TSH/IGF-1 Receptor Cross-Talk Rapidly Activates Extracellular Signal-Regulated Kinases in Multiple Cell Types

Abstract

We previously showed that thyrotropin (TSH)/insulinlike growth factor (IGF)-1 receptor cross-talk appears to be involved in Graves’ orbitopathy (GO) pathogenesis and upregulation of thyroid-specific genes in human thyrocytes. In orbital fibroblasts from GO patients, coadministration of TSH and IGF-1 induces synergistic increases in hyaluronan secretion. In human thyrocytes, TSH plus IGF-1 synergistically increased expression of the sodium-iodide symporter that appeared to involve ERK1/2 activation. However, the details of ERK1/2 activation were not known, nor was whether ERK1/2 was involved in this synergism in other cell types. Using primary cultures of GO fibroblasts (GOFs) and human thyrocytes, as well as human embryonic kidney (HEK) 293 cells overexpressing TSH receptors (HEK-TSHRs), we show that simultaneous activation of TSHRs and IGF-1 receptors (IGF-1Rs) causes rapid, synergistic phosphorylation/activation of ERK1 and ERK2 in all three cell types. This effect is partially inhibited by pertussis toxin, an inhibitor of TSHR coupling to G_i/G_o proteins. In support of a role for G_i/G_o proteins in ERK1/2 phosphorylation, we found that knockdown of G_i(1–3) and G_o in HEK-TSHRs inhibited ERK1/2 phosphorylation stimulated by TSH and TSH plus IGF-1. These data demonstrate that the synergistic effects of TSH plus IGF-1 occur early in the TSHR signaling cascade and further support the idea that TSHR/IGF-1R cross-talk is an important mechanism for regulation of human GOFs and thyrocytes.

We showed that TSH/IGF-1 receptor cross-talk leads to rapid, synergistic stimulation of ERK1/2 phosphorylation in three human cell types.

Cross-talk between G protein–coupled receptors (GPCRs) and receptor tyrosine kinases (RTKs) is generally defined as interactions between these signaling pathways leading to an additive or synergistic effect that occurs rapidly after agonist binding at a step proximal to or at the receptors (1–3). In this construct, cross-talk is thought to occur by one of three mechanisms: (1) activation of GPCR signaling by its agonist leads to secretion of an agonist for a RTK on the same cell, (2) activation of GPCR leads to Tyr phosphorylation of the RTK, or (3) RTK signaling uses components of the GPCR transduction pathway (e.g., G proteins, β-arrestins). It is possible, however, for interactions leading to additive or synergistic effects on a biologic response to occur more slowly at a step downstream of the receptor in the signaling cascade.

Positive interactions in responses to simultaneous stimulation by thyrotropin (TSH) and insulinlike growth factor (IGF)-1 have been found in a number of cell types. In rodent FRTL-5 cells, IGF-1 was found to synergize with TSH to increase [³H]thymidine incorporation into DNA (4). In sheep thyrocytes, TSH plus IGF-1 was found to stimulate thyroid function, whereas either hormone alone was ineffective (5). Physical interactions between TSH receptors (TSHRs) and IGF-1 receptors (IGF-1Rs) have been shown in primary cultures of fibroblasts/preadipocytes obtained at decompression surgery from patients with Graves’ orbitopathy (GO) and in human thyrocytes (6).

We have studied the effects of simultaneous activation of TSHRs and IGF-1Rs in human thyrocytes and in GO fibroblasts (GOFs). In human thyrocytes, we showed that simultaneous stimulation by TSH and IGF-1 caused a synergistic increase in phosphorylation of ERK1/2 and of sodium-iodide symporter (7) and that the increased sodium-iodide symporter messenger RNA (mRNA) could be blocked by an inhibitor of mitogen-activated protein kinase kinase 1 (MEK1), which is the enzyme that phosphorylates ERK1/2 (6). We found that simultaneous stimulation by TSH and IGF-1 caused a synergistic response in GOFs also (8, 9). Specifically, we showed that IGF-1 caused a marked increase in potency of TSH to stimulate secretion of hyaluronan [hyaluronic acid (HA)] and an increase in the amount of HA secreted (efficacy) that was more than the additive effects of TSH and IGF-1 alone. [Accumulation of HA in the retro-orbital space is a major component of the pathology of GO (10).] We have referred to this phenomenon as “TSHR/IGF-1R cross-talk” based on the observation that an IGF-1R inhibitor inhibits TSH stimulation of gene expression in human thyrocytes and HA secretion by GOFs even though the effects on gene expression and HA secretion were observed hours to days after activation of these receptors.

