| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Molecular Endocrinology Group (D.A.S., C.B.H., A.J.S., P.J.O’S., J.C.B., A.J.W., G.R.W.), Division of Medicine and Medical Research Council Clinical Sciences Centre, Faculty of Medicine, Imperial College London, London W12 0NN, United Kingdom; Epistem Ltd. and School of Biological Sciences (G.B.), University of Manchester, Manchester M13 9PT, United Kingdom; Laboratoire de Biologie Moléculaire et Cellulaire de l’ENS de Lyon (J.S., O.C.), Unité Mixte de Recherche 5665 Centre National Recherche Scientifique, LA 913 Institut National de la Recherche Agronomique, 69364 Lyon Cedex 07, France; and Unité Institut National de la Santé et de la Recherche Médicale Unité-443 (O.C.), Université Victor Segalen Bordeaux 2, 33076 Bordeaux Cedex 02, France
Address all correspondence and requests for reprints to: Dr. G. R. Williams, Molecular Endocrinology Group, Medical Research Council Clinical Sciences Centre, Clinical Research Building, 5th Floor, Hammersmith Hospital, Du Cane Road, London W12 0NN, United Kingdom. E-mail: graham.williams{at}imperial.ac.uk.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
gene are essential for bone development whereas adult hyperthyroidism increases the risk of osteoporotic fracture. We isolated fibroblast growth factor receptor-1 (FGFR1) as a T3-target gene in osteoblasts by subtraction hybridization. FGFR1 mRNA was induced 2- to 3-fold in osteoblasts treated with T3 for 6–48 h, and FGFR1 protein was stimulated 2- to 4-fold. Induction of FGFR1 was independent of mRNA half-life and abolished by actinomycin D and cycloheximide, indicating the involvement of an intermediary protein. Fibroblast growth factor 2 (FGF2) stimulated MAPK in osteoblasts, and pretreatment with T3 for 6 h induced a more rapid response to FGF that was increased in magnitude by 2- to 3-fold. Similarly, T3 enhanced FGF2-activated autophosphorylation of FGFR1, but did not modify FGF2-induced phosphorylation of the docking protein FRS2. These effects were abolished by the FGFR-selective inhibitors PD166866 and PD161570. In situ hybridization analyses of TR-knockout mice, which have impaired ossification and skeletal mineralization, revealed reduced FGFR1 mRNA expression in osteoblasts and osteocytes, whereas T3 failed to stimulate FGFR1 mRNA or enhance FGF2-activated MAPK signaling in TR-null osteoblasts. These findings implicate FGFR1 signaling in T3-dependent bone development and the pathogenesis of skeletal disorders resulting from thyroid disease. |
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
In hyperthyroidism, bone remodeling is accelerated and activities of bone-forming osteoblasts and bone-resorbing osteoclasts are disproportionately increased, leading to a net loss of 10% of mineralized bone per remodeling cycle (8). T3 stimulates bone resorption in organ cultures, but studies of isolated osteoclasts have shown this effect requires cocultured osteoblasts (9). T3 has been found to stimulate, inhibit, or exert no effect on osteoblastic cell proliferation, but a consensus suggests that T3 stimulates osteoblast activity (1). Thus, T3 has been reported to increase production of osteocalcin (10, 11), collagenase 3 (matrix metalloproteinase-13), gelatinase B (matrix metalloproteinase-9), tissue inhibitor of metalloproteinase-1 (12), alkaline phosphatase (10), IGF-I, IGF-binding protein-2 and -4 (13, 14), and IL-6 and -8 (15) in various systems, although no studies have addressed the mechanisms of their activation in detail or the downstream consequences of their stimulation.
T3 actions are mediated by nuclear TRs, which act as hormone-inducible transcription factors (16). Several TR and TRß mRNA isoforms are expressed in tissue-specific patterns throughout development and adulthood. In the skeleton, TR and -ß mRNAs and proteins are expressed at sites of new bone formation in vivo (17) and in primary cultured osteoblasts (18), osteoblastic bone marrow stromal cells (19), and growth plate chondrocytes in vitro (20), indicating that cells of the osteoblast and chondrocyte lineages are major T3-responsive cells in bone. Data from TR-null mice (21, 22, 23, 24) support the notion that T3 exerts direct actions in bone and indicate that TR is required for bone development and mineralization.
