help button home button Endocrine Society Molecular Endocrinology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Molecular Endocrinology, doi:10.1210/me.2005-0515
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
20/12/3279    most recent
Author Manuscript (PDF)
Right arrow Purchase Article
Right arrow View Shopping Cart
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Paternot, S.
Right arrow Articles by Roger, P. P.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Paternot, S.
Right arrow Articles by Roger, P. P.
Molecular Endocrinology 20 (12): 3279-3292
Copyright © 2006 by The Endocrine Society

Differential Utilization of Cyclin D1 and Cyclin D3 in the Distinct Mitogenic Stimulations by Growth Factors and TSH of Human Thyrocytes in Primary Culture

Sabine Paternot, Jacques E. Dumont and Pierre P. Roger

Institute of Interdisciplinary Research (IRIBHM), Université Libre de Bruxelles, Campus Erasme, B-1070 Brussels, Belgium

Address all correspondence and requests for reprints to: Pierre P. Roger, Institute of Interdisciplinary Research, Université Libre de Bruxelles, Campus Erasme, 808 route de Lennik, B-1070 Brussels, Belgium. E-mail: proger{at}ulb.ac.be.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 
Two distinct mitogenic modes coexist in thyroid epithelial cells. TSH via cAMP induces proliferation and differentiation expression, whereas growth factors including epidermal growth factor (EGF) induce proliferation and dedifferentiation. Divergent models of TSH/cAMP-dependent mitogenesis have emerged from different thyroid cell culture systems. In the FRTL-5 rat cell line, cAMP cross-signals with transduction pathways of growth factors to induce cyclin D1 and p21cip1 and down-regulate p27kip1. By contrast, in canine primary cultures, mitogenic pathways of cAMP and growth factors are fully distinct. cAMP does not induce D-type cyclins and p21, it up-regulates p27, and it stimulates the formation and activity of cyclin D3-cyclin-dependent kinase (CDK) 4 complexes. In primary cultures of normal human thyrocytes, EGF + serum increased cyclin D1 and p21 accumulation, and it stimulated the assembly and activity of cyclin D1-CDK4-p21 complexes. By contrast, TSH repressed or did not induce cyclin D1 and p21, and it rather up-regulated p27. TSH did not increase cyclin D1-CDK4 activity, but it stimulated the activating phosphorylation of CDK4 and the pRb-kinase activity of preexisting cyclin D3-CDK4 complexes. As recently demonstrated in dog thyrocytes and other systems, cyclin D1 and cyclin D3 differently oriented the site specificity of CDK4 pRb-kinase activity, which might differently impact some pRb functions. Cyclin D1 or cyclin D3 are thus differentially used in the distinct mitogenic stimulations by growth factors or TSH, and potentially in hyperproliferative diseases generated by the overactivation of their respective signaling pathways. At variance with dog thyroid primary cultures, rat thyroid cell lines might not be valid models of TSH-dependent mitogenesis of human thyrocytes.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 
ABNORMALLY ACTIVATED OR deregulated thyroid cell proliferation is associated with several pathological states including goitre and the various thyroid adenomas and carcinomas. Most if not all cell proliferation effects of TSH, TSH receptor-stimulating autoantibodies (TSAbs), activating mutations of TSH receptor, or Gs{alpha} are mediated by elevation of cellular cAMP levels (1), according to a paradigm first demonstrated using primary cultures of canine thyroid cells (2) and confirmed in primary cultures of normal human thyrocytes (3). The clinical relevance includes TSH-dependent goitre, Graves’ disease (4), autonomous hyperfunctional adenomas (5), or congenital hyperthyroidism (6). Elevated TSH and TSAbs also favor the development of at least some differentiated thyroid carcinomas (7, 8). Whereas the cAMP-dependent stimulation is associated with enhanced thyroid function and differentiation expression (9), the cAMP-independent mitogenic stimulation by growth factors such as epidermal growth factor (EGF) is associated with reversible dedifferentiation of the cells (10, 11). The coexpression of some growth factors and their receptors might explain the relative TSH independence of multinodular goitre (12). Moreover, the subversion of tyrosine kinase pathways similar to those operated by local growth factors [i.e. the activation of Ret/PTC (13, 14), or TRK (15), the permanent activation of B-Raf by mutation (16) or chromosomal rearrangement (17), the overexpression of Met/hepatocyte growth factor receptor sometimes in association with hepatocyte growth factor (18), or ErbB1/EGF receptor in association with its ligand TGF{alpha} (19)] is causally associated with TSH-independent thyroid papillary carcinomas.

To investigate cellular and molecular mechanisms of mitogenesis stimulated by TSH/cAMP, several cell systems have been used, including rat thyroid cell lines (FRTL-5, WRT, and PC Cl3 cells), and the primary cultures of dog and human thyrocytes that also allow the comparison with growth factor-dependent mechanisms (20, 21). Disappointingly, the confrontation of available data on signaling cascades, cell cycle kinetics and cell cycle-regulatory proteins has led to divergent models of TSH/cAMP-dependent mitogenesis in rat cell lines and canine primary cultures (21, 22). In WRT cells, TSH reportedly activates Ras (23) and phosphatidylinositol 3-kinase (PI3-kinase) (24). In FRTL-5 cells, TSH/cAMP activates ERKs (25) and p38 MAPK (26), and induces early genes such as c-jun (27), egr-1 (28), and c-myc (29). It induces cyclin D1 (30) and down-regulates the cyclin-dependent kinase (CDK) inhibitor p27kip1 (31) [while increasing the related protein p21cip1 (32)], as generally observed in the response to growth factors that activate Ras and the MAPK and PI3-kinase pathways (33). Possibly through these mechanisms, TSH exerts a priming effect that makes the cell more competent to progress into G1 phase in response to insulin/IGF-I alone, which is thus qualified as the only genuine mitogen (30, 34). Furthermore, TSH presence is dispensable during G1 phase progression, and maintenance of high cAMP levels even delays DNA synthesis initiation (35). In these cells, TSH also potentiates IGF-I action, which involves a cAMP-dependent augmentation of IGF-I signaling pathways leading to the activations of MAPKs and PI3-kinase (34).