Herein, we extend these observations to show that IGF-1 increases the effect of TSH to rapidly phosphorylate ERK1/2 in GOFs and in human embryonic kidney (HEK) 293 cells overexpressing TSHRs (HEK-TSHRs) and confirm this effect in human thyrocytes. Moreover, we show that this effect is dependent in part on receptor coupling to G_i/G_o proteins. We use these additional cell types to provide evidence in support of the idea that TSHR/IGF-1R cross-talk is a general phenomenon.

Materials and Methods

Materials

Dulbecco’s modified Eagle medium (DMEM), 100-fold penicillin-streptomycin solution, l-glutamine, 1 M HEPES buffer, and Hanks’ balanced salt solution (HBSS) were obtained from Mediatech Inc. (Manassas, VA). Hydrocortisone, insulin, bovine thyrotropin (bTSH), cholera toxin, pertussis toxin (PTX), Fungizone, Tween^® 20, PhosSTOP phosphatase inhibitor cocktail tablets, and cOmplete protease inhibitor tablets were obtained from Sigma-Aldrich (St. Louis, MO). Fetal bovine serum (FBS), Ham’s F-12 nutrient mixture, hygromycin, NuPAGE precast Bis-Tris gels, and NuPAGE MOPS sodium dodecyl sulfate (SDS) running buffer were obtained from Thermo Fisher Scientific (Waltham, MA). Recombinant human epithelial growth factor and recombinant human IGF-1 were obtained from PeproTech (Rocky Hill, NJ). Y-27632 dihydrochloride was obtained from Bio-Techne (Minneapolis, MN). BSA was obtained from MP Biomedicals (Santa Ana, CA). Radioimmunoprecipitation assay buffer and rabbit monoclonal anti-GAPDH antibody (clone EPR16891) were obtained from Abcam (Cambridge, MA). Mouse monoclonal antiphospho-ERK 1/2 (clone E10), and mouse monoclonal antiphospho-Akt (clone 587F11) antibodies were purchased from Cell Signaling Technologies (Danvers, MA). Odyssey blocking buffer and IRDye secondary antibodies were purchased from Li-COR Biotechnology (Lincoln, NE). Phospho-ERK1 (T202/Y204)/ERK2 (T185/Y187) DuoSet IC enzyme-linked immunosorbent assay (ELISA) and human/mouse/rat total ERK2 DuoSet IC ELISA were purchased from R&D Systems (Minneapolis, MN), and 1 MDa HAwas purchased from Lifecore Biomedical (Chaska, MN). HA test kits were purchased from Corgenix (Broomfield, CO). DharmaFECT transfection reagent and all inhibitory RNA were received from Dharmacon (Lafayette, CO).

Primary cell isolation and cell culture

GOFs were isolated as previously described (11) from retro-orbital adipose tissue obtained from GO patients who underwent orbital decompression surgery at the joint thyroid–eye clinic of the Johannes Gutenberg University Medical Center ( Mainz, Germany). The collection of patients’ samples has been approved by the Ethical Committee of the Medical Chamber of the State Rhineland-Palatinate, Germany, and by the Institutional Review Board of the Johannes Gutenberg University Medical Center. Written informed consent was received from all patients with GO prior to collection. Use of human tissues was approved by the National Institute of Diabetes and Digestive and Kidney Diseases Institutional Review Boards. Briefly, monolayer outgrowths of adherent cells from minced tissue were serially passaged with trypsin/EDTA and cultured in F-media comprised of DMEM with FBS [10% volume-to-volume ratio (v/v)], penicillin (100 U/mL), streptomycin (100 μg/mL), l-glutamine (2 mM), Ham F-12 nutrient mixture (25% v/v), hydrocortisone (25 ng/mL), epithelial growth factor (0.125 ng/mL), insulin (5 μg/mL), cholera toxin (11.7 nM), gentamicin (10 μg/mL), Fungizone (250 ng/mL), and Y-27632 (5 μM). Cells were maintained in a humidified 7% CO₂ incubator at 37°C. For experiments, GOFs were plated to confluence in DMEM containing FBS (10% v/v), penicillin (100 U/mL), and streptomycin (100 μg/mL) the day before and used no later than passage 3.