To investigate T3 action in bone, we sought to identify and characterize T3-responsive signaling pathways in osteoblasts using mRNA subtraction hybridization. Here we describe a new pathway that links T3 and fibroblast growth factor receptor-1 (FGFR1) signaling in bone and that has potential for therapeutic targeting in the treatment of osteoporosis and fracture repair. FGFRs are membrane tyrosine kinase receptors that are expressed widely during embyrogenesis. Three FGFRs are essential for skeletal development (25, 26). FGFR1 is the main isoform in developing limb mesenchyme, and FGFR2 is expressed during mesenchyme condensation (26). Later on in limb development, persistent FGFR1 and FGFR2 expression in the perichondrium and periosteum indicate their future expression in cells of the osteoblast lineage (26). FGFR1 and -2 are expressed in developing cranial bone and regulate intramembranous ossification (25, 26). FGFR3 is largely restricted to proliferating and prehypertrophic chondrocytes in the growth plate, where its localization does not overlap with expression of FGFR1 (26, 27, 28). Activating mutations of FGFR1 cause Pfeiffer’s craniosynostosis syndrome, which is characterized by premature fusion of the skull sutures leading to facial abnormalities and mental retardation (29). Mutations of FGFR2 and FGFR3 cause other craniosynostoses that differ according to the presence of associated abnormalities in the hands and feet, whereas FGFR3 mutations also cause achondroplasia, the most common genetic form of dwarfism (25, 26, 30). We show that FGFR1 lies downstream of T3 in a signaling pathway in osteoblasts that requires TR. The identification of this pathway provides new insights into understanding how thyroid hormones regulate skeletal development and why thyrotoxicosis in children causes advanced bone formation, whereas hypothyroidism delays ossification.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
T3 treatment (0.1–1000 nM, 6 h) stimulated a concentration-dependent increase in FGFR1 mRNA expression in UMR106 and ROS 17/2.8 cells, to a maximum of 2.4- and 3.2-fold, respectively (Fig. 1A
, left). T3 did not stimulate FGFR1 mRNA expression after 1 or 2 h treatment, but increased FGFR1 expression was evident after 6 h treatment and maintained over 48 h (Fig. 1A, right). The time course and magnitude of FGFR1 mRNA expression after T3 treatment correlated with a maximum 4.5-fold increased expression of an approximately sized 120–130 kDa FGFR1 protein, also evident over 6–48 h (Fig. 1B).
|
). FGFR2 and FGFR3 mRNAs were both expressed at low levels, and T3 treatment did not influence their expression (Fig. 1, C and D), as assessed by both RT-PCR and Northern blotting, during the time period over which FGFR1 mRNA was induced by T3 (FGFR2: 0 h, 1.14 ± 0.12; 1 h, 1.07 ± 0.09; 2 h, 0.96 ± 0.11; 6 h, 1.16 ± 0.14; FGFR3: 0 h, 1.04 ± 0.10; 1 h, 1.05 ± 0.16; 2 h, 1.10 ± 0.14; 6 h, 1.14 ± 0.18 4; FGFR mRNA concentration ratio ± SEM [+T3:Control] at each time point in Northern blots, normalized to 18S RNA; n = 3). FGFR4 was not expressed in terminally differentiated ROS17/2.8 cells, fibroblastic ROS25/1, or in fetal tibial chondrocytes, although FGFR4 mRNA was observed in preosteoblastic UMR106 cells, in which expression was apparently increased 2-fold by T3 when assessed by RT-PCR over 30 cycles (Fig. 1C, right). FGFR4 mRNA expression was below the sensitivity for detection by Northern blotting.
T3 Stimulates FGFR1 Gene Expression Indirectly
We investigated the mechanism of T3-regulated FGFR1 mRNA expression using the metabolic inhibitors, actinomycin D and cycloheximide, which block gene transcription or protein synthesis, respectively. We preincubated UMR106 or ROS17/2.8 cells or primary cultured osteoblasts from either Lewis or WKY rat strains with actinomycin D or cycloheximide for 1 h, before addition of T3 (100 nM, 6 h) in the continued presence of inhibitor. Treatment with either inhibitor did not reduce basal levels of FGFR1 mRNA but abolished T3-stimulated FGFR1 expression (Fig. 2A). Thus, continuing gene transcription and protein synthesis are required for T3 stimulation of FGFR1 mRNA expression in both osteosarcoma cells and primary cultured osteoblasts.
|
), indicating that T3-mediated induction of FGFR1 mRNA did not result from altered mRNA stability in the presence of hormone. Taken together, these data suggest that T3 stimulation of FGFR1 mRNA expression is indirect and requires the transcription and translation of a T3-responsive intermediary factor before FGFR1 expression is increased.
Fibroblast Growth Factor 2 (FGF2)-Stimulated Activation of MAPK Signaling Is Enhanced by T3
We next tested whether the stimulatory effects of T3 on FGFR1 mRNA and protein expression produce functional consequences. Binding of FGF to FGFR1 results in phosphorylation and functional activation of the receptor, which then signals via the MAPK pathway and other second messenger systems (31). Treatment of ROS17/2.8 cells with FGF2 (0.05–5.0 ng/ml) for 10 min resulted in a concentration-dependent 28-fold maximum increase in phosphorylation of the p44 and p42 components of the MAPK pathway. There was a 2-fold increase in MAPK signaling after stimulation by FGF2 at all concentrations after cells were pretreated for 6 h with T3 (100 nM) (Fig. 3A), indicating that T3 enhances FGF-stimulated MAPK signaling in ROS17/2.8 cells.
|
). In cells pretreated with T3 (100 nM) for 6 h, there was a 2-fold increase in FGF2-stimulated MAPK signaling after 5, 10, 30, and 60 min of FGF2 treatment. Additionally, FGF2-stimulated MAPK signaling was seen after only 2 min exposure to growth factor in cells preincubated with T3. At the 90-min time point, T3 enhancement of FGF2-stimulated MAPK signaling declined and was no longer evident by 120 min (Fig. 3B). Thus, T3 increased the sensitivity of the MAPK signaling pathway to stimulation by FGF2 such that MAPK was activated earlier by FGF2 and to a greater degree. T3, however, did not prolong the stimulatory actions of FGF2.
