Abstract
The aim of the present study was to examine whether triiodo-l-thyronine (T3) or l-thyroxine (T4) rapidly activated the mitogen-activated protein kinase (MAPK) intracellular signalling cascade in osteoblast-like cells and investigate whether this activation was initiated at the integrin αVβ3 cell surface receptor. Using PCR and western blotting, the expression of integrin αVβ3 mRNA and protein was demonstrated in the human osteoblast-like cell lines MG-63 and SaOS-2. The treatment of MG-63 cells with T3 (10 nM) or T4 (100 nM) for 10 min stimulated extracellular signal-regulated kinase activity (ERK, a component of the MAPK pathway) as determined by fluorescent immunocytochemistry and an immunocomplex activity assay (T3 by 10.7-fold, P<0.01 and T4 by 10.4-fold, P<0.01 compared with control). T3 (10 nM) and T4 (100 nM) also significantly stimulated thymidine incorporation into MG-63 cells by 2.3±0.7-fold (P<0.01) and 2.1±0.1-fold (P<0.05) respectively. To establish whether transient ERK activation via the integrin αVβ3 cell surface receptor mediated these effects, MG-63 cells were pretreated for 30 min with the specific MAPK kinase inhibitor, U0126 (1 μM), or an anti-integrin αVβ3-blocking antibody. Both pretreatments significantly inhibited T3- and T4-stimulated ERK activation and abolished T3-stimulated thymidine incorporation (P<0.01). T4-stimulated incorporation was significantly inhibited from 2.1- to 1.3-fold above control (P<0.05). Thus, our results suggest that T3 and T4 rapidly stimulate ERK activation in MG-63 cells via integrin αVβ3 and that one functional effect of this ERK activation is increased DNA synthesis.
Introduction
Non-genomic, receptor-independent actions of members of the steroid hormone family are well recognised. For example, triiodo-l-thyronine (T3) and l-thyroxine (T4) are reported to rapidly activate intracellular signalling pathways, such as mitogen-activated protein kinase (MAPK; Lin et al. 1999, Bergh et al. 2005, Mousa et al. 2006), independent of the thyroid hormone receptor (TR). A number of molecular mechanisms underlying the non-genomic activation of signalling cascades by thyroid hormone have been reported (reviewed in Davis et al. 2005) with one proposed mechanism involving the existence of plasma membrane-binding sites for thyroid hormone. Evidence that the heterodimeric structural proteins, integrins, might contain a cell surface receptor site for thyroid hormones was initially obtained from a study showing T4-dependent regulation of integrin–laminin interactions (Farwell et al. 1995). This study also reported that a RGD (Arg-Gly-Asp) peptide interfered with these interactions. Later studies showed that purified integrin αVβ3 protein bound radiolabelled T4 and that this binding was again blocked by the RGD peptide and integrin antibodies (Bergh et al. 2005), thus suggesting that the receptor site was located at or near the RGD recognition site on integrin αVβ3. The functional consequences of thyroid hormone binding to a cell surface site were subsequently illustrated by the same authors with knock-down experiments using small interfering RNAs directed against the genes encoding the αV or β3 subunits, which prevented thyroid hormone activation of MAPK in CV-1 cells. An alternative approach used a specific integrin αVβ3 antagonist to inhibit the activation of MAPK and proangiogenic actions of T4 in human microvascular endothelial cells (Mousa et al. 2006). Thus, increasing evidence would suggest that the extracellular domain of integrin αVβ3 can act as a thyroid hormone cell surface receptor and initiate rapid activation of MAPK signal transduction cascades in human cell lines.
Several in vitro studies have demonstrated direct effects of thyroid hormones on cells of the osteoblast lineage. In response to T3, osteoblasts are able to increase DNA synthesis and alkaline phosphatase activity (Sato et al. 1987), osteocalcin production (Ohishi et al. 1994) and cytokine synthesis (Siddiqi et al. 1998). While the expression of TR isoforms has been documented in both rodent (Williams et al. 1994) and human (Siddiqi et al. 2002) osteoblast cell lines and primary cultures of rat (Bland et al. 1997) and human (Siddiqi et al. 2002) osteoblasts, there is only one report from more than a decade ago, which has provided evidence for non-genomic effects of T3 on bone (Lakatos & Stern 1991). More recently, Hoffman et al. (2002) reported that the RGD peptide blocked bone demineralisation in rats induced by thyroid hormone raising the possibility that the integrin αVβ3 protein might be the initiation site for non-genomic actions of thyroid hormone on osteoblasts. The aim of our study therefore was to examine whether T3 or T4 were able to rapidly activate the MAPK intracellular signalling cascade and investigate whether this activation was initiated at the integrin αVβ3 cell surface receptor.