Human thyrocytes were isolated as previously described (7). Briefly, thyroid tissue samples were collected from normal thyroid tissue from patients undergoing total thyroidectomy for thyroid cancer at the National Institutes of Health Clinical Center. Patients provided informed consent on an Institutional Review Board–approved protocol, and materials were received anonymously with approval of research activity through the Office of Human Subjects Research, National Institutes of Health. Adherent cells were isolated from tissue following collagenase digestion. Cells were maintained in DMEM supplemented with FBS (10% v/v), penicillin (100 U/mL), and streptomycin (100 μg/mL) at 37°C in a humidified 5% CO₂ incubator. For experiments, thyrocytes were plated in 12-well plates at 1.5 × 10⁵ cells per well in DMEM containing 10% FBS, then stepped-down for 24 hours in in serum-free media comprised of DMEM, BSA (0.1%), penicillin (100 U/mL), and streptomycin (100 μg/mL).

HEK-TSHR cells were generated as previously described (12). HEK-TSHR cells were cultured at 37°C in a 5% CO₂ incubator and grown in DMEM supplemented with FBS (10% v/v), penicillin (100 U/mL), and streptomycin (100 µg/mL). Hygromycin (250 µg/mL) was used as a selection marker and was supplemented in all HEK-TSHR media unless mentioned otherwise. For all HEK-TSHR experiments, cells were seeded in 24-well plates at a density of 220,000 cells per well. Twenty-four hours after seeding, cells were stepped-down for an additional 24 hours in serum-free media.

Phospho-kinase stimulation in cells

GOFs were stepped down in serum-free media for 24 hours. For experiments testing ERK effects on crosstalk, GOFs were treated with increasing doses of bTSH with or without IGF-1 (100 ng/mL) for 5 minutes at 37°C. Following stimulation, cells were washed two times with ice-cold HBSS containing 10 mM HEPES, lysed with radioimmunoprecipitation assay buffer supplemented with PhosSTOP and cOmplete inhibitor cocktails, and assayed by Western blot. Thyrocytes and HEK-TSHR cells were plated and stepped-down as stated previously. Cells were stimulated for 5 minutes at 37°C with increasing doses of bTSH with and without IGF-1. Following stimulation, cells were washed two times with ice-cold HBSS containing 10 mM HEPES, then lysed and assayed by phospho-ERK1/2 and total ERK 2 ELISA according to the manufacturer’s directions.

Immunoblotting analysis

Whole-cell lysates were separated by SDS polyacrylamide gel electrophoresis under reducing conditions on 4% to 12% Precast NuPAGE gels using MOPS SDS running buffer, then transferred to nitrocellulose membranes using a Mini Trans-Blot^® electrophoretic transfer cell (Bio-Rad). Membranes were blocked with Odyssey blocking buffer. Antibody mixtures were diluted with Odyssey blocking buffer with 0.1% Tween® 20 and incubated with membranes with gentle shaking overnight at 4°C. Membranes were washed with TBS containing 0.1% Tween® 20. IRDye secondary antibodies were diluted 1:20,000 in Odyssey blocking buffer with 0.1% Tween® 20 for 30 minutes, washed with TBS containing 0.1% Tween® 20, then TBS, and imaged on an Odyssey CLx using the AutoScan function. Densitometry was performed using ImageJ. Phosphorylated protein was normalized to total GAPDHand normalized to maximal bTSH only response.