Thyroid hormones have also been reported to activate MAPK activity rapidly, possibly via nongenomic actions at the cell surface (32, 33, 34). We investigated this putative pathway in ROS17/2.8 cells by preincubating cells with T3 (100 nM) for 30 min before incubation with FGF2 (5 ng/ml). T3 alone failed to stimulate phosphorylation of p44 and p42 (Fig. 3C), in accord with the lack of MAPK activation seen in cells treated with T3 for 6 h (Fig. 3B). Furthermore, pretreatment of cells with T3 for 30 min failed to enhance FGF2-stimulated MAPK signaling (Fig. 3C), in contrast to the effects of T3 after 6 h pretreatment (Fig. 3B). Thus, T3 induces functionally active FGFR1 mRNA and protein after 6 h and does not mediate significant nongenomic actions that influence FGF or MAPK signaling in osteoblasts.
T3 Does Not Influence MAPK Signaling in Cells Treated with Epidermal Growth Factor (EGF) or Platelet-Derived Growth Factor (PDGF)
To determine whether other growth factors might be involved in activation of MAPK in ROS17/2.8 cells, we examined the effects of EGF and PDGF in the absence or presence of T3. EGF treatment over a 30-min time period did not stimulate MAPK activity above basal values, and pretreatment of cells with T3 for 6 h did not affect MAPK activity in the absence or presence of EGF (Fig. 4A). Similarly, treatment of cells with PDGF, in the absence or presence of T3, failed to stimulate MAPK activity above basal levels (Fig. 4B). These negative findings are in keeping with data indicating that ROS17/2.8 cells do not express EGF receptors (35). Similarly, very few studies have investigated the effects of PDGF in ROS17/2.8 cells (36, 37), and there are no data to indicate that PDGF signaling is significant in these cells. Thus, FGF-mediated activation of MAPK appears to be a major route by which MAPK signaling is activated in ROS17/2.8 cells by receptor tyrosine kinases.
|
). FGF2 stimulated MAPK by 27-fold in ROS 17/2.8 cells, and this response was doubled when cells were preincubated with T3 for 6 h. Cotreatment with PD166866 abolished FGF2-stimulated MAPK signaling in the absence or presence of T3 with an IC50 of 50 nM. Similar findings were observed with PD161570, with an IC50 of 150 nM. These IC50 values correspond well with published IC50 concentrations for the inhibition of FGFR1 by each antagonist in other cell systems (38, 39, 40). In primary osteoblasts obtained from Lewis or WKY rats, FGF2 stimulated MAPK signaling by approximately 28-fold (Lewis: 25.8-fold, n = 2; WKY: 30.5-fold, n = 1), and this response was doubled in the presence of T3 [Lewis, 2.35-fold; WKY, 1.97-fold; mean increase in activated/basal MAPK ratio after pretreatment of primary osteoblasts with T3 (100 nM) before stimulation with FGF2 (5 ng/ml, 10 min); Lewis, n = 2; WKY, n = 1)]. In contrast to osteoblastic osteosarcoma cells, basal MAPK activity in primary osteoblasts was also increased by 2-to 3-fold in the presence of T3. This finding probably results from differing cell culture conditions; osteosarcoma cells were cultured serum free, and primary osteoblasts require 15% fetal calf serum (FCS), suggesting that the presence of growth or related factors in serum accounts for T3-enhanced basal MAPK activity in primary cultures. Nevertheless, cotreatment of primary osteoblasts with FGF2 and PD161570 in both the absence and presence of T3 abolished the FGF2-stimulated increase in MAPK (Fig. 5B). Thus, FGF2-stimulated MAPK signaling in osteoblasts, and its enhancement by T3, requires FGFR1.
|
). These data indicate that, similar to its effect on MAPK signaling, T3 increased the sensitivity of FGFR1 to FGF2 such that the receptor was autophosphorylated earlier by FGF2 and to a greater degree (by 
2-fold) in cells that were pretreated with T3. The response required FGFR1 because autophosphorylation of the receptor was inhibited by either PD166866 or PD161570 (Fig. 6A).
|
). These data indicate that, although T3 enhances both FGFR1 autophosphorylation and MAPK activation by FGF2 with similar kinetics and to a similar degree, the two events are not coupled to FRS2 because T3 does not enhance FGF2-stimulated phosphorylation of FRS2.
FGF2 Does Not Activate Phospholipase C
(PLC2) Signaling in ROS17/2.8 Cells
An alternative pathway for FGFR signaling is by direct coupling, via autophosphorylated tyrosine residues, to other signaling molecules such as PLC (31). However, PLC2 was not expressed in ROS17/2.8 cells pretreated with T3 and stimulated with FGF2. Furthermore, activated phosphorylated PLC2 was not detected in treated ROS17/2.8 cells by immunoprecipitation. In contrast, both basal and activated PLC2 expression was present in FGF2-treated chondrogenic ATDC5 cells in the absence or presence of T3 (Fig. 6C). These data indicate that FGF signaling via PLC2 is not a significant pathway in osteoblastic ROS17/2.8 cells.