In dog thyrocytes, TSH and insulin/IGF-I also combine to induce proliferation (36). However, TSH main action is essentially different in these cells. Unlike growth factors and/or insulin, TSH does not activate Ras (37), PI3-kinase, and protein kinase B (38), or the different MAPKs (39). It down-regulates the expression of c-myc (after an initial induction) (40), c-jun, and egr-1 (41, 42). As a likely consequence, TSH rather inhibits the accumulation of D-type cyclins (43), but it stimulates the expression of p27kip1 (44). However, CDK4 activity (45) and cyclin D3 (43, 46) are required for TSH-stimulated DNA synthesis in dog thyrocytes. TSH has to continuously elevate cAMP levels to directly control the passage through the restriction point (47), via critical but distinct actions on the assembly (43), nuclear translocation (48, 49) and activating phosphorylation of CDK4 (50) bound to cyclin D3 (43), itself synthesized in response to insulin (46). Therefore, at least in a first stage, the formation and activation of cyclin D3-CDK4 complexes and the phosphorylation of pRb in dog thyrocytes result from distinct but complementary actions of TSH and insulin, rather than an interaction at earlier steps of the signaling cascades, as reported from FRTL-5 cells (21). On the other hand, cyclin D3 is not required in the mitogenic stimulation of dog thyrocytes by growth factors that induce cyclin D1 (43) and more transiently p21cip1 (51) but rather reduce p27 levels. Paradoxically, the related CDK inhibitors p27 or p21 are differentially used as positive regulators that determine the nuclear translocation of CDK4 and support its activity in the distinct mitogenic stimulations by TSH or growth factors, respectively (51).

A crucial question is to discriminate which of these divergent models, if any, could apply to normal human thyrocytes. The induction of the different D-type cyclins and the expected resulting activation of CDK4 are the most generally considered endpoints of mitogenic signaling cascades (52). They thus reflect the activation state of the different mitogenic pathways. In the present study, we show the differential utilization of cyclin D1 or cyclin D3 in CDK4 activation in the mitogenic stimulations of primary cultures of human thyrocytes by EGF + serum or TSH. This observation indicates that the involved upstream signaling cascades must be largely distinct, as found in dog thyrocytes, but at variance with the situations reported from rat thyroid cell lines.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 
The present study was based on the analysis of 19 primary cultures obtained between 1998 and 2005 from histologically normal tissue. As previously described (3, 53, 54), the seeding of follicles in low-serum conditions and the subsequent culture in serum-free medium ensure that the primary cultures consisted of at least 99% of cytokeratin-positive thyrocytes that can express thyroid differentiation (thyroglobulin, thyroperoxidase, and iodide transport in response to TSH stimulation). In all these primary cultures, cells were stimulated at d 5 or 6 by TSH or EGF + serum in the presence of insulin. As previously shown, both treatments could induce DNA synthesis at levels that approach those observed in canine thyroid primary cultures (2) and rat thyroid cell lines (see Fig. 4Go). However, as compared with our previous experience with similar primary cultures obtained from normal thyroid glands of donors deceased for non-thyroid-related causes (3, 53) (mostly traffic accident victims; this material was no longer available in the present study, as an indirect consequence of a modification of Belgian laws), DNA synthesis responses were more variable (Fig. 1Go). In several primary cultures, DNA synthesis was increased either by TSH (and the adenylyl cyclase activator forskolin) or by EGF + serum (Fig. 1Go). Thus, no correlation could be observed between the cell responsiveness to cAMP-dependent or cAMP-independent mitogenic stimulations. In two experiments, both TSH and EGF + serum failed to appreciably induce DNA synthesis. When investigated, an increased proportion of DNA-replicating cells [bromodeoxyuridine (BrdU)-positive nuclei] was first detected about 26 h after stimulation both by TSH (or forskolin) and EGF + serum, and this proportion only slowly but steadily augmented afterwards (data not shown). Primary cultures that are numbered in Fig. 1Go are those for which the investigation of cell cycle regulatory proteins is further illustrated in this study.


Figure 4
View larger version (38K):
[in this window]
[in a new window]
  		Fig. 4. Cell Cycle Regulation in Rat Thyroid FRTL-5 Cells

Cells were made quiescent by a three-day culture without serum and hormones ({alpha} treatment) (A and B), and then stimulated with none ({alpha}), 5 µg/ml insulin ({alpha}I), 1 mU/ml TSH ({alpha}T) or both ({alpha}IT). Alternatively, cells were made quiescent by a three-day culture with 5 µg/ml insulin only (I treatment) (A and C), and then remained in the presence of insulin (I), or were stimulated by TSH in the presence of insulin (IT). A, DNA synthesis was evaluated in cells that were stimulated for 48 h. BrdU was present for the last 24 h and the percentage of nuclei having incorporated BrdU was determined (mean + range of duplicate dishes). B and C, Cells were stimulated for 20 h, and the Thr826-phosphorylation of pRb (specifically ascribed to CDK4) (P-Rb826), CDK4, cyclin D1 (cyc D1), cyclin D3 (cyc D3), p21 and p27 were detected by Western blotting from whole cell extracts. CDK4 was also detected from cyclin D1 immunoprecipitations (IP) (cyc D1-bound CDK4), cyclin D3 IP (cyc D3-bound CDK4) and p21 IP (p21-bound CDK4).

 

Figure 1
View larger version (27K):
[in this window]
[in a new window]
  		Fig. 1. DNA Synthesis Responses to TSH (T), Forskolin (F), EGF + Serum (ES), or None (Control; C) from 19 Independent Primary Cultures of Human Thyrocytes

Cells were stimulated at d 5 or 6 for 48 h, with BrdU for the last 24 h. The percentage of nuclei having incorporated BrdU was determined. When indicated, ranges are from duplicate Petri dishes. Numbered cultures are those for which the investigation of cell cycle regulatory proteins is further illustrated in this study.