Materials and Methods
Materials
Epidermal growth factor (EGF) was purchased from CN Biosciences (Nottingham, UK), T3 and T4 were purchased from Sigma–Aldrich Company Ltd. The specific MAPK kinase (MEK) inhibitor, U0126, was obtained from Promega and used at 1 μM, a concentration previously reported to produce almost 100% suppression of MAPK kinase 1 (also known as MKK1 or MEK1 (Davies et al. 2000)). All the above components were prepared as stock solutions in sterile water and applied to the cells in culture media. Specific goat polyclonal antibodies to the αV (SC-6618) and β3 subunits (SC-6627) with their respective blocking peptides were purchased from Santa Cruz Biotechnology Inc. (Santa Cruz, CA, USA) and a monoclonal blocking antibody to αVβ3 (ab20143) was obtained from Abcam (Cambridge, Cambs, UK). Horseradish peroxidase (HRP)-conjugated donkey anti-goat Ig and HRP-conjugated rabbit anti-mouse Ig were purchased from Dako (UK) Ltd (Ely, Cambs, UK).
Cell culture
The human osteoblasts (hOb)-like cell lines, MG-63 and SaOS-2 (both kindly donated by F J Hughes, School of Dentistry, Queen Mary, University of London, UK), and human dermal fibroblasts (kindly donated by L E Russell, Centre of Dermatology, Queen Mary, University of London, UK) were maintained in αMEM supplemented with 10% (v/v) heat-inactivated fetal calf serum (FCS), 1000 U/l penicillin and 1 mg/l streptomycin (hereafter referred to as medium). All culture reagents were purchased from Invitrogen Life Technologies Inc. The cultures were maintained at 37 °C in a humidified atmosphere containing 5% CO2 in air. The medium was changed every 3 days, and the cell lines were passaged at 70% confluence.
Reverse transcription and real-time PCR
Total RNA was extracted from cells using the RNeasy kit from Qiagen according to the manufacturer's instructions. Three hundred nanograms of total RNA were transcribed using the First Strand DNA synthesis kit (Amersham Biosciences) according to the manufacturer's instructions. Reverse transcription was also performed on diethylpyrocarbonate (DEPC)-treated water for use as a negative control (NC) in subsequent PCRs. The presence of integrin αV and β3 subunits and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) cDNA in the samples was confirmed by PCR. The primer sequences used in the PCR were based on published human-specific sequences: αV (NM_002210): forward 5′-CACCAGCAGTCAGAGATGGA, reverse 5′-GGCAACCGTGTCATTCTTTT (441 bp product); β3 (NM_000212): forward 5′-AAGGATAACTGTGCCCCAGA, reverse 5′-CACAGGCTTGTCCACAAATG (348 bp product); GAPDH (AF261085): forward 5′-TGCACCACCAACTGCTTAG, reverse 5′-CACCACAATGTTGTCGTAG (523 bp product). After a 3-min incubation at 95 °C, the PCR was run for 35 cycles under the following conditions: 95 °C for 30 s, 60 °C for 30 s, 72 °C for 60 s, followed by a final extension step at 72 °C for 10 min. The PCR products were visualised on a 1.4% (wt/vol) agarose gel stained with ethidium bromide. The identity of each PCR product generated was confirmed by direct DNA sequencing.
Total protein extraction and western blotting
Total proteins were harvested from dermal fibroblasts, MG-63 and SaOS-2 cells, subjected to SDS-PAGE (on a 10% (w/v) resolving gel) and transferred to nitrocellulose membranes as described previously (Fowkes et al. 2003). After blocking for 1 h with 5% (w/v) non-fat milk in Tris-buffered saline containing 0.1% (v/v) Tween 20 (TBS-T), the membranes were incubated overnight at 4 °C with the indicated primary antibodies (in the presence or absence of blocking peptides), washed and incubated for 1 h at room temperature (RT) with the appropriate secondary antibodies. Immunoreactive protein was detected using enhanced chemiluminescence (ECL Western Blotting Reagents, Amersham Bioscience).