FGFR1 mRNA Expression Is Reduced in TR0/0 Osteoblasts in Vivo
To determine whether the regulatory effects of T3 in vitro were evident in vivo, we examined skeletal FGFR1 mRNA expression in TR0/0 mice and wild-type littermate controls by in situ hybridization. We have shown previously that TR0/0 mice exhibit growth retardation with reduced bone mineralization (22). FGFR1 mRNA was expressed clearly in osteoblasts lining trabecular bone surfaces in the secondary ossification centers of the epiphyses of wild-type mice, whereas expression was barely detectable in a minority of such osteoblasts in TR0/0 mice (Fig. 7A). Furthermore, FGFR1 mRNA was expressed strongly in osteoblasts lining cortical diaphyseal bone surfaces and was clearly evident in osteocytes located within cortical bone lacunae in wild-type mice. In contrast, FGFR1 expression was barely detectable in equivalent osteoblasts in TR0/0 cortical bone (Fig. 7A). Thus, FGFR1 mRNA expression is reduced in osteoblasts and osteocytes in TR0/0 mice.
|
2 and FGFR2 mRNAs in osteoblasts in wild-type and TR0/0 mice by in situ hybridization (Fig. 7B). Both collagen I2 and FGFR2 were expressed clearly in osteoblasts lining both trabecular bone and cortical diaphyseal bone surfaces. There were no differences in expression of either collagen I2 and FGFR2 mRNAs between wild-type and TR0/0 osteoblasts (Fig. 7B). These data indicate that the reduced expression of FGFR1 mRNA seen in TR0/0 osteoblasts (Fig. 7A) is not a generalized and nonspecific finding.
Enhancement of FGF2-Stimulated MAPK Signaling by T3 Is Abolished in Primary TR0/0 Osteoblasts
To determine whether FGFR1 mRNA was induced by T3 in TR0/0 osteoblasts, we prepared primary osteoblasts from wild-type and TR0/0 mice for Northern blotting studies. In contrast to levels of FGFR1 mRNA expression in vivo (Fig. 7A), there were similar basal levels of FGFR1 mRNA expression in primary cultured wild-type and TR0/0 osteoblasts. However, T3 failed to stimulate FGFR1 mRNA in TR0/0 osteoblasts whereas there was a 2.4-fold stimulation of expression in wild-type osteoblasts treated with T3 (Fig. 8A). Thus, T3-stimulation of FGFR1 expression is impaired in TR0/0 osteoblasts. The apparent discrepancy between the basal levels of FGFR-1 mRNA expression in wild-type and TR0/0 osteoblasts in vitro and in vivo can be explained as follows. In wild-type mice, normal circulating T3 concentrations in the presence of TR1 produce a sustained stimulation of FGFR-1 expression. In contrast, in TR0/0 mice, which also have normal circulating T3 concentrations, the absence of TR1 prevents this sustained stimulation of FGFR1. As a consequence, basal expression of FGFR1 is lower in TR0/0 mice compared with wild-type (Fig. 7A). In contrast, primary osteoblasts from wild-type or TR0/0 mice were cultured in the absence or presence of T3 (Fig. 8A). In the absence of T3, FGFR-1 expression is at its basal, low level in both wild-type and TR0/0 cells because of the lack of hormone and irrespective of the absence or presence of TR1. In wild-type osteoblasts in the presence of T3, FGFR1 expression is activated, but this does not occur in TR0/0 cells because they lack TR1. Taken together, the data from in vivo and in vitro studies support the attribution of a central role for FGFR-1 in the mediation of T3 effects on osteoblast activity and bone metabolism.
|
0/0 osteoblasts, we next examined FGF2-stimulated MAPK signaling in primary osteoblasts from wild-type and TR0/0 mice. FGF2 (5.0 ng/ml) stimulated MAPK signaling in wild-type osteoblasts to a maximum of 32-fold over a 30-min time course, which is comparable to levels of stimulation observed in ROS 17/2.8 cells and primary rat osteoblasts. Furthermore, as in rat osteoblasts, there was a 2-fold increase in FGF-stimulated MAPK signaling in cells that were preincubated for 6 h with T3 (100 nM). In contrast, FGF2-stimulated MAPK signaling was not enhanced in TR0/0 osteoblasts that were preincubated for 6 h with T3, although FGF-stimulated MAPK activity in the absence of T3 was similar to wild type (Fig. 8). These data indicate that T3 enhancement of FGF2-stimulated MAPK signaling in osteoblasts requires TR. Thus, the FGF-FGFR-MAPK signaling pathway remains active in TR0/0 osteoblasts but is not sensitive to the regulatory effects of thyroid hormones. |
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
. The physiological importance of these findings is indicated by the key roles that thyroid hormones and FGFs play in skeletal development (1, 26). Accelerated skeletal maturation and short stature in childhood thyrotoxicosis is associated with premature closure of the growth plates (2) and craniosynostosis (5), whereas activating mutations of FGFR1 cause Pfeiffer’s craniosynostosis syndrome, which results from advanced bone formation (29, 30). Furthermore, endochondral ossification is disrupted in hypothyroidism (17), and skeletal development and bone mineralization are delayed in TR0/0 mice (22). On the other hand, in adult thyrotoxicosis bone loss results in an increased risk of osteoporotic fracture (1, 6). Our findings, therefore, implicate altered FGFR1 signaling in the pathogenesis of skeletal abnormalities associated with thyroid disease, and this pathway provides a new therapeutic target for the stimulation of bone formation and mineralization.