G_i and G_o inhibition with pertussis toxin

GOFs, thyrocytes, and HEK-TSHR cells were seeded as described previously, then stepped-down in serum-free media with or without PTX (100 ng/mL). Following 24 hours of starvation and PTX pretreatment, cells were stimulated 5 minutes at 37°C, washed, and assayed by ELISA as described previously.

Small inhibitory RNA mediated knockdown of G_i and G_o

HEK-TSHR cells were seeded in 10-cm dishes at a density of 2.3 × 10⁶ cells per dish. Following 24 hours of incubation, cells were transfected using DharmaFECT transfection reagent 1, along with 2 µM G_i1-3 small inhibitory RNA (siRNA), 2 µM G_o siRNA, or 2 µM nonsilencing RNA. Cells were incubated for 24 hours in a transfection cocktail consisting of DMEM supplemented with transfection reagent and RNA, and were then split into 24-well plates in normal culture conditions omitting hygromycin. After 24 hours, cells were serum-starved as described previously. After 24 hours of serum-starvation, cells were stimulated for 5 minutes at 37°C. Total RNA was collected at this time, and extent of knock-down determined by qPCR. Stimulated cells were washed two times with HBSS containing 10 mM HEPES, lysed, and assayed according to manufacturer instructions.

Hyaluronan stimulation in Graves’ orbital fibroblasts

GOFs were plated at confluence then pretreated with U0126 (1 µM) for 1 hour in serum-free media. GOFs were treated with increasing doses of bTSH, with or without IGF-1 (100 ng/mL) and/or U0126 (1 µM) in DMEM containing 10% FBS and 10 mM HEPES, pH 7.4 for 4 days. Conditioned media were collected, stored at −20°C, and assayed using a modified Corgenix HA ELISA kit as previously described (8).

Statistical analysis

Statistical analysis was performed using GraphPad Prism version 7.02 for Windows (GraphPad Software, San Diego, CA). Concentration curves were fit to a three-parameter dose-response curve, and statistical analysis comparing EC₅₀s was performed using the extra sum-of-squares F test, with P ≤ 0.05 considered significant. Student t test was used to compare the effects of PTX, G_i1-3 siRNA, and G_o siRNA.

Results

TSH receptor signaling has been shown to involve several pathways, including ERK1/2 (7, 13) and Akt (14, 15). Both pathways were found to be activated when cells were stimulated by TSH and IGF-1 simultaneously (16), and we sought to determine whether they play a role in TSHR/IGF-1R crosstalk. Unlike previously reported overexpression systems, GOF stimulation by a maximal dose of TSH did not result in significant Akt phosphorylation compared to basal, even after 30 minutes (Supplemental Fig. 1A ^{(90.1KB, pdf)} ). Akt phosphorylation was stimulated by IGF-1, but TSH/IGF-1 combination treatment was not significantly different than IGF-1 alone. In contrast, both TSH and IGF-1 individually stimulated ERK1/2 phosphorylation, where combined TSH and IGF-1 led to greater-than-additive effects at short times (Supplemental Fig. 1B ^{(90.1KB, pdf)} ). To show that ERK phosphorylation was involved in stimulation of HA secretion by GOFs, we used an inhibitor of MEK1, the enzyme that phosphorylates ERK1/2. Fig. 1 illustrates the effects of a MEK1 inhibitor on stimulated HA secretion. The MEK1 inhibitor U0126 blocked TSH stimulation of HA secretion and partially inhibited TSH plus IGF-1 stimulation. Thus, ERK1/2 appeared to play a role in TSH and TSH plus IGF-1 stimulation of HA secretion.

Figure 1.

MEK inhibitor reduces bTSH-dependent HA secretion in GOFs. Secreted HA was measured in GOFs incubated for 4 days with 0.54 nM to 5.4 µM bTSH (1.8 µM = 100 mU/mL) alone (filled circles), with 100 ng/mL IGF-1 (filled squares), with 1 µM U0126 (unfilled circles), and with 100 ng/mL IGF-1 plus 1 µM U0126 (unfilled squares). Data are plotted as the mean ± standard error of the mean of the percentage of the mean maximal bTSH only response; n ≥ 6 from three patient strains. EC₅₀ for bTSH and bTSH + IGF-1 concentration curves are 117 nM and 2 nM, respectively, and are significantly different (F test P < 0.0001).