 
Expression of Cell Cycle Regulatory Proteins in Human Thyrocytes Stimulated by TSH or EGF + Serum
In 13 of these primary cultures, the phosphorylation of pRb and the expression of proteins that regulate the activity of CDK4 (and CDK2 in a subset of these experiments) were investigated by Western blotting. The most comprehensive analysis was obtained 26 h after stimulation in culture no. 1 (see Fig. 1Go), which responded well to the mitogenic stimulation by TSH and more weakly to EGF + serum (Fig. 2Go). Both mitogenic treatments induced the hyperphosphorylation of pRb (as shown by the appearance of a slower migrating band), consistent with the increased entry of cells in S phase. Cyclin E presence was only slightly increased by both mitogenic stimulations (Fig. 2Go), probably due to the weak synchronicity of the cell cycle progression. CDK4 and CDK2 were slightly increased by TSH or EGF + serum, respectively, but these weak effects were not consistently observed in other experiments. Cyclin H and CDK7 that constitute the CDK activating kinase (CAK), were not up-regulated by mitogenic treatments (CDK7 level was even moderately reduced by EGF + serum) (Fig. 2Go), as observed in many cell systems. CDC25A, a phosphatase involved in CDK2 activation, was unexpectedly down-regulated by EGF + serum, as previously observed in dog thyrocytes (our unpublished data). On the other hand, cyclin D1 accumulation was stimulated by EGF + serum but not by TSH, whereas cyclin D3 expression was constitutive (Fig. 2Go). Cyclin D2 was undetectable by Western blotting in human thyrocytes (negative data not shown), whereas CDK6 was very weakly expressed compared with other human cells (not shown). In another primary culture (no. 2 in Fig. 1Go), cyclin D1, cyclin D3, and CDK4 were analyzed 8, 16, or 26 h after stimulation by fetal calf serum (FCS) (10%), forskolin, or TSH (Fig. 3Go). After 16 and 26 h of stimulation, cyclin D1 was induced by FCS, but markedly repressed by forskolin and TSH, despite a stronger stimulation of DNA synthesis by forskolin and TSH compared with FCS. Cyclin D3 and CDK4 levels did not show such a down-regulation (Fig. 3Go). Overall, cyclin D1 accumulation was stimulated by EGF + serum in all the primary cultures (13 cultures) but two (in which no DNA synthesis was induced). It was never increased by TSH and forskolin, and in eight of 13 experiments, it was even repressed by TSH. Conversely, cyclin D3 expression was never stimulated by EGF + serum, but slightly enhanced by TSH (maximum 50%) in about half the cultures (seven of 13). The levels of the CDK inhibitors of the CIP/KIP family (p21, p27, and p57) were not decreased after stimulation. Indeed, p21 expression was even increased after EGF + serum stimulation (reproduced in five of five cultures), and p27 was most often (seven of 10 cultures) up-regulated by TSH (Fig. 2Go).


Figure 2
View larger version (41K):
[in this window]
[in a new window]
  		Fig. 2. Cell Cycle Regulatory Proteins in a Human Thyroid Primary Culture

Cells from culture no. 1 were stimulated or not (C) for 26 h with TSH (T) or EGF + serum (ES). pRb, cyclin D1, cyclin D3, cyclin E, cyclin H, CDK4, CDK2, CDK7, p21, p27, p57, and CDC25A were detected by Western blotting from whole cell extracts. The band corresponding to hyperphosphorylated forms of pRb is indicated (P).

 

Figure 3
View larger version (23K):
[in this window]
[in a new window]
  		Fig. 3. Induction of DNA Synthesis by TSH and Forskolin Despite Repression of Cyclin D1

Human thyrocytes from culture no. 2 were stimulated or not (C) for 48 h with 10% serum (S), forskolin (F), or TSH (T). BrdU was present for the last 24 h and the percentage of nuclei having incorporated BrdU was determined. In the same primary culture, cells were stimulated or not (C) for 8, 16, or 26 h with serum, forskolin, or TSH. Cyclin D1, cyclin D3, and CDK4 were detected by Western blotting from whole cell extracts.

 
The observations described above are reminiscent of the situation found in dog thyroid primary cultures where D-type cyclins are not induced during the cAMP-dependent mitogenic stimulation, whereas cyclin D1 is induced by growth factors (43). Also in canine thyrocytes, p21 expression is increased by EGF + serum, whereas TSH up-regulates p27 (44, 51).

Comparison with the Regulation of Cell Cycle Regulatory Proteins in FRTL-5 Cells
The regulation by TSH of the expression of D-type cyclins and p21 and p27 is reported to be very different in rat FRTL-5 cells (21). We have thus analyzed the accumulation of these proteins and their association with CDK4 in response to insulin and TSH in FRTL-5 cells that have been made quiescent either by deprivation of serum, TSH and insulin (Fig. 4BGo), or deprivation of serum and TSH while maintaining insulin (Fig. 4CGo). The latter condition more closely corresponded to the treatment of human thyrocytes by TSH in the presence of insulin, as performed in the present study. In these FRTL-5 cells, the induction of DNA synthesis absolutely depended on the combined stimulation by both TSH and insulin (Fig. 4AGo), in accordance with the initial characterization (55). Similarly, CDK4 activity, as reflected by the CDK4-specific Thr826-phosphorylation of pRb (56), was strongly stimulated by TSH in the presence of insulin, but not in response to insulin or TSH used alone (Fig. 4Go, B and C). However, insulin alone moderately induced cyclin D1 and cyclin D3 and their complex with CDK4, and it decreased p27 levels (Fig. 4BGo). TSH alone was almost inactive on these proteins (but it up-regulated p21) (Fig. 4BGo). In the presence of insulin, TSH strongly increased the accumulation of cyclin D1, cyclin D3, and p21, which associated with CDK4, and it further reduced p27 levels (Fig. 4CGo). These results are consistent with previous reports (30, 31, 32) and sharply contrast with the opposite regulation of these proteins observed in human thyrocytes (Figs. 2Go and 3Go).