Immunocytochemistry
Specific rabbit polyclonal anti-phospho-extracellular signal-regulated kinase (pERK) antibody (V803A) was obtained from Promega. MG-63 cells were plated at a density of 2×105 cells onto sterile coverslips in 35-mm culture dishes and allowed to settle overnight in medium, followed by incubation for 24 h in serum-free αMEM. Subsequently, the cells were pretreated for 30 min in the presence or absence of 1 μM U0126. The cells were then treated with either no stimulant (NC), T3 (10 nm) or EGF (10 nM), as a positive control, for 10 min before being fixed in 4% (v/v) paraformaldehyde for 30 min. Ice-cold methanol was then applied to the fixed cells for 10 min to permeabilise them. The coverslips were then incubated in blocking buffer (1% (w/v) BSA and 5% (v/v) donkey serum in PBS) for 2 h, washed in PBS and incubated overnight in anti-pERK antibody (Promega) diluted 1:500. Control cells were processed in the absence of primary antibody. After three 15-min washes in PBS, the cells were incubated for 1 h in donkey anti-rabbit Cy3 secondary antibody (Vector Laboratories Ltd, Peterborough, Lincs, UK) before mounting the coverslips onto microscope slides with Fluoromount-G (Cambridge Bioscience Ltd, Cambridge, UK) and viewing with a fluorescent microscope.
Total protein extractions and measurement of ERK activity
MG-63 cells were plated in 35-mm dishes (2 million cells per dish) and allowed to grow overnight in culture medium containing 10% (v/v) FCS, which was then changed to serum-free medium for 24 h. Subsequently, the cells were pretreated for 30 min in the presence or absence of 1 μM U0126 or a monoclonal blocking antibody to αVβ3 (diluted 1:500). Following this pretreatment period, the cells were treated for 10 min with serum-free medium alone, 10 nM T3 or 100 nM T4, in the continued presence or absence of 1 μM U0126 or the monoclonal blocking antibody to αVβ3. Total proteins were then extracted in the presence of phosphatase inhibitors as described previously and ERK activity was measured using the p44/42 MAP Kinase Assay kit (New England Biolabs, Hitchin, UK) as described previously (Fowkes et al. 2001). Briefly, active MAPK was immunoprecipitated from cell lysates with an immobilised phospho-p44/42 MAP kinase (ERK; Thr202/Tyr204) monoclonal antibody. The immunoprecipitated pellets were incubated with 200 μM ATP and 2 μg Elk-1 fusion protein, before terminating the reaction with SDS loading buffer. The levels of phosphorylated Elk-1 were analysed by immunoblotting with anti-phospho-Elk-1 antibody. Immunoreactive protein was visualised by enhanced chemiluminescence, using the supplied LumiGLO and peroxidase reagents. Autoradiographs were analysed by scanning densitometry using Gel Base/Gel Blot Pro software (Synoptics Ltd, Cambridge, UK). Peak height intensities were used to calculate ERK phosphorylation of Elk-1, and compared with the untreated controls.
[3H]thymidine incorporation
MG-63 cells were plated in 24-well cell culture dishes (10 000 cells/dish) and allowed to grow overnight in media containing 10% FCS, which was then changed to serum-free medium for 24 h. Subsequently, cells were treated with serum-free medium alone, T3 (1, 10, 100 nM), T4 (1, 10, 100 nM) or medium containing 10% (v/v) FCS as a positive control for 18 h before the addition of 1 μCi/well of [3H]thymidine (Amersham Bioscience) for a further 6 h. Cells were trypsinised and harvested, before counting in the presence of scintillation fluid using a MicroBeta 1450 β-counter (Wallac, Beaconsfield, Bucks, UK). In later experiments, the cells were pretreated for 30 min in the presence or absence of 1 μM U0126 or a monoclonal blocking antibody to αVβ3 (diluted 1:500) prior to treatment with serum-free medium alone, T3 (10 nM), T4 (100 nM) or medium containing 10% FCS as a positive control in the continued presence or absence of U0126 or the blocking antibody to αVβ3 for 18 h before the addition of 1 μCi/well of [3H]thymidine (Amersham Bioscience) for a further 6 h.
Data presentation and statistical analysis
All graphical data were prepared using GraphPad Prism 4.03 (GraphPad, San Diego, CA, USA) and analysed using pre-programmed analysis equations within Prism. Data are presented as results which are representative of several experiments or as normalised data pooled from multiple experiments. Where appropriate, an ANOVA was performed on data followed by Tukey's multiple comparisons test, accepting P<0.05 as significant.