We previously reported that TSH + IGF-1 crosstalk leading to stimulate HA secretion by GOFs manifested in a shift in potency for the TSH concentration curve in addition to an increase in TSH efficacy. We next looked to see if a similar phenomenon applied to ERK phosphorylation. Fig. 2(a) is a representative Western blot for phosphorylated ERK1, ERK2, and Akt in GOFs treated with increasing doses of TSH with or without IGF-1. TSH alone had no effect on Akt phosphorylation whereas IGF-1 caused a robust increase in pAkt that was not affected by TSH. The densitometry from a compilation of these experiments on stimulation of ERK phosphorylation in five GOF strains is illustrated in Fig. 2(b) and 2(c). These data confirm that TSH or IGF-1 alone had modest effects on pERK1 and pERK2. Most importantly, IGF-1 caused synergistic increases in TSH-stimulated pERK1 and pERK2. In the presence of IGF-1, TSH stimulated pERK1 and pERK2 1.5- to 2.0-fold over TSH alone. The potencies of TSH were also increased by IGF-1 from EC₅₀ of at least 12.0 µM to 0.2 µM for pERK1 (F test P = 0.0962) and from at least 2.4 µM to 0.1 µM for pERK2 (F test P = 0.0268). Thus, in GOFs, IGF-1 caused synergistic increases in the potencies and efficacies of TSH to stimulate ERK1 and ERK2 phosphorylation rapidly.

Figure 2.

Synergistic ERK phosphorylation in GOFs by bTSH + IGF-1. GOFs were incubated with 0.0018 to 1.8 µM of bTSH (1.8 µM = 100 mU/mL) with (filled squares) or without (filled circles) 100 ng/mL IGF-1 for 5 min at 37°C. (a) Phosphorylation of ERK1 and ERK2 were measured by Western blot. Labels describe bTSH dose in mU/mL. (b) pERK1 and (c) pERK2 stimulation in response to bTSH was determined by immunoblotting analysis. EC₅₀ for bTSH and bTSH + IGF-1 are 2.4 µM and 0.1 µM, respectively, and are significantly different (F test P = 0.0268). Data are shown as mean ± standard error of the mean of the percentage of the mean maximal bTSH only response (% bTSH max); n = 5 from five patient strains.

As TSHR is known to interact with all four classes of G proteins (17) and the majority of GPCRs that have been shown to cross-talk with IGF-1Rs have been shown to interact via G_i proteins (2), we determined whether TSHR/IGF-1R cross-talk in GOFs involves G_i/o proteins. We treated GOFs with PTX that uncouples G_i/G_o proteins from GPCRs. PTX has been shown to affect TSHR signaling in FRTL-5 rat thyroid cells (18). Figure 3(a) illustrates that pretreatment of GOFs with PTX overnight lowered basal pERK1 and pERK2 and inhibited stimulation by bTSH of pERK1 and pERK2 under all conditions. Inhibition of IGF-1 alone was significant for pERK1 but not pERK2, although a trend to inhibition was evident. To support our findings in GOFs, we determined whether PTX would affect ERK1/2 phosphorylation in human thyrocytes. Figure 3(b) shows that PTX pretreatment inhibited pERK1/2 in human thyrocytes under all conditions, including IGF-1 alone. Thus, in both human cell types expressing endogenous levels of TSHR, stimulation of ERK1/2 activation by TSH and TSH plus IGF-1 was inhibited by PTX.

Figure 3.

ERK1/2 stimulation by bTSH ± IGF-1 is inhibited by PTX in GOFs and thyrocytes. GOFs and thyrocytes were stimulated with HBSS, 100 ng/mL IGF-1, 0.0054 µM bTSH (bTSH_low), 1.8 µM bTSH (bTSH_max), or bTSH_max plus IGF-1 combination with (striped bars) or without (filled bars) 100 ng/mL PTX for 5 min at 37°C. Data are plotted as the mean ± standard error of the mean (SEM) of the percentage of the mean maximal bTSH only response (% bTSH max). (a) ERK1 and ERK2 stimulation in GOFs was determined by immunoblotting analysis. Data are shown as mean ± SEM; n = 7 from three patient strains. (b) ERK1/2 stimulation in thyrocytes was determined by ELISA, and data are shown as mean ± SEM; n = 5 from five patients. * P ≤ 0.05.