Cyclin D3 Supports the TSH-Stimulated pRb-Kinase Activity of CDK4, Whereas Cyclin D1 Supports the Activity of CDK4 Stimulated by EGF + Serum in Human Thyrocytes
In human thyrocytes, none of the modulations of the levels of investigated cell cycle regulatory proteins could explain the stimulatory effects of TSH on pRb phosphorylation and DNA synthesis. We have thus compared, from human thyrocytes stimulated by TSH or EGF + serum, the composition and the pRb-kinase activity of CDK4 complexes coimmunoprecipitated using cyclin D1, cyclin D3, p21, and p27 antibodies. Sufficient amounts of cells for such an analysis were only obtained in four different occasions during the last 3 yr, and the detection of the pRb-kinase activity had required an improvement (51) of the classical method based on 32P labeling of pRb detected by autoradiography (57). Cyclin D1/D3-CDK4 complexes were also investigated in two other primary cultures, with similar results (not shown), but their activity was not assessed. Because of the variations between individual primary cultures, these four different primary cultures have been compared in Fig. 5Go, A–D. In the cultures of Fig. 5Go, A (culture no. 1 in Fig. 1Go also illustrated in Fig. 2Go) and B (culture no. 3 in Fig. 1Go), TSH was the most potent mitogen. TSH was a very weak DNA synthesis inducer in Fig. 5CGo (culture no. 5 in Fig. 1Go) where an appreciable mitogenic response to EGF + serum was observed, and in Fig. 5DGo (culture no. 4 in Fig. 1Go) DNA synthesis was more potently stimulated by EGF + serum over an abnormally high rate of DNA synthesis in nonstimulated control cells.


Figure 5
View larger version (32K):
[in this window]
[in a new window]
  		Fig. 5. Assembly and pRb-Kinase Activity of CDK4 Complexes in Four Independent Primary Cultures of Human Thyrocytes

Cells from culture no. 1 (A), no. 3 (B), no. 5 (C), and no. 4 (D) were stimulated or not (C) with TSH (T) or EGF + serum (ES). In culture no. 4 (D), cells were also treated by TSH in the presence of TGFß (Tß). BrdU incorporation was determined from cells stimulated for 48 h with BrdU for the last 24 h (upper panels in A–D). Extracts from cells stimulated for 26 h were immunoprecipitated (IP) with anti-cyclin D1 (D1), anti-cyclin D3 (D3), anti-p21 or anti-p27 antibodies, assayed for pRb-kinase activity, separated by SDS-PAGE and immunoblotted. Cyclin D1 (cyc D1), cyclin D3 (cyc D3), p21, CDK4 and the in vitro phosphorylation at Ser780 of the pRb fragment (P-Rb780) were detected using specific antibodies. Immunodetections were quantitated by scanning densitometry (arbitrary optical density units), as indicated. Arrows point to the different migration profile of the in vitro phosphorylated pRb fragment.

 
In these four primary cultures (Fig. 5Go, A–D), complexes containing CDK4 and cyclin D3 or to a lesser extent cyclin D1, and p21 or p27 (IP p27 in Fig. 5Go, A and B) were already formed in control nonstimulated cells. A significant basal pRb-kinase activity was only detected in cyclin D3 coimmunoprecipitates (Fig. 5Go). In response to EGF + serum, the levels of cyclin D1 and p21 were increased at various levels (geometric mean of the stimulation ratios in the four experiments: cyclin D1, 1.9; p21, 1.4). The associations of CDK4 and p21 with cyclin D1, and cyclin D1 with p21, were more increased than the levels of these proteins in the four cultures (average stimulation ratios were: CDK4 in IP cyclin D1, 3.3; p21 in IP cyclin D1, 2.8; cyclin D1 in IP p21, 2.4), indicating that EGF + serum also stimulated the assembly of cyclin D1-CDK4-p21 complexes (Fig. 5Go, A–D). Both cyclin D1 and p21 supported the pRb-kinase activity stimulated by EGF + serum (Fig. 5Go, A–D). Interestingly, this activity was even more potently stimulated than the formation of the cyclin D1-CDK4-p21 complexes (average stimulation ratios were: 8.1 in IP cyclin D1 and 12.9 in IP p21). By contrast, EGF + serum did not increase the binding of CDK4 and p21 to cyclin D3, nor did it significantly stimulate the pRb-kinase activity associated with cyclin D3 (average stimulation ratio: 1.1) (Fig. 5Go, A–D).

On the other hand, TSH reduced or did not increase the levels of cyclin D1 and p21, and it did not at all stimulate the pRb-kinase activity associated with cyclin D1 (average stimulation ratio: 0.95) (Fig. 5Go, A–D), at variance with its effects in FRTL-5 cells (Fig. 4CGo). By contrast, TSH increased the pRb-kinase activity associated with cyclin D3 (Fig. 5Go, A–D). In the four primary cultures, this stimulated activity correlated with the amplitude of the DNA synthesis response to TSH (stimulatory factors ranged from 2.9 in the experiment of Fig. 5BGo in which DNA synthesis was the most potently induced by TSH, to 1.1 in the culture of Fig. 5CGo in which TSH only marginally increased DNA synthesis). Only in the cultures of Fig. 5Go, A and D, the presence of cyclin D3-CDK4 complexes was somewhat elevated by TSH, but this was accounted for by an increase of immunoprecipitable cyclin D3 in these two experiments, indicating that TSH did not stimulate the assembly of these complexes. TSH also increased the pRb-kinase activity associated with p27 (Fig. 5Go, A and B) or p21 (Fig. 5Go, A–C, but not in Fig. 5DGo, in which TSH more strongly reduced p21 levels). Nevertheless, these activities were weak compared with the strong pRb-kinase activity coimmunoprecipitated by the cyclin D3 antibody, or to the p21-associated pRb-kinase activity strongly stimulated by EGF + serum (Fig. 5Go, A–D).