Results
Expression of integrin αV and β3 subunits in MG-63 and SaOS-2 cells
To investigate the presence of αV and β3 subunit mRNAs in these human osteoblast-like cell lines, RT-PCR was performed on MG-63 and SaOS-2 cDNA using specific intron-spanning primers designed to amplify the different subunits. After 30 cycles, DNA fragments of the expected size were detected for integrin αV (441 bp) and integrin β3 (348 bp) in both cell lines (Fig. 1A). As reported previously (Asano et al. 2005), both subunits were also detected in the positive control, dermal fibroblast cells (Fig. 1A). When NCs were performed by omitting reverse transcriptase from the RT reaction no PCR products were detected (data not shown). RT-PCR of GAPDH (expected product size, 523 bp) was also performed to confirm loading of cDNA from each cell type (Fig. 1B). Sequencing of the DNA fragments confirmed product identity in all cases. Western blotting with integrin receptor subunit-specific antibodies (Fig. 2) confirmed the presence of proteins of the expected size for integrin αV (125–135 kDa) and integrin β3 (125 kDa) in both cell lines and in the positive control, dermal fibroblasts. Western blotting of dermal fibroblast cell lysates in the presence of specific integrin αV or integrin β3 blocking peptides confirmed the identity of the protein.
The presence of integrin αV and β3 mRNA in human osteoblast-like cell lines. (A) Amplification of cDNA for integrin αV and integrin β3 subunits by PCR in dermal fibroblasts (DF), SaOS-2 cells (SA) and MG-63 cells (MG). (B) Amplification of cDNA for the reference housekeeping gene GAPDH in dermal fibroblasts (DF), SaOS-2 cells (SA) and MG-63 cells (MG). M is the DNA size marker.
Citation: Journal of Endocrinology 196, 3; 10.1677/JOE-07-0344
Expression of integrin αV and β3 proteins in human osteoblast-like cell lines. Whole-cell lysates from dermal fibroblasts (DF), MG-63 cells (MG) and SaOs-2 cells (SA) were electrophoresed through a 10% polyacrylamide gel, transferred to a nitrocellulose membrane and probed with anti-αV (A) or anti-β3 (B) antibodies. The blots shown are representative of three independent experiments. Western blots of dermal fibroblast cell lysates were also performed in the presence of blocking peptides to integrin αV or β3 (DF+BP).
Citation: Journal of Endocrinology 196, 3; 10.1677/JOE-07-0344
Thyroid hormones stimulate pERK in MG-63 cells
In preliminary studies, we used fluorescent immunocytochemistry with a specific antibody to pERK to investigate the effects of T3 on MG-63 cells. MG-63 cells were treated for 10 min with either serum-free media alone, T3 (10 nM) or EGF (10 nM) as a positive control. To demonstrate the specificity of these effects, MG-63 cells were also pretreated for 30 min with U0126 (1 μM), a MEK inhibitor, and then for 10 min with either serum-free media alone, T3 (10 nM) or EGF (10 nM) in the continued presence of U0126. Following treatment, the cells were permeabilised, fixed and incubated overnight with rabbit anti-pERK antibody. The cells were then incubated with rhodamine-labelled donkey anti-rabbit Cy3-conjugated antibody. They were subsequently examined using a fluorescent microscope at 100× magnification. As shown in Fig. 3, both T3- and EGF-stimulated the phosphorylation of ERK in MG-63 cells following 10 min incubation (Fig. 3B and C respectively). The specific MEK inhibitor, U0126, was able to inhibit these stimulatory effects (Fig. 3E and F). There was no staining visible on control slides processed in the absence of primary antibody (Fig. 3, NC).
Fluorescent immunocytochemistry to demonstrate the effect of T3 on pERK in MG-63 cells. MG-63 cells were treated for 10 min with either (A) serum-free media alone (SFM), (B) T3 (10 nM; T3) or (C) EGF (10 nM; EGF) as a positive control. MG-63 cells were pretreated for 30 min with U0126 (1 μM), a MEK inhibitor and then for 10 min with either (D) serum-free media alone (SFM+U0126), (E) T3 (10 nM; T3+U0126) or (F) EGF (10 nM; EGF+U0126). Following treatment, the cells were stained for pERK and subsequently examined using a fluorescent microscope at 100× magnification. Negative control (G) was processed in the absence of primary antibody.
Citation: Journal of Endocrinology 196, 3; 10.1677/JOE-07-0344
Subsequent quantitative studies were performed with an immunoprecipitation kinase activity assay. MG-63 cells were pretreated for 30 min with serum-free media alone in the presence or absence of U0126 (1 μM) or anti-integrin αVβ3 antibody (1:500 dilution). Subsequently, the cells were treated for 10 min with serum-free media alone, T3 (10 nM) or T4 (100 nM) in the continued presence or absence of U0126 (1 μM) or anti-integrin αVβ3 antibody (1:500 dilution). The presence of activated ERK was quantitated by its ability to phosphorylate Elk-1, which was detected by western blotting with phospho-Elk-1 antibody and scanning densitometry. The activation of ERK at 10 min was enhanced by T3 (10.7-fold, P<0.01) and T4 (10.4-fold, P<0.01) compared with the control (Fig. 4). T3-stimulated ERK activation was completely blocked by pretreatment with U0126 (1.1-fold above control, P<0.01 compared with T3 stimulation alone) and partially inhibited by pretreatment with anti-integrin αVβ3 antibody (5.2-fold above control, P<0.05 compared with T3 stimulation alone; Fig. 4). T4-stimulated ERK activation was also inhibited following pretreatment with U0126 (3.1-fold above control, P<0.01 compared with T4 stimulation alone) and showed a partial reduction to 7.3-fold above control with anti-integrin αVβ3 antibody, although this decrease did not achieve statistical significance.