We decided to study HEK-TSHRs because we thought any effects on signaling pathways might be exaggerated in cells overexpressing TSHRs and that the involvement of PTX-sensitive G-proteins might be more readily delineated. Supplemental Fig. 3 ^{(90.1KB, pdf)} illustrates that IGF-1 alone had only a small effect on the level of pERK1/2, whereas TSH stimulated a rapid and robust increase in pERK1/2. IGF-1 further increased pERK1/2 in the presence of TSH at all time points measured. Figure 4 illustrates the dose-response of TSH to increase pERK1/2 and the effects of IGF-1 on the TSH dose-response curve. IGF-1 converted a monophasic curve for TSH stimulation of pERK1/2 into a biphasic curve (F test P = 0.0164) with a greater than 100-fold increase in potency for the higher potency component of the curve (EC₅₀=0.22 nM) compared with the lower potency phase in the presence of IGF-1 or the monophasic potency in the absence of IGF-1. We had previously found a 100-fold shift in potency in the concentration curve for TSH in the presence of IGF-1 for TSH stimulation of HA secretion by GOFs (8). However, we did not find a consistent effect of TSH or IGF-1 on HA secretion in HEK-TSHRs in which HA secretion was very low (not shown).

Figure 4.

Biphasic ERK phosphorylation by bTSH + IGF-1 in HEK-TSHR cells. HEK-TSHR cells were incubated with 0, 0.000018, 0.000054, 0.00018, 0.00054, 0.0018, 0.0054, 0.018, 0.054, 0.18, 0.54, 1.8, and 5.4 µM of bTSH (1.8 µM = 100 mU/mL) with (filled squares) or without (filled circles) 100 ng/mL IGF-1 for 5 minutes at 37°C. Data are shown as mean ± standard error of the mean of the percentage of the mean maximal bTSH only response (% bTSH max); n = 5. Concentration curve for bTSH + IGF-1 fit a biphasic model (dotted line); F test P = 0.0164.

To confirm in HEK-TSHRs that TSHR couples, in part, to ERK1/2 phosphorylation via a PTX-sensitive G_i/o protein, we studied the effects of PTX and of knockdown of all three G_i proteins and G_o. Figure 5 illustrates the effects of PTX pretreatment on TSH, IGF-1, and TSH plus IGF-1 stimulation of pERK1/2 levels. PTX had no effect on basal levels of pERK1/2 but partially inhibited stimulation by TSH, IGF-1, and TSH plus IGF-1. The finding that PTX did not decrease basal pERK1/2 levels but inhibited IGF-1 stimulation of pERK1/2 in the absence of TSH provided additional evidence of the cross-talk between IGF-1R and TSHR.

Figure 5.

PTX inhibition of bTSH and bTSH + IGF-1 mediated ERK phosphorylation in HEK-TSHR cells. HEK-TSHR cells were incubated for 24 hours in 100 ng/mL Pertussis toxin (striped bars) or in HBSS (filled bars). Following 24 hours incubation, cells were stimulated for 5 minutes at 37°C with HBSS, 0.0054 (bTSH_low) and 1.8 µM bTSH (bTSH_max;1.8 µM = 100 mU/mL) with and without 100 ng/mL IGF-1. ERK1/2 phosphorylation was measured by ELISA. Data are shown as mean ± standard error of the mean of the percentage of the mean maximal bTSH only response (% bTSH max); n = 8. * P ≤ 0.05.