In one of these primary cultures, we have also tested the effect of TGFß1 in the presence of TSH on DNA synthesis and pRb-kinase activity (Fig. 5DGo). TGFß inhibits cell proliferation in the different thyroid cell culture systems (48, 58). In normal human thyroid cells, TGFß prevents both the DNA synthesis and the differentiation expression induced by TSH and cAMP (59). In the presence of TSH, TGFß decreased DNA synthesis, as well as the pRb-kinase activity associated with cyclin D3, below their basal levels (Fig. 5DGo). TGFß also reverted the weak stimulatory effect of TSH on cyclin D3 accumulation and the reduction by TSH of the accumulation of cyclin D1 and p21 (Fig. 5DGo).

Different pRb-Kinase Activities Are Associated with Cyclin D1 and Cyclin D3
We have recently demonstrated that cyclin D1 and cyclin D3 drive partly different phosphorylations of pRb by CDK4 in dog and monkey thyrocytes and human fibroblasts (51). Whereas both cyclin D1 and cyclin D3 support the phosphorylation of pRb at Ser780, only cyclin D1-CDK4 efficiently phosphorylates pRb at Ser807/811. The phosphorylations on Ser807/811 and Thr826 but not on Ser780 or Ser795 contribute to generate an upward electrophoretic shift of the pRb fragment in pRb-kinase activity assays (51). Similarly, in the assays of pRb-kinase activity detected using the Ser780-phosphospecific pRb antibody, as shown in Fig. 5Go, A–D, the migration of the pRb fragment phosphorylated in vitro was different in immunoprecipitations of cyclin D1 (one upward-shifted band) or cyclin D3 (one more rapidly migrating doublet or large band). Thus, whereas the phosphorylation of pRb at Ser780 was stimulated by both TSH (supported by cyclin D3) and EGF + serum (supported by cyclin D1), additional phosphorylation(s) of pRb were induced only by EGF + serum through the activation of cyclin D1-CDK4 (Fig. 5Go). In immunoprecipitations of p21, which contain both cyclin D1 and cyclin D3, the signature of the pRb-kinase activity of cyclin D1-CDK4 activated by EGF + serum was prevalent, but a weaker contribution of the pRb-kinase activity of cyclin D3-CDK4 was also obvious (Fig. 5Go).

Weak Association of CDK4 to p16 in Human Thyrocytes Compared with Human Fibroblasts
High amounts of cyclin D3-CDK4 complexes were unexpectedly present in nonstimulated human thyrocytes (Fig. 5Go). p16, a CDK inhibitor of the CDK4 inhibitory (INK4) family (60), is an important regulator of CDK4 activity. It binds CDK4/6 and impairs their association with D-type cyclins (61). By Western blotting, we failed to detect the p16 protein in different primary cultures of human thyrocytes, although it was present in extracts of human normal diploid fibroblasts (IMR-90) put as a positive control on a same gel (not shown). We have thus compared the amount of CDK4 that could be coimmunoprecipitated by p16 and cyclin D antibodies from human thyrocytes and human fibroblasts (IMR-90) (Fig. 6Go). In human fibroblasts, CDK4 was mostly associated with p16 both in quiescent and in serum-stimulated cells. Even after stimulation, only a minor fraction of CDK4 was associated with cyclin D1 [and even less with cyclin D3 (51)]. By contrast, in human thyrocytes a larger amount of CDK4 was bound to cyclin D3 than to p16 even in nonstimulated cells (Fig. 6Go). This indicates that an insufficient concentration and CDK4-binding capacity of p16 in human thyrocytes could facilitate the constitutive formation of cyclin D3-CDK4 complexes.


Figure 6
View larger version (21K):
[in this window]
[in a new window]
  		Fig. 6. Comparison of CDK4 Association to p16 in Human Fibroblasts (IMR-90) Stimulated or Not (C) for 16 h with 20% Serum (S), and Human Thyrocytes Stimulated or Not (C) for 26 h with TSH (T) or EGF + Serum (ES)

Cell extracts were immunoprecipitated (IP) with anti-p16, or anti-cyclin D1 (D1) or anti-cyclin D3 (D3) antibodies, separated by SDS-PAGE and immunoblotted. CDK4 was detected using a specific antibody.

 
Activating Thr172-Phosphorylation of CDK4 Is Stimulated by TSH or EGF + Serum, and Inhibited by TGFß
The phosphorylation of CDK4 does not affect its electrophoretic migration in sodium dodecyl sulfate (SDS)-polyacrylamide gels. Previously, using two-dimensional (2D) gel electrophoresis, we have separated different forms of CDK4. We have identified the form 3 of CDK4 as the Thr172-phosphorylated CDK4 using 32P-phosphate incorporation, a new phospho-specific antibody, and 2D gel analysis of T172A-mutated CDK4 (49, 62). In two experiments, sufficient amounts of cells allowed to evaluate the proportion of the Thr172-phosphorylated form 3 of CDK4 (culture no. 1 also illustrated in Figs. 2Go and 5AGo; culture no. 4 also illustrated in Fig. 5DGo) (Fig. 7Go, A and B). In both experiments, the relative abundance of CDK4 form 3 perfectly correlated with the very different rates of DNA synthesis in nonstimulated cells, and with the respective amplitudes of the stimulation of DNA synthesis by TSH and EGF + serum (Fig. 7Go, A–C). In the presence of TSH, TGFß reduced the presence of this phosphorylated form 3 below the level observed in nonstimulated cells (Fig. 7BGo). The regulation of the Thr172 activating phosphorylation of CDK4 by mitogenic and antimitogenic signals could thus explain the important modulations of pRb-kinase activities of CDK4 that exceeded more modest variations of the abundance of cyclin D-CDK4 complexes.