We showed that T3 stimulates FGFR1 expression in osteoblastic cells and in rat and murine primary calvarial osteoblasts by an indirect mechanism (Figs. 1 and 2). Studies with transcription and protein synthesis inhibitors indicated that stimulation of FGFR1 mRNA by T3 is independent of mRNA stability and requires the prior synthesis of a T3-inducible factor. Such a factor could regulate FGFR1 gene transcription directly to cause the observed T3 stimulation of steady-state FGFR1 mRNA levels. Alternatively, T3 acting via TR could regulate splicing or stability of the FGFR1 primary transcript by a mechanism that involves additional cofactors, as described recently for other nuclear receptors (41), or may influence mRNA export to the cytoplasm. Clarification of these issues will require detailed studies to address the precise mechanism of FGFR1 gene regulation by T3.
To determine whether increased FGFR1 expression resulted in increased functional activity, we studied the downstream MAPK signaling pathway and the effect of two selective FGFR1 antagonists (38, 39, 40). These experiments suggested strongly that FGF2 activates MAPK signaling in osteoblasts via FGFR1 and indicated that enhanced signaling in the presence of T3 is also mediated via FGFR1 (Figs. 3 and 5). Nevertheless, the possibility that activation of MAPK could be mediated in part by other FGFRs, in addition to FGFR1, needs to be considered. The specificity of FGFR1 signaling was investigated using the only two described tyrosine kinase receptor inhibitors that are reported to be selective for FGFR1, PD166866 and PD161570. It is important to note that these antagonists have been tested only for FGFR1 specificity against a series of other tyrosine kinase receptors, but not in comparison with other FGFRs (38, 39, 40). In this context, we excluded the possibility that MAPK activation could have been mediated by the EGF and PDGF receptor tyrosine kinases in these studies (Fig. 4), supporting the conclusion that enhanced FGF-activated MAPK signaling in the presence of T3 is mediated by FGFR. Nevertheless, it is not possible to conclude definitively whether FGFR signaling in these studies is restricted to FGFR1, although the IC50 values identified here are equivalent to those reported for antagonism of FGFR1 (38, 39, 40) and support the contention that FGFR1 is functionally predominant. In spite of this, FGFR2 is also known to be expressed in osteoblasts (25, 26), and some of the observed effects could, therefore, have been mediated via FGFR2 in addition to FGFR1. This particular issue can only be resolved if new specific FGFR1- or FGFR2-restricted antagonists become available. Significant contributions from FGFRs 3 and 4 to our findings, however, can be excluded since FGFR3 expression is largely restricted to growth plate chondrocytes and does not overlap with FGFR1 (26), whereas FGFR4 is not implicated in bone development (25, 30). Despite these observations, we did detect expression of FGFRs 2 and 3 in ROS17/2.8 cells by RT-PCR, although their levels were low. FGFR4 mRNA expression was not detectable in ROS17/2.8 cells by RT-PCR or Northern blotting studies (Fig. 1). T3 did not regulate expression of FGFRs 2 and 3 (Fig. 1D) and, taken together with in situ hybridization data indicating no change in expression of FGFR2 in TR0/0 mice (Fig. 7B), these findings strongly support the view that interactions between T3 and FGF signaling in osteoblasts occur via FGFR1. Interestingly, FGFR4 mRNA was expressed in preosteoblastic UMR106 cells, but not in terminally differentiated ROS17/2.8 cells, and expression was increased 2-fold after T3 treatment in nonquantitative RT-PCR experiments (Fig. 1). These findings raise the possibility that T3 modulation of FGFR signaling may involve FGFR4 in addition to FGFR1 in specific populations of preosteoblastic cells, although this possibility will require detailed future investigation.
We also considered whether activation of MAPK by T3 could have been a direct response to the hormone since T3 has been reported to exert rapid nongenomic actions at the cell surface, which occur within 10–30 min (32, 33, 34) before most genomic responses could be anticipated. This pathway was excluded as being responsible for T3-enhanced FGF responsiveness in experiments in which we preincubated cells with T3 for 30 min before FGF2 stimulation instead of the 6 h preincubation required for induction of FGFR1 mRNA and protein (Fig. 3C). Moreover, the requirement for nuclear receptors to mediate T3 enhancement of FGF2 signaling was further demonstrated by experiments that showed skeletal FGFR1 expression is markedly reduced in vivo in osteoblasts from TR0/0 mice (Fig. 7A), in which ossification and bone mineralization are impaired (22). Furthermore, T3 failed to stimulate FGFR1 mRNA expression or enhance FGF2-stimulated MAPK activity in TR0/0 osteoblasts (Fig. 8). Taken together, these data demonstrate that T3 enhances an FGF2-stimulated and FGFR1-mediated MAPK signaling pathway in osteoblasts via a genomic mechanism that requires TR.