To show that the effects of PTX were caused, at least in part, by uncoupling G_i(1–3)/G_o proteins from activation by TSHR, these proteins were knocked down in HEK-TSHR cells. The percent knock-down for G_i(1–3)/G_o mRNA compared with their nontargeting siRNA controls were as follows: G_i1, 38.76%; G_i2, 82.4%; G_i3, 60.15%; and G_o, 47.76% with P < 0.05 for all listed mRNAs. Figure 6 illustrates that G_i(1–3)/G_o knockdown had no effect on the basal levels of pERK1/2. Importantly, G_i(1–3)/G_o knockdown partially inhibited stimulation by IGF-1 alone, stimulation by a low dose of TSH, and low dose TSH plus IGF-1 stimulation of pERK1/2; however, G_i(1–3)/G_o knockdown did not inhibit stimulation of ERK1/2 phosphorylation caused by a maximally effective concentration of TSH without or with concomitant addition of IGF-1. These findings show that the pERK1/2 signal transduction cascade initiated by TSHR/IGF-1R crosstalk is in part dependent on G proteins sensitive to PTX.

Figure 6.

Knockdown of G_i(1–3) and G_o proteins lowers high-potency bTSH and bTSH + IGF-1 mediated ERK phosphorylation in HEK-TSHR cells. HEK-TSHR cells were transfected with 2 µM nonsilencing RNA (filled bars), 2 µM G_i(1–3) siRNA and G_o siRNA (unfilled bars). Following 72 hours posttransfection incubation at 37°C, cells were stimulated for 5 minutes at 37°C with HBSS, 0.0054 (bTSH_low), and 1.8 µM bTSH (bTSH_max; 1.8 µM = 100 mU/mL) with and without 100 ng/mL IGF-1. ERK phosphorylation was measured by ELISA. Data are shown as mean ± standard error of the mean as the percentage of maximal bTSH response (% bTSH max) alone response; n = 8. * P ≤ 0.05.

Discussion

Herein, we have shown that synergistic interactions between TSHR and IGF-1R occur rapidly at phosphorylation of ERK1 and ERK2, closely resembling pharmacologically defined GPCR/RTK cross-talk previously described in non-TSHR expressing cells (1–3). Before this work, it was an open question whether TSHR/IGF-1R signaling interactions occurred proximally or downstream of the receptors, though work done by Tsui et al. (6) demonstrating a close, physical association between TSHR and IGF-1R suggested proximal interactions were possible. We had previously shown synergism primarily at end-biologic responses, namely stimulated secretion of HA by GOFs and upregulation of thyroid-specific genes in human thyrocytes, that were routinely measured days after stimulation by TSH and IGF-1 was initiated. The changes in signaling that directly cause synergistic increases in HA secretion and gene expression are not known. In GOFs, it may involve increased transcription of HA synthases, decreased transcription of hyaluronidases that degrade HA, changes in the rates of translation of these enzymes, or posttranslational modifications that could alter their enzyme activities. In human thyrocytes, it may involve increased thyroid-specific gene transcription or decreased mRNA degradation. Until these mechanisms are elucidated it will not be possible to understand TSHR/IGF-1R interactions fully. Nevertheless, our findings that similar increases in TSH potency and efficacy occur for ERK1 and ERK2 phosphorylation in HEK-TSHR cells as for HA secretion by GOFs may allow us to determine the changes in the early postreceptor steps that cause synergy as ERK phosphorylation is a more proximal step in the signal transduction cascade. Moreover, the IGF-1 effects on potency and efficacy of TSH-stimulated increases in pERK1/2 are consistent with the idea that IGF-1 effected changes in the TSHR-initiated signaling cascade, that is, by IGF-1R cross-talk with TSHR.

There are several mechanisms by which activation of GPCRs leads to ERK phosphorylation (1, 2, 19, 20). In general, these are different for GPCRs that couple to different G proteins and to β-arrestins. Because TSHR can couple to all four G protein families (17) and to β-arrestin-1 for activatory signals (21), the mechanism of TSHR stimulation of ERK phosphorylation is not obvious. Moreover, because IGF-1 causes a marked increase in potency to TSH stimulation of ERK phosphorylation while retaining the lower potency stimulation as well, it is likely that more than one pathway is involved in this effect. We suggest that the lower potency stimulation found in the absence of IGF-1 is mediated via coupling of TSHR to the G_q/11–phospholipase C–diacylglycerol–protein kinase C pathway, as we have shown that in human preosteoblast-like U2OS cells this pathway stimulated by TSH alone leads to ERK activation (22). The pathway that mediates the high potency component of TSH-stimulated ERK phosphorylation in the presence of IGF-1 remains unknown; however, it appears to be initiated by coupling to G_i/o proteins that in turn may act via the phosphatidylinositol-4,5- bisphosphate 3-kinase pathway. Experiments to further delineate this pathway in detail is the goal of future studies.

