR-stimulated activation of PI3K in thyroid cells.
In contrast to the above results in the WRT thyroid cells, this study shows that S6K1 activation by cAMP is dependent on PI3K in the FRTL-5 thyroid cells. Wortmannin and LY294002 are structurally unrelated molecules that, at low concentrations, are relatively specific, cell-permeable inhibitors of PI3K. Wortmannin may directly inhibit the FRAP/mTOR autokinase activity at an IC50 ∼100-fold higher than its IC50 for inhibiting PI3K (∼200 nmin vitro, ∼300 nmin vivo). LY294002 inhibits FRAP/mTOR autokinase activity in vitro with an IC50 of 5 μm (
). LY294002 and wortmannin inhibit the TSH-stimulated phosphorylation of S6K1 at 500 nm and 100 nm, respectively, which suggests the involvement of PI3K (Fig. 6).
FRAP/mTOR regulates the protein translation by two independent mechanisms involving the direct or indirect activation of S6K1 and the inactivation of the translational repressor PHAS-1/4E-BP1. Rapamycin completely inhibits the phosphorylation of S6K1 induced by TSH, 8-Br-cAMP, forskolin, and insulin in FRTL-5 thyroid cells (Fig. 6). The mechanisms by which TSH regulates FRAP/mTOR kinase is not known. However, FRAP/mTOR senses the cellular ATP level (
), and 2-deoxyglucose (100 mm), which is an inhibitor of the glycolytic pathway, partially inhibits S6K1 phosphorylation in the TSH-treated cells (data not shown). Rapamycin may inhibit the S6K1 phosphorylation by activating the phosphatases in the TSH-treated cells (
). The TSH-induced changes in the cellular ATP level might affect the TSH-induced phosphorylation of S6K1 in thyroid cells. These findings suggest that FRAP/mTOR kinase may not act down-stream of PI3K/PDK1 in TSH-treated cells. It is possible that FRAP/mTOR senses the nutrient signals that change in the TSH-treated cells.
Different thyrocyte cell systems have different requirements for cell proliferation (i.e. TSH, insulin/IGF-1, serum). TSH and insulin stimulate the proliferation by mechanisms, which are not completely clear. In dog thyroid cells, cAMP activates S6K1 but does not promote DNA synthesis (
). TSH also activates S6K1 and increases the number of cells in the S phase slightly but does not significantly increase the number of cells in G2 (Fig. 10). In contrast, TSH/insulin markedly increases the number of cells in S and G2 and TSH/insulin stimulates the synergistic prolonged activation of S6K1 (Fig. 10D, lane 4). Rapamycin inhibits the TSH and TSH/insulin-induced cell cycling, suggesting a role for the S6K1 pathways. However, rapamycin does not inhibit the activation of Akt/PKB in the insulin-treated thyroid cells (data not shown), suggesting that the synergistic activation of the S6K1 signaling pathways by TSH/insulin may stimulate the cell cycle progression in FRTL-5 thyroid cells.
) reported that Akt/PKB was not phosphorylated in the TSH-treated FRTL-5 thyroid cells. A constitutively active Akt/PKB also promotes the hormone-independent proliferation in the PC C1 3 thyroid cells (
). These observations suggest that Akt/PKB might play a role in thyrocyte proliferation. However, the role of Akt/PKB in TSH and TSH/insulin-treated thyroid cells remains to be clarified.
This study used Sprague-Dawley rats maintained on MMI-containing drinking water as an in vivo experimental system to study the long term stimulation of the thyroid follicles. The MMI-treated rats had higher TSH levels in response to the inadequate thyroid hormone production. The thyroid glands from the MMI-treated rats have a follicular structure that suggests TSH stimulation, and the follicular cells have an increased phosphorylation of the ribosomal protein S6 (Fig. 11C). These findings suggest that the endogenous TSH stimulates S6K1 in the thyroid follicles. Follicular changes induced by sustained TSH stimulation were partially reversed by the short term intraperitoneal delivery of rapamycin. This suggests that some of the effects of TSH on the thyroid follicles are mediated by S6K1.
This study also shows that the TSHR antibodies modulate the S6K1 activity in thyroid cells. TSAb induces and TSBAb inhibits the phosphorylation of S6K1 and the S6 protein. The TSAb-stimulated phosphorylation of S6K1 was inhibited by LY294002 and wortmannin but not by PD98059. This suggests that the TSAb-induced activation of S6K1 requires PI3K. However, it was not possible to determine whether the S6K1 and TSAb activities correlated with each other in this study as a result of insufficient data. Nevertheless, the TSAb-stimulated activation and TSBAb-stimulated inhibition of S6K1 might be related to goiter and atrophy in Graves disease and primary myxedema, respectively.
We thank Dr. John Blenis for providing the polyclonal S6K1 antibodies, Dr. Leonard D. Kohn for helpful discussion and critical review of the manuscript, and Yong Mi Kang for technical assistance.