T3 enhancement of FGFR-mediated activation of MAPK was qualitative as well as quantitative and occurred 2 min after FGF stimulation in the presence of T3 but only after 5 min in the absence of T3 (Fig. 3B). While it is clear that the increased magnitude of response in the presence of T3 correlates well with the increased expression of FGFR1 mRNA and protein, the mechanism underlying the qualitative more rapid effect is not known. Thus, it is not known whether the qualitative rapid autophosphorylation of FGFR1 is a direct nongenomic effect of T3 or whether it requires new protein synthesis similar to T3 induction of FGFR1 expression (Fig. 2). One possibility is that, in addition to stimulating the increase in FGFR1 gene expression, T3 could affect additional components of the signaling cascade. We investigated this possibility further. T3 enhanced the qualitative and quantitative FGF2-induced tyrosine autophosphorylation of FGFR1 to a similar degree to its effects on MAPK activation and over the same time course (Fig. 6A). Surprisingly, and in spite of this finding, T3 did not enhance FGF2-induced tyrosine phosphorylation of FRS2 (Fig. 6B), the docking protein that couples FGFR activation to MAPK signaling. This finding indicates that T3 enhancement of FGFR1 autophosphorylation and MAPK signaling in response to FGF2 occurs via a pathway in which FGFR1 activation of MAPK is uncoupled from FRS2. Indeed, FRS2-independent MAPK stimulation that is rapid and distinct from FRS2-dependent signaling in wild-type cells has been demonstrated clearly in FRS2 knockout mice, although its mechanism has not been elucidated (42). Thus, it is possible that T3 may modify the interaction between FGF and FGFR1 such that the receptor becomes more sensitive to FGF and responds more rapidly by undergoing rapid autophosphorylation and activating the FRS2-independent MAPK pathway. In previous studies we demonstrated delayed endochondral ossification and structural alterations in the growth plates of hypothyroid rats that included the deposition of an abnormal cartilage matrix rich in heparan sulfate proteoglycans (17). We identified similar features in TR0/0 null mice (22). Crystallographic studies have demonstrated that heparan sulfate is required for binding of FGF to FGFR and for ligand-induced receptor activity (43, 44). Collectively, these studies suggest that T3-regulated production of heparan sulfate, or modification of its structure, could be an important mechanism whereby T3 modulates FGFR1 signaling in a qualitative and quantitative manner.
There are, however, additional complexities to FGF signaling. More than 20 FGF ligands can activate the four FGFRs (25). FGFRs signal through three main routes in bone cells; the MAPK cascade, via phospholipase C (PLC), or by activation of the signal transducers and activators of transcription (STAT1, 5a and 5b) (31, 45, 46, 47, 48). FGFs induce proliferation or apoptosis in skeletal cells according to the balance of activated MAPK and STAT pathways (31, 48), but the consequences of PLC-activation are unknown. It will be important to clarify 1) whether T3 influences the FGFR1 response to different FGF ligands in a specific fashion; 2) whether FGFR1 signaling via the PLC and STAT pathways is modified in a similar way to effects on the MAPK pathway; 3) whether T3 exerts differential responses on individual FGFR1 downstream signaling pathways; or 4) whether the activities of other FGFRs are T3 regulated. Our demonstration that FGF2 activates PLC2 in chondrocytic, but not osteoblastic, cells (Fig. 6C) suggests further that cell specificity of FGFR signaling is likely to be an important factor involved in skeletal responses to FGFs and thyroid hormones. Such issues demonstrate that our studies provide a new physiologically important field of investigation with which to study the role of FGFR signaling in mediating the effects of T3 on bone development, mineralization, and turnover.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
Cell Culture
Osteoblastic ROS17/2.8 and UMR106 rat osteosarcoma cells were maintained in Ham’s F12 medium (Invitrogen, Paisley, Scotland, UK) supplemented with 5% FCS (Globepharm, Guildford, UK), and cultures were transferred to serum-free medium for 24 h before treatments (49). Murine chondrogenic ATDC5 cells (50) were maintained in DMEM/Hams F12 (1:1; Invitrogen) containing 5% FCS, 10 µg/ml transferrin (Sigma, Poole, UK), and 30 nM sodium selenite (Sigma), as described (51). Primary rat calvarial osteoblasts were obtained from neonatal Lewis and WKY rats as described (52), and murine calvarial osteoblasts were prepared by the same procedure from wild-type and TR0/0 mice, which lack all products of the Thra locus (22). Primary osteoblasts were transferred to medium containing serum stripped of thyroid hormones (53) for 24 h before treatments. Cells were incubated with T3 (0.1–1000 nM, 1–48 h, Sigma), FGF2 (0.05–5 ng/ml, 1–120 min, Invitrogen), EGF (1–50 ng/ml, 1–30 min, Upstate Biotechnology, Inc., Dundee, UK), PDGF-BB (1–50 ng/ml, 1–30 min, Upstate Biotechnology, Inc.), cycloheximide (10 µM, 1–7 h, Sigma), actinomycin D (10 µM, 1–18 h, Sigma), PD166866 (25–250 nM, 10 min, Parke-Davis, Ann Arbor, MI), or PD161570 (40–400 nM, 10 min, Parke-Davis) alone or in various combinations.
mRNA Subtraction Hybridization
Differentially expressed T3-inducible mRNAs were isolated by poly A PCR subtraction hybridization (54). cDNA was prepared by reverse transcribing total RNA extracted from T3- (100 nM, 6 h) and vehicle-treated ROS17/2.8 and UMR106 cells using a dT24 primer, and poly A tails were added using terminal deoxytransferase (Invitrogen). Primary PCR was performed with a NotIdt primer (5'-CATCTCGAGCGGCCGCTTTTTTTTTTTTTTTTTTTTTTTT-3'), and reamplification was performed with either a driver (5'-CTTCGAAGTTTTTTTTTTTTTTTT-3') or tracer (5'-CATCTCGAGCGGCCGCTTTTTTTT-3') primer. Driver cDNA was photobiotinylated with photobiotin (40 µg, Sigma) using an Osram 400W HQ (MB-U) lamp (10 min, 4 C) at a distance of 10 cm. Tracer cDNA (400 ng) was combined with biotinylated driver cDNA (4 µg) and tRNA (5 µg, Sigma) and annealed by heating in a PCR machine to 98 C for 5 min, 80 C for 5 min, ramping from 80-68 C over 15 min, and incubating at 68 C for 1 h. Biotinylated cDNA complexes were precipitated with streptavidin (4 µg) and removed by phenol/chloroform extraction, and three additional rounds of subtraction were performed after addition of further biotinylated driver cDNA (4 µg) to extracted tracer. Experiments in which driver cDNA is obtained from control cells generate subtracted cDNAs that are enriched in T3-treated cells.