Figure 7
View larger version (20K):
[in this window]
[in a new window]
  		Fig. 7. Activating Phosphorylation of CDK4 in Two Independent Primary Cultures of Human Thyrocytes

Cells from culture no. 1 (A) (also illustrated in Figs. 2Go and 4AGo), and from culture no. 4 (B) (also illustrated in Fig. 4DGo) were stimulated or not (C) for 26 h with TSH (T), EGF + serum (ES), or TSH + TGFß (Tß). Extracts were immunoprecipitated with an anti-CDK4 antibody and the different forms of CDK4 were separated by 2D gel electrophoresis. CDK4 was detected by Western blotting using a specific antibody. Arrows indicate the Thr172-phosphorylated form of CDK4. Values indicate the proportion of this phosphorylated form relative to total CDK4 (% P-CDK4), as quantitated by scanning densitometry of CDK4 immunodetections. In both primary cultures (C), DNA synthesis responses perfectly correlated with the proportion of CDK4 phosphorylated forms.

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
MATERIALS AND METHODS
 REFERENCES
 
Primary cultures of normal human thyrocytes are a rather awkward experimental system, marred by scarce availability and generally insufficient amounts of the initial tissue left by the pathologist, the slow proliferation rates of the cells even after strong stimulations, and their very limited division capacity. Yet, the present primary culture system is so far the only in vitro model allowing the investigation, in normal (nontumoral) human cells, of the mitogenic action exerted by TSH and mediated by cAMP elevations (3, 21). The present study was also complicated by the variability of mitogenic responses observed in primary cultures of thyroid from different patients. Several cultures responded poorly to the mitogenic actions of TSH and forskolin (yet displayed the expected induction of differentiation genes) or EGF + serum. No correlation was observed between the responsiveness to either mitogenic treatments, arguing for their independent mechanisms. The reasons for this unexpected variability are unclear. The responsiveness of the cultivated cell might depend on its in vivo history (e.g. age of patient; normal tissue surrounding a tumor that could produce growth factors; Graves’ disease tissue). For instance, the responsiveness of dog thyrocytes to cAMP as a mitogenic stimulus is specifically extinct after their proliferation stimulated by growth factors (63). Alternatively, transcriptome analyses of human thyrocytes stimulated by TSH or EGF + serum have suggested the induction of multiple growth limiting mechanisms, the impact of which could vary in different cultures (Ref. 64 ; and Hebrant, A., V. Detours, and C. Maenhaut, unpublished observations).

At Variance with the Case of FRTL-5 Cells, TSH Does Not Induce Cyclin D1 Nor Repress p27 in Human Thyrocytes
Three kinds of (nonexclusive) mechanisms may mediate a positive action of cAMP on cell proliferation (65): 1) cAMP may act indirectly, for instance by up-regulating an autocrine growth factor or a growth factor receptor (see Ref. 12 for a review in thyroid); 2) as now most frequently considered, cAMP may also act by positively intervening at different steps of the signaling cascades of growth factors (66, 67, 68). This is the situation reported from FRTL-5 and WRT cells, although mechanisms are partly divergent in both rat cell lines and even in FRTL-5 cells used by different laboratories (21); 3) as demonstrated in canine thyroid primary cultures, cAMP could operate its own distinct mitogenic cascade and cell cycle regulation (21). The two first types of mechanisms are expected to converge with pathways generally operated by growth factors, and thus should culminate in the accumulation of cyclin D1, the transient induction of p21 [which stabilizes cyclin D1-CDK4 complexes in the nucleus (69, 70)], and the down-regulation of p27. These different effects are indeed observed in canine thyrocytes stimulated by growth factors (43, 51), but in response to TSH or forskolin in FRTL-5 cells (Fig. 4Go) (30, 31, 32). Activated (mutated) Ras has been previously shown in human thyrocytes to induce cyclin D1 and p21, but not cyclin D3, and to down-regulate p27 (71). In the present study, similar but weaker responses are generated by the stimulation of human thyrocytes by growth factors, but they are not at all observed in response to TSH. Indeed, TSH did not induce—or even markedly reduced in a majority of primary cultures—the accumulation of cyclin D1 and p21, and it rather up-regulated p27, in sharp contrast with the situation observed in FRTL-5 cells. This is consistent with the lack of MAPK activation in human thyroid primary cultures (39), and the repression by TSH and forskolin of c-Jun and c-Fos (72), which are important transactivators of the cyclin D1 gene. These major differences indicate that the poorly known signaling cascades of TSH-dependent mitogenesis must be quite distinct from growth factor pathways in human thyrocytes, as in canine thyroid primary cultures, but at variance with rat thyroid cell lines (21).

Whereas Growth Factors Selectively Activate Cyclin D1-CDK4, TSH Selectively Activates Cyclin D3-CDK4 in Human Thyrocytes
EGF + serum did not only increase levels of cyclin D1 and (to a lesser extent) p21; it also stimulated the formation of cyclin D1-CDK4-p21 complexes and even more potently their pRb-kinase activity. By contrast, TSH stimulated the activity of cyclin D3-CDK4, but little or not the presence of this abundant complex. Intriguingly, EGF + serum did not activate the abundant cyclin D3-CDK4 complexes, whereas TSH did not enhance the assembly and activity of cyclin D1-CDK4. The distinct mitogenic pathways of EGF + serum or TSH thus appear to be channeled to use only cyclin D1 or cyclin D3. Mechanisms responsible for the regulation of the assembly and activation of D-type cyclin-CDK complexes remain largely enigmatic (62). In dog thyrocytes, we have recently demonstrated that cAMP stimulates, whereas TGFß inhibits, the phosphorylation and activity of CDK4 in cyclin D3-CDK4 complexes (49, 50, 62), at variance with the generally hold concept that the CDK activating kinase (CAK, consisting of cyclin H-CDK7-Mat1) is constitutively active during cell cycle progression and mitogenic stimulation (73, 74). In the present study, levels of CDK7 and cyclin H were also not increased by EGF + serum or TSH in human thyrocytes, but the stimulation of DNA synthesis by EGF + serum and TSH and its inhibition by TGFß perfectly correlated with the activating phosphorylation of CDK4 (Fig. 7CGo). In human thyrocytes, as we recently observed it in other cells (62), the activating phosphorylation of CDK4, rather than the induction of D-type cyclins, could thus be the main regulated step that determines the activity of cyclin D1/D3-CDK4 complexes and the entry of cells into S phase. Very unexpectedly, our observations imply that the mechanisms of the assembly and activation of CDK4 complexes could be separate in the distinct mitogenic stimulations and specifically addressed to a particular D-type cyclin.