It is interesting that PTX pretreatment partially inhibited stimulation of ERK1/2 phosphorylation by IGF-1 alone, which is most readily appreciated in HEK-TSHR cells in which PTX had no effect on basal pERK1/2 levels, and by a low and maximally effective concentration of TSH alone and with IGF-1, whereas knockdown of G_i(1-3)/G_o inhibited stimulation by a low dose of TSH (without or with IGF-1) but not by a maximally effective dose of TSH (Fig. 5). These findings are consistent with the shift in TSH potency caused by IGF-1 (Fig. 4) in that the main effect of IGF-1 is likely to be to increase coupling between TSHR and G_i(1-3)/G_o. As PTX inhibited ERK1/2 phosphorylation by all concentrations of TSH, it is likely that PTX has an effect in addition to decreasing IGF-1R-dependent TSHR coupling to G_i(1–3)/G_o. Future experiments will be directed at the identification of the additional effect of PTX on TSHR/G protein coupling.

The involvement of both TSHR and IGF-1R in the pathogenesis of GO has been described by several laboratory groups (6, 9, 11, 23–25); however, the mechanism of this involvement has been controversial. One idea states that stimulatory antibodies to both receptors are present in the blood of GO patients and that TSHR- and IGF-1R-stimulating antibodies activate both receptors concomitantly. A second idea, which we favor, is that TSHR-stimulating antibodies but not IGF-1R-stimulating antibodies are present in the blood of Graves’ patients and the involvement of IGF-1R is via cross-talk with TSHR, which is activated by binding of TSHR-stimulating antibodies to TSHRs only. In addition to the data presented herein and in our other reports involving thyroid-specific gene expression (7) and HA secretion (8, 9, 11), we recently showed that a monoclonal human TSHR-stimulating antibody, M22, stimulates TSHR/IGF-1R cross-talk (9) and binds to TSHR but not to IGF-1R (26). In fact, we have found no evidence of IGF-1R-stimulating antibodies in highly enriched immunoglobulin fractions from GO patients (9).

In conclusion, we have found that the synergy between TSHR and IGF-1R is mediated, at least in part, by rapid activation of the ERK1/2 pathway that is dependent on G protein coupling. These findings indicate that TSHR/IGF-1R interactions are likely similar to other GPCR/RTK interactions that have been described as receptor cross-talk.

Appendix.

Antibody Table

Peptide/Protein Target	Antigen Sequence (if Known)	Name of Antibody	Manufacturer, Catalog No.	Species Raised in; Monoclonal or Polyclonal	Dilution Used	RRID
Phospho-p44/42 MAPK (Erk1/2) (Thr202/Tyr204, Thr185/Tyr187)	Synthetic phosphopeptide corresponding to residues surrounding Thr202/Tyr204 of human p44 MAP kinase	Phospho-p44/42 MAPK (Erk1/2; Thr202/Tyr204l; E10)	Cell Signaling Technology, 9106	Mouse; monoclonal	1:2000	AB_331768
Phospho-Akt (Ser473)	Synthetic phosphopeptide corresponding to residues around Ser473 of mouse Akt	Phospho-Akt (Ser473; 587F11)	Cell Signaling Technology, 4051	Mouse; monoclonal	1:1000	AB_331158
GAPDH	Recombinant fragment within mouse GAPDH aa 100 to the C-terminus. The exact sequence is proprietary.	Anti-GAPDH antibody (EPR16891)	Abcam, ab181602	Rabbit; monoclonal	1:10,000	AB_2630358

Abbreviation: RRID, Research Resource Identifier.