Elk-1 phosphorylation in MG-63 cells by T3 and T4. MG-63 cells were pretreated for 30 min with serum-free media alone or U0126 (1 μM) or anti-integrin αVβ3 antibody (1:500) and then treated for 10 min with serum-free media, T3 (10 nM) or T4 (100 nM) in the continued presence or absence of U0126 (U) or anti-integrin αVβ3 antibody (Ab). ERK activation was determined by an immunoprecipitation kinase assay. An autoradiograph of pElk-1 is shown (upper panel) with accompanying scanning densitometry data from three independent experiments (lower panel). *P<0.05, **P<0.01 compared with T3- or T4-treated cells. A positive control for the kinase assay (MAPK) is also shown where active MAPK (20 ng) was added to untreated cell extract.
Citation: Journal of Endocrinology 196, 3; 10.1677/JOE-07-0344
Thyroid hormones stimulate thymidine incorporation in MG-63 cells
Previous studies have reported that T3 can stimulate DNA synthesis in MG-63 cells (5) but we could find no published reports for effects of T4 on DNA incorporation in osteoblast-like cells. Thus, our initial studies investigated dose–response effects of T3 and T4 on MG-63 cells over a 24-h time period. Both 10 and 100 nM T3 significantly stimulated thymidine incorporation (by 1.8- and 1.9-fold respectively, P<0.001; Fig. 5A). The positive control (10% serum) increased thymidine incorporation by 2.6-fold (P<0.001). T4 significantly increased thymidine incorporation at all three concentrations tested (by 1.3-fold at 1 nM, P<0.01; by 1.6-fold at 10 nM, P<0.001 and by 1.5-fold at 100 nM, P<0.01; Fig. 5B). Subsequent experiments using T3 at 10 nM and T4 at 100 nM were performed in the presence or absence of U0126 (1 μM) or anti-integrin αVβ3 antibody (1:500 dilution; Fig. 6). Again, T3 and T4 significantly stimulated DNA synthesis by 2.3-fold (P<0.01, compared with control) and 2.1-fold (P<0.05, compared with control) respectively. Neither U0126 nor anti-integrin αVβ3 antibody affected basal thymidine incorporation. Treatment of MG-63 cells with U0126 blocked the effects of T3 entirely (to 0.95-fold compared with control, P<0.01 compared with T3 stimulation alone) and significantly attenuated the effects of T4 (to 1.3-fold above control, P<0.05 compared with T4 stimulation alone), suggesting that ERK activation is involved in mediating the effects of both T3 and T4 on DNA synthesis in MG-63 cells. The effects of the positive control were also significantly reduced by U0126, suggesting that serum-stimulated thymidine incorporation in MG-63 cells is partially mediated by ERK activation. Treatment of MG-63 cells with anti-integrin αVβ3 antibody also blocked the effects of T3 (to 0.96-fold compared with control, P<0.01 compared with T3 stimulation alone) and significantly reduced the effects of T4 (to 1.5-fold above control, P<0.05 compared with T4 stimulation alone). As expected, the serum-positive control was unaffected by anti-integrin αVβ3 antibody. These results suggest that blocking the integrin αVβ3 cell surface receptor inhibits the effects of T3 and T4 on thymidine incorporation in MG-63 cells, thus supporting a role for integrin αVβ3 as the initiation site for T3- and T4-induced ERK activation and cell proliferation in human osteoblast-like cells.
Dose–response effects of T3 and T4 on [3H]thymidine incorporation into MG-63 cells. Cells were treated with either serum-free media alone (0), T3 (1, 10 or 100 nM) or T4 (1, 10 or 100 nM) for 24 h. A medium containing 10% FCS was used as the positive (+ve) control. [3H]thymidine (1 μCi/well) was added for the last 6 h of culture. Results are expressed as c.p.m. and show means±s.e.m. of at least 12 individual incubations (**P<0.01, ***P<0.001).