Cloning of FGFR cDNAs
Subtracted cDNAs were blotted onto duplicate filters and probed with 32P-labeled first-strand cDNA prepared from control and T3-treated cells. A T3-inducible clone (T7), originating from the 3'-UTR of FGFR1, was isolated and a cDNA from the 5'-UTR and coding region of rat FGFR1 (nucleotides 104–603) was prepared subsequently by RT-PCR and used to probe Northern blots and in situ hybridizations. cDNAs encoding mouse FGFR2 (nucleotides 379–742), FGFR3 (nucleotides 2678–2697), and FGFR4 (nucleotides 2685–3027) were prepared by RT-PCR using RNA extracted from murine chondrogenic ATDC5 cells (50) and used to probe Northern blots. A mouse FGFR2 cDNA (nucleotides 768-1258) was also prepared by RT-PCR for in situ hybridization studies.
Western Blotting
Cells were lysed in lysis buffer [1% Triton X-100, 0.5% sodium dodecyl sulfate, 0.75% deoxycholate, 10 mM Tris-Cl (pH 7.4), 75 mM NaCl, 10 mM EDTA, 0.5 mM phenylmethylsulfonylfluoride, 2 mM sodium orthovanadate, 4 mg/ml leupeptin, 10 mg/ml aprotinin, 50 mM NaF, and 30 mM sodium pyrophosphate]. Extract (20 µg) was resolved by SDS-PAGE, transferred to polyvinylidine difluoride filters, and analyzed by Western blotting using an enhanced chemiluminescence detection system (Amersham Pharmacia Biotech, Arlington Heights, IL) as described (49). A rabbit FGFR1 antibody (1:1000 dilution, Sigma) was used to determine FGFR1 expression. For analysis of MAPK activation, filters were incubated with polyclonal antibodies to nonphosphorylated p42 and p44 components of the MAPK pathway (1:1000 dilution, New England Biolabs, Hitchin, UK). Filters were then stripped at 56 C for 30 sec in 6.25 mM Tris-Cl (pH 6.8), 10 mM ß-mercaptoethanol, 2% sodium dodecyl sulfate, and reprobed with antiphospho-p42/p44 antibodies that recognize phosphorylated p42 and p44 proteins (1:1000 dilution, New England Biolabs).
Immunoprecipitations
Treated cells, from a single well of a six-well tissue culture plate, were lysed in 500 µl lysis buffer as above. Lysates were precleared overnight by incubating with 5 µl goat antimouse IgG (Upstate Biotechnology, Inc.) at 4 C, followed by 1 h incubation with 50 µl protein G-Sepharose (Bio-Rad Laboratories, Inc., Hemel Hempstead, UK). Samples were spun at 12,000 rpm for 3 min and the pellet was discarded. The IgG clearance step was repeated for a further 4 h, followed by 1 h incubation with 50 µl protein G-Sepharose. The supernatant was incubated overnight at 4 C with 5 µl antibody [anti-FRS2/SNT-1 (Upstate Biotechnology, Inc.), anti-PLC 2 (Santa Cruz Biotechnology, Inc., Santa Cruz, CA), or anti-FGFR1 (Sigma)] and precipitated by addition of 50 µl protein G-Sepharose suspension for 1 h at 4 C. Samples were centrifuged at 12,000 rpm for 3 min. The pellet was resuspended in sample loading buffer and denatured at 96 C for 5 min. The sample was recentrifuged at 12,000 rpm for 3 min, and 15 µl of supernatant were resolved by 10% SDS-PAGE. Activated phosphorylated forms of the immunoprecipitated proteins were detected by Western blotting, as described above, using an antiphosphotyrosine 4G10 antibody (Upstate Biotechnology, Inc.).
In Situ Hybridization
The FGFR1 and FGFR2 cDNAs were linearized with SpeI and SmaI. Digoxigenin-labeled probes were synthesized using T7 and SP6 RNA polymerases, respectively (Boehringer Mannheim, Lewes, UK) and used to probe sections obtained from the lower limbs of 3-wk-old TR0/0 mice (22) and their littermate controls, as described (17). In situ hybridizations were also performed using a partial cDNA encoding collagen I2 (nucleotides 3892–4400, kindly provided by Dr. D. W. Rowe, University of Connecticut Health Center) that was linearized with PstI to allow transcription of a cRNA probe using T3 RNA polymerase. Hybridization was detected using alkaline phosphatase-conjugated antidigoxigenin Fab fragments (Boehringer Mannheim), and a bacterial neomycin resistance gene cRNA probe was used as a negative control for all hybridizations (17).