Possible Implications of the Specific Involvement of Cyclin D3 in TSH-Dependent Mitogenesis
The specific involvement of cyclin D3 in the mitogenic stimulation of human thyrocytes by TSH is congruent with our initial findings from dog thyroid primary cultures. In this system, cyclin D3 is specifically required for the mitogenic stimulation by TSH and cAMP, but the neutralization of cyclin D3 does not impair the response to various growth factors that induce cyclin D1 and more weakly cyclin D2 (43). A more partial requirement for cyclin D3 has been confirmed in the TSH-dependent proliferation of PC Cl3 cells, likely because TSH induces both cyclin D3 and cyclin D1 in this cell line (Ref. 75 ; and our own unpublished observations). D-type cyclins have been suggested to play distinct roles in cell differentiation (76). In several adult mammalian tissues, cyclin D1 localizes to proliferative layers, whereas cyclin D3 is often highly expressed in the adjacent compartment where differentiation takes place (76, 77, 78). Cyclin D3 replaces cyclin D1 during myogenesis (79), adipogenesis (80), differentiation of the leukemia cell line HL-60 (76), or at the pre-TCR developmental stage during T lymphocyte maturation (81). It might thus be more specifically restricted to specialized cell cycles associated with maintenance or induction of differentiation. Using human thyrocytes, we confirm here our recent demonstration (51) that the differential utilization of cyclin D1 or D3 affects the site specificity of the pRb-kinase activity of CDK4. Besides Ser780, cyclin D1 but not cyclin D3 supported in vitro other phosphorylation(s) of pRb that affect its electrophoretic mobility. We did not identify these additional phosphorylations in the present study, but in dog and monkey thyrocytes, as in human fibroblasts and Chinese hamster ovary cells transfected with CDK4 and cyclin D1 or cyclin D3, cyclin D1-CDK4 more efficiently drives the phosphorylation of pRb at Ser807/811 and Thr826 (51). In general, the role of individual phosphorylations of pRb remains poorly defined or debated. Nevertheless, it is admitted that different functions of pRb could be regulated by phosphorylation on distinct residues (Ref. 82 ; and see a more complete discussion in Ref. 51). In addition to inhibiting E2F-dependent gene transcription related to cell cycle progression, pRb plays positive roles in the induction of tissue-specific gene expression by directly interacting with a variety of transcription factors (83), including Pax 8 in thyroid cells (84). In the present primary culture conditions of human thyrocytes, TSH induces or maintains differentiation expression, whereas EGF strongly inhibits it (3, 53, 54, 85). Whether the selective utilization of cyclin D3 in the TSH cascade, associated with a more restricted pRb-kinase activity, could allow the preservation of some differentiation-related functions of pRb thus remains to be examined.

Preactivation of Primary Cultured Human Thyrocytes
As compared with canine thyrocytes in similar primary culture conditions, human thyrocytes displayed higher basal rates of DNA synthesis and appreciable levels of cyclin D1 and cyclin D3 in the present control conditions (with insulin), and even in the absence of insulin (our unpublished data). In nonstimulated human thyrocytes, CDK4 was mostly nuclear (immunofluorescence microscopy observations, not shown), and cyclin D3-CDK4 complexes were already assembled, which contrasts with the stimulation by TSH of both the nuclear import of CDK4 and its association with cyclin D3 in dog thyrocytes (43). In the present primary culture conditions, control human thyrocytes thus seemed to be primed for cell cycle progression. Several nonexclusive hypotheses could explain this preactivation: 1) contrary to canine thyrocytes which rapidly spread out when seeded in serum-free conditions, and thus likely secreted their own extracellular matrix proteins, human thyroid follicles require their seeding in the presence of 1% serum to allow the spreading of cells as a monolayer, and cells are then cultured without serum (3). Serum contains various extracellular matrix proteins, including fibronectin; moreover, human thyrocytes produce fibronectin only when stimulated by serum (86). In normal human thyroid cells, spreading of the cells on fibronectin and the resulting activation of integrins can stimulate MAPK and PI-3kinase pathways and suffice to support a weak proliferation (87); cultured human thyrocytes are reported to express a variety of growth factors, proteases that can activate growth factor precursors, and receptors for some of these growth factors, indicating that they could be subjected to several autocrine stimulations (88); and as also observed by Jones et al. (71), the INK4 protein p16 was weakly expressed in normal human thyrocytes and its levels were insufficient to bind most of CDK4. Although we could not investigate the other related INK4 proteins (p15, p18, p19), this is expected to facilitate the formation of cyclin D3-CDK4 complexes as we also observe it for the constitutive assembly of these complexes in p16-defective human T98G glioma cells (62).