Citation: Journal of Endocrinology 196, 3; 10.1677/JOE-07-0344
The involvement of integrin αVβ3 and ERK in T3- and T4-stimulated [3H]thymidine incorporation into MG-63 cells. These cells were pretreated for 30 min with serum-free media alone or U0126 (1 μM) or anti-integrin αVβ3 antibody (1:500 dilution), followed by treatment for 24 h with serum-free media, T3 (10 nM) or T4 (100 nM) in the continued presence or absence of U0126 or anti-integrin αVβ3 antibody. Media containing 10% FCS (serum) was used as a positive control. Results are expressed as fold increase over cells treated with serum-free media alone and show means±s.e.m. of three independent experiments, each performed in quadruplicate (*P<0.05, **P<0.01 compared with T3, T4 or serum alone).
Citation: Journal of Endocrinology 196, 3; 10.1677/JOE-07-0344
Discussion
The results reported here provide novel evidence that T3 and T4 stimulate ERK activation in human osteoblast-like cells as revealed by fluorescent immunocytochemistry for pERK or by phosphorylation of the ERK-regulated transcription factor, Elk-1. These findings are consistent with previous studies in HeLa cells, CV-1 monkey fibroblasts and chick embryo hepatocytes that have demonstrated the ability of thyroid hormones to activate the MAPK signalling cascade and promote the phosphorylation of ERK within 5–20 min after treatment (Lin et al. 1999, Alisi et al. 2004, Bergh et al. 2005). We used concentrations of T3 and T4 that were reported previously to produce the maximal ERK activation response in other cell types. The concentration of T3 needed (10 nM) to achieve ERK activation is in the supraphysiological circulating range (lower concentrations failed to achieve a significant effect (data not shown)), while that of T4 (100 nM) is within the physiological circulating range. The MAPK signalling pathway has been demonstrated previously in normal human osteoblasts and bone marrow stromal cells (Chaudhary & Avioli 1998) but activation of this pathway in bone cells is generally reported to involve growth factor receptors such as platelet-derived growth factor or fibroblast growth factor (FGF). A 6-h pretreatment with T3 was shown to enhance FGF2-stimulated MAPK (Stevens et al. 2003) although these authors failed to show a stimulation of ERK phosphorylation with T3 alone. However, these studies were performed in rat osteoblast cell lines and used only T3 (and not T4) at a very high concentration of 100 nM for 30 min. Thus, there may be species differences in response, or as reported previously (Lin et al. 1999, Alisi et al. 2004, Bergh et al. 2005), thyroid hormone effects on ERK activation are rapid and transient with activation returning to control levels by 30 min.
The early reports of rapid thyroid hormone activation of ERK in HeLa and CV-1 cells (which lack functional TRs) suggested that this effect of T3 and T4 was cell surface initiated and not mediated via TRs. The existence of binding sites for thyroid hormone on the cell surface has been known for many years in the red blood cell membrane (Yoshida & Davis 1981) and in the synaptosome (Giguere et al. 1996) and indeed, the existence of plasma membrane-associated receptors for T3 in bone was proposed in 1991 (Lakatos & Stern 1991). However, it was not until recently that the identity of one cell surface protein capable of initiating non-genomic actions of thyroid hormone was revealed as integrin αVβ3 (Bergh et al. 2005, Davis et al. 2006, Mousa et al. 2006). Evidence supporting this concept was provided by the stimulatory effects of thyroid hormone on angiogenesis and MAPK activation being inhibited either in the presence of small interfering RNAs directed against the genes encoding the αV or β3 subunits or with the use of a specific integrin αVβ3 antagonist or the RGD peptide (Bergh et al. 2005, Davis et al. 2006, Mousa et al. 2006). The presence of integrin αVβ3 in human osteoblasts has not been reported, thus as an initial experiment we examined the expression levels of αV and β3 subunits in both MG-63 and SaOs-2 cells. We found that αV and β3 subunit mRNA and protein were expressed in both the cell types. To function as active receptors, integrins have to be present on the cell surface as dimers, which dissociate when cell lysates are examined by western blotting. Therefore, to examine whether heterodimeric integrin αVβ3 was mediating the rapid thyroid hormone activation of ERK in osteoblasts, we used a blocking antibody to the dimer αVβ3. Integrin αVβ3 is also known to function as an active receptor for latent transforming growth factor-β (TGF-β) and this anti-αVβ3 antibody has been used previously to reduce the transcriptional effects of TGF-β in scleroderma fibroblasts (Asano et al. 2005). The attenuating effects of the anti-αVβ3 antibody were greater on T3 than T4-stimulated ERK activation. This was also the case when cells were treated with the specific MEK inhibitor, U0126, and might suggest that molecular mechanisms other than those initiated at the integrin αVβ3 receptor exist for T4 action at the cell surface as proposed by Davis et al. (2005). In other cell types, activation of the integrin αVβ3 receptor is reported to activate the MAPK pathway via protein kinase C (PKC; Alisi et al. 2004) and we have preliminary data using a specific PKC inhibitor (data not shown) that would suggest that this is also the case in osteoblast cells.