Statistics
Data were analyzed using the unpaired Student’s t test. Values are expressed as mean ± SEM and P < 0.05 was considered statistically significant.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
D.A.S. and C.B.H. contributed equally to this work.
Abbreviations: EGF, Epidermal growth factor; FCS, fetal calf serum; FGF, fibroblast growth factor; FGFR, FGF receptor; PDGF, platelet-derived growth factor; PLC, phospholipase C; TR, T3 receptor; UTR, untranslated region.
Received for publication April 15, 2003. Accepted for publication May 29, 2003.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
locus. Mol Cell Biol 21:4748–4760 gene encoding a thyroid hormone receptor is essential for post-natal development and thyroid hormone production. EMBO J 16:4412–4420[CrossRef][Medline]
2 and a concomitant overexpression of 1 yields a mixed hypo- and hyperthyroid phenotype in mice. Mol Endocrinol 15:2115–2128 in FGF receptor-mediated signal transduction pathways. Proc Natl Acad Sci USA 98:8578–8583NURSA Molecule Pages Link:
This article has been cited by other articles:
 |
|
 |
|
  V. Niederkinkhaus, R. Marx, G. Hoffmann, and I. D. Dietzel Thyroid Hormone (T3)-Induced Up-Regulation of Voltage-Activated Sodium Current in Cultured Postnatal Hippocampal Neurons Requires Secretion of Soluble Factors from Glial Cells Mol. Endocrinol., September 1, 2009; 23(9): 1494 - 1504. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J H D. Bassett and G. R Williams The skeletal phenotypes of TR{alpha} and TR{beta} mutant mice J. Mol. Endocrinol., April 1, 2009; 42(4): 269 - 282. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A Scarlett, M P Parsons, P L Hanson, K K Sidhu, T P Milligan, and J M Burrin Thyroid hormone stimulation of extracellular signal-regulated kinase and cell proliferation in human osteoblast-like cells is initiated at integrin {alpha}V{beta}3 J. Endocrinol., March 1, 2008; 196(3): 509 - 517. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. H. D. Bassett, A. J. Williams, E. Murphy, A. Boyde, P. G. T. Howell, R. Swinhoe, M. Archanco, F. Flamant, J. Samarut, S. Costagliola, et al. A Lack of Thyroid Hormones Rather than Excess Thyrotropin Causes Abnormal Skeletal Development in Hypothyroidism Mol. Endocrinol., February 1, 2008; 22(2): 501 - 512. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. J. O'Shea, C. J. Guigon, G. R. Williams, and S.-y. Cheng Regulation of Fibroblast Growth Factor Receptor-1 (Fgfr1) by Thyroid Hormone: Identification of a Thyroid Hormone Response Element in the Murine Fgfr1 Promoter Endocrinology, December 1, 2007; 148(12): 5966 - 5976. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. H. D. Bassett, K. Nordstrom, A. Boyde, P. G. T. Howell, S. Kelly, B. Vennstrom, and G. R. Williams Thyroid Status during Skeletal Development Determines Adult Bone Structure and Mineralization Mol. Endocrinol., August 1, 2007; 21(8): 1893 - 1904. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. H. D. Bassett, P. J. O'Shea, S. Sriskantharajah, B. Rabier, A. Boyde, P. G. T. Howell, R. E. Weiss, J.-P. Roux, L. Malaval, P. Clement-Lacroix, et al. Thyroid Hormone Excess Rather Than Thyrotropin Deficiency Induces Osteoporosis in Hyperthyroidism Mol. Endocrinol., May 1, 2007; 21(5): 1095 - 1107. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. Rabier, A. J Williams, F. Mallein-Gerin, G. R Williams, and O Chassande Thyroid hormone-stimulated differentiation of primary rib chondrocytes in vitro requires thyroid hormone receptor {beta}. J. Endocrinol., October 1, 2006; 191(1): 221 - 228. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. A. Mousa, L. O'Connor, F. B. Davis, and P. J. Davis Proangiogenesis Action of the Thyroid Hormone Analog 3,5-Diiodothyropropionic Acid (DITPA) Is Initiated at the Cell Surface and Is Integrin Mediated Endocrinology, April 1, 2006; 147(4): 1602 - 1607. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. J. O'Shea, J. H. D. Bassett, S. Sriskantharajah, H. Ying, S.-y. Cheng, and G. R. Williams Contrasting Skeletal Phenotypes in Mice with an Identical Mutation Targeted to Thyroid Hormone Receptor {alpha}1 or {beta} Mol. Endocrinol., December 1, 2005; 19(12): 3045 - 3059. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. C. Barnard, A. J. Williams, B. Rabier, O. Chassande, J. Samarut, S.-y. Cheng, J. H. D. Bassett, and G. R. Williams Thyroid Hormones Regulate Fibroblast Growth Factor Receptor Signaling during Chondrogenesis Endocrinology, December 1, 2005; 146(12): 5568 - 5580. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
) or T3 (
). FGFR1 mRNA expression was determined by Northern blotting, and expression at each time point was normalized to FGFR1 mRNA concentration after actinomycin D pretreatment. Data (mean ± SEM) are from three independent experiments, and similar findings were seen in UMR106 cells (n = 3).
), in osteoblasts lining cortical bone in the diaphysis (
), and in osteocytes within cortical bone (
). tb, Trabecular bone; cb, cortical bone; bm, bone marrow. B, Expression of collagen I