Both T3 and T4 are reported to influence proliferation in a number of different cell types (Garcia-Silva et al. 2002, Alisi et al. 2004, Davis et al. 2006). T3 has been found to stimulate, inhibit or exert no effect on osteoblastic cell proliferation but a consensus suggests that T3 stimulates osteoblast activity (Harvey et al. 2002). There are no reports in the literature of T4 effects on osteoblast proliferation. Our results demonstrate that both T3 and T4 are able to significantly increase DNA synthesis in MG-63 cells although the concentration of T3 (10 nM) needed to achieve this effect was again supraphysiological. Similarly, in glioma cells, T4 (1–100 nM) was reported to increase thymidine incorporation while the concentrations of T3 needed to achieve this effect were again supraphysiological (Davis et al. 2006). These authors had previously shown that the affinity of αVβ3 for T3 is substantially lower than the affinity for T4 (Bergh et al. 2005), which may explain these concentration-dependent differences between T3 and T4. In glioma cells, the stimulatory effects of T4 on DNA synthesis were blocked by the RGD peptide (Davis et al. 2006). In our study, the stimulatory effects of T3 were completely abolished in the presence of either anti-αVβ3 blocking antibody or U0126, suggesting that T3 effects are mediated entirely via the integrin αVβ3 receptor and subsequent ERK activation. T4-stimulated thymidine incorporation was only partially inhibited by either the anti-αVβ3 blocking antibody or U0126, despite this inhibitor being used at a concentration reported to completely suppress the activation of MKK1 (Davies et al. 2000). These results would again suggest that additional mechanisms other than those initiated at the integrin αVβ3 receptor might exist for T4 effects on thymidine incorporation.
In summary, our results suggest that T3 and T4 rapidly stimulate ERK activation in MG-63 cells and that one of the functional effects of this ERK activation is increased DNA synthesis and cell proliferation. In addition, this study has demonstrated the presence of integrin αVβ3 in human osteoblast-like cells and provides in vitro evidence to support a role for integrin αVβ3 as the apparent initiation site for T3- and T4-induced ERK activation and cell proliferation in human osteoblast-like cells.
Acknowledgements
The authors thank Prof. F J Hughes and Ms L E Russell (Queen Mary, University of London) for provision of the hOb-like cell lines and human dermal fibroblasts respectively. A S was supported by a grant from the St Bartholomew's and Royal London Alumnae Association. P H and K S were supported by the Research Advisory Board of the St Bartholomew's and the Royal London Charitable Foundation. The authors declare that there is no conflict of interest that would prejudice the impartiality of the present work.
References
Alisi A, Spagnuolo S, Napoletano S, Spaziani A & Leoni S 2004 Thyroid hormones regulate DNA synthesis and cell cycle proteins by activation of PKCα and p42/44 MAPK in chick embryo hepatocytes. Journal of Cellular Physiology 201 259–265.
Asano Y, Ihn H, Yamane K, Jinnin M, Mimura Y & Tamaki K 2005 Increased expression of integrin αVβ3 contributes to the establishment of autocrine TGF-β signaling in scleroderma fibroblasts. Journal of Immunology 175 7708–7718.
Bergh JJ, Lin H-Y, Lansing L, Mohamed SN, Davis FB, Mousa S & Davis PJ 2005 Integrin αVβ3 contains a cell surface receptor site for thyroid hormone that is linked to activation of mitogen-activated protein kinase and induction of angiogenesis. Endocrinology 146 2864–2871.
Bland R, Sammons RL, Sheppard MC & Williams GR 1997 Thyroid hormone, vitamin D and retinoid receptor expression and signaling in primary cultures of rat osteoblastic and immortalized osteosarcoma cells. Journal of Endocrinology 154 63–67.
Chaudhary LR & Avioli LV 1998 Identification and activation of mitogen activated protein (MAP) kinase in normal human osteoblastic and bone marrow stromal cells: attenuation of MAPK activation by cAMP, parathyroid hormone and forskolin. Molecular and Cellular Biology 178 59–68.
Davies SP, Reddy H, Caivano M & Cohen P 2000 Specificity and mechanism of action of some commonly used protein kinase inhibitors. Biochemical Journal 351 95–105.
Davis PJ, Davis FB & Cody V 2005 Membrane receptors mediating thyroid hormone action. Trends in Endocrinology and Metabolism 16 429–435.
Davis FB, Tang HY, Shih A, Keating T, Lansing L, Hercbergs A, Fenstermaker RA, Mousa A, Mousa SA & Davis PJ et al. 2006 Acting via a cell surface receptor, thyroid hormone is a growth factor for glioma cells. Cancer Research 66 7270–7275.
Farwell AP, Tranter MP & Leonard JL 1995 Thyroxine-dependent regulation of integrin–laminin interactions in astrocytes. Endocrinology 136 3909–3915.
Fowkes RC, Burch J & Burrin JM 2001 Stimulation of extracellular signal-regulated kinase by PACAP in αT3-1 gonadotropes. Journal of Endocrinology 171 R5–R10.
Fowkes RC, Sidhu KK, Sosabowski JK, King P & Burrin JM 2003 Absence of PACAP-stimulated transcription of the human glycoprotein α-subunit (αGSU) gene in LβT2 gonadotrophs reveals disrupted cyclic 3′5′-adenosine monophosphate (cAMP)-mediated gene transcription. Journal of Molecular Endocrinology 31 263–278.
Garcia-Silva S, Perez-Juste G & Aranda A 2002 Cell cycle control by thyroid hormone in neuroblastoma cells. Toxicology 181 179–182.
Giguere A, Fortier S, Beaudry C, Gallo-Payet N & Bellabarba D 1996 Effect of thyroid hormones on G proteins in synaptosomes of chick embryo. Endocrinology 137 2558–2564.
Harvey CB, O'Shea PJ, Scott AJ, Robson H, Siebler T, Shalet SM, Samarut J, Chassande O & Williams GR 2002 Molecular mechanisms of thyroid hormone effects on bone growth and function. Molecular Genetics and Metabolism 75 17–30.
Hoffman SJ, Vasko-Moser J, Miller WH, Lark MW, Gowen M & Stroup G 2002 Rapid inhibition of thyroxine-induced bone resorption in the rat by an orally active vitronectin receptor antagonist. Journal of Pharmacology and Experimental Therapeutics 302 205–211.
Lakatos P & Stern PH 1991 Evidence for direct non-genomic effects of triiodothyronine on bone rudiments in rats: stimulation of the inositol phosphate second messenger system. Acta Endocrinologica 125 603–608.
Lin HY, Davis FB, Gordinier JK, Martino LJ & Davis PJ 1999 Thyroid hormone induces activation of mitogen-activated protein kinase in cultured cells. American Journal of Physiology 276 C1014–C1024.
Mousa SA, O'Connor L, Davis FB & Davis PJ 2006 Proangiogenesis action of the thyroid hormone analog 3′5-diiodothyropropionic acid (DITPA) is initiated at the cell surface and is integrin mediated. Endocrinology 147 1602–1607.
Ohishi K, Ishida H, Nagata T, Yamauchi N, Tsurumi C, Nishikawa S & Wakano Y 1994 Thyroid hormone suppresses the differentiation of osteoprogenitor cells to osteoblasts, but enhances functional activities of mature osteoblasts in cultured rat calvaria cells. Journal of Cellular Physiology 161 544–552.
Sato K, Han DC, Fujii Y, Tsushima T & Shizume K 1987 Thyroid hormone stimulates alkaline phosphatase activity in cultured rat osteoblastic cells (ROS 17/2.8) through 3,5,3-triiodo-l-thyronine nuclear receptors. Endocrinology 120 1873–1881.
Siddiqi A, Burrin JM, Wood DF & Monson JP 1998 T3 regulates the production of IL-6 and IL-8 in human bone marrow stromal and osteoblast-like cells. Journal of Endocrinology 157 453–461.
Siddiqi A, Parsons MP, Lewis JL, Monson JP, Williams GR & Burrin JM 2002 TR expression and function in human bone marrow stromal and osteoblast-like cells. Journal of Clinical Endocrinology and Metabolism 87 906–914.
Stevens DA, Harvey CB, Scott AJ, O'Shea PJ, Barnard JC, Williams AJ, Brady G, Samarut J, Chassande O & Williams GR 2003 Thyroid hormone activates fibroblast growth factor receptor-1 in bone. Molecular Endocrinology 17 1751–1766.
Williams GR, Bland R & Sheppard MC 1994 Characterization of thyroid hormone (T3) receptors in three osteosarcoma cell lines of distinct osteoblast phenotype: interactions among T3, Vitamin D3 and retinoid signaling. Endocrinology 135 2375–2385.
Yoshida K & Davis PJ 1981 Partition of thyroid hormone among erythrocyte cytosol, erythrocyte membrane and human plasma binding sites. Hormone and Metabolic Research 13 394–395.