Possible Relevance to Hyperproliferative Thyroid Diseases
The selective activation of CDK4 by cyclin D1 or cyclin D3, leading to partly different pRb-kinase activities in the distinct mitogenic stimulations by growth factors or TSH, could be relevant to human hyperproliferative thyroid diseases. Cyclin D1 is overexpressed in papillary carcinomas (89) caused by overactivation of growth factor signaling cascades by B-Raf mutation or Ret/PTC activation. It was proposed to predict the metastatic behavior of papillary microcarcinomas (90). On the other hand, cyclin D3 is prominent in hyperfunctional goitre caused in transgenic mice by thyroid-targeted overactivation of the cAMP pathway (91). It is also elevated during TSH-dependent goitrogenesis in rats, and overexpressed in human thyroid follicular adenomas (75). Further studies should investigate whether cyclin D3 could be specifically involved in thyroid proliferative diseases that maintain tissue follicular architecture and thyroid differentiated functions, by contrast with papillary carcinomas associated with overactivation of ERK/MAPK pathway and cyclin D1 accumulation.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
REFERENCES
 
Cell Cultures
For primary cultures of human thyrocytes, thyroid tissue samples were obtained from 14 adult patients (aged 19–53 yr) undergoing surgery for a hypofunctioning nodule, following a protocol approved by the Ethics Committee of the Medical School of the Free University of Brussels. Only histologically normal perinodular tissue was used. In five other cases, thyroid tissue was obtained from Graves’ disease patients. The thyrocytes were cultured as described from follicles released by collagenase/dispase digestion of minced tissue (3). Follicles were seeded (at a density corresponding to about 2 x 104 cells/cm2) in the control medium, i.e. DMEM + Ham’s F12 + MCDB104 medium (2:1:1 by vol) supplemented with bovine insulin (5 µg/ml; Sigma, St. Louis, MO) (92), human transferrin (1.25 µg/ml; Sigma), ascorbic acid (40 µg/ml), and antibiotics. One percent FCS was also added to ensure the spreading of the cells as a monolayer (3). After 1 d, medium was renewed and the culture was pursued in completely serum-free conditions with medium renewal every 2 d. In all the experiments, cells were stimulated at d 5 or 6 with either bovine TSH (0.3 mU/ml; Sigma), the general adenylyl cyclase activator forskolin (10–5 M; Calbiochem, La Jolla, CA), 10% FCS or the combination of murine EGF (25 ng/ml; Sigma) and 10% FCS. In one experiment, cells were treated in the presence of BSA (500 µg/ml, crystallized; ICN Biomedicals, Irvine, CA) with TSH in the absence or presence of recombinant human TGFß1 (2 ng/ml; R&D Systems, Minneapolis, MN).

FRTL-5 cells were initially received in 1994 from Dr. R. Di Lauro (Stazione Zoologica Anton Dohrn, Napoli, Italy) and kept frozen after a very few number of passages since this date. They were cultured as described (93) in Coon’s modified Ham F12 medium (Invitrogen, Carlsbad, CA) supplemented with 5% calf serum, transferrin (5 µg/ml), insulin (5 µg/ml) and bovine TSH (1 mU/ml) (3H-medium). Cells were made quiescent by switching to medium containing transferrin, BSA (500 µg/ml), with or without insulin (5 µg/ml) for 3 d, and then stimulated for 20 h (analysis of cell cycle regulatory proteins) or 48 h (evaluation of DNA synthesis) using insulin (5 µg/ml), TSH (1 mU/ml) or both.

Human diploid fibroblasts (IMR-90) (American Type Culture Collection, Manassas, VA) were cultured as described (94). After reaching semiconfluency, they were synchronized by starvation in 0.2% FCS for 3 d. Quiescent cells were then growth stimulated by addition of FCS (20%).

DNA Synthesis
Cells in 3-cm Petri dishes were stimulated for 48 h and BrdU was added for the last 24 h. The incorporation of BrdU was detected by immunofluorescence, and BrdU-labeled nuclei (1000/dish) were counted as described (63).

Western Blotting Detections of Proteins
Ten to 20 µg of total proteins were separated by PAGE and the proteins of interest were immunodetected after Western blotting as previously described (49). The following antibodies were used: C-15 polyclonal anti-pRb (Santa Cruz Biotechnology, Santa Cruz, CA), DCS-6 monoclonal anticyclin D1 (from J. Bartek, Danish Cancer Society, Copenhagen, Denmark; or NeoMarkers, Fremont, CA), DCS-22 monoclonal anti-cyclin D3 (NeoMarkers), DCS-3 monoclonal anti-cyclin D2 (from J. Bartek), HE-12 monoclonal anti-cyclin E (from J. Bartek), C-18 polyclonal anti-cyclin H (Santa Cruz), DCS-156 monoclonal anti-CDK4 (Cell Signaling Technology, Beverly, MA) or C-22 polyclonal anti-CDK4 (Santa Cruz), M2 polyclonal anti-CDK2 (Santa Cruz), DCS-83 monoclonal anti-CDK6 (NeoMarkers), DCS-61 monoclonal anti-p21 (from J. Bartek), C-15 polyclonal anti-p27 (Santa Cruz), DCS-230 monoclonal anti-p57 (from J. Bartek) and DCS-120 monoclonal anti-CDC25A (NeoMarkers).

Immunoprecipitation (49)
For analyses of protein complexes and their pRb kinase activity, subconfluent cultures of thyrocytes or IMR-90 fibroblasts in 9-cm Petri dishes that contain the same number of cells were lysed on ice in 1 ml Nonidet P-40 (NP-40) lysis buffer containing 150 mM NaCl, 50 mM Tris-HCl (pH 7.5), 0.5% NP-40, 50 mM NaF, 1 mM sodium orthovanadate, 10 mM dithiothreitol, protease inhibitors, and 10% glycerol. The homogenized (glass/glass) cellular lysate was sonicated twice, precleared with protein A Sepharose (Amersham Biosciences, Uppsala, Sweden) and then incubated at 4 C for 3 h with protein A-Sepharose that had been preincubated overnight with 2 µg of antibody [DCS-11 monoclonal anti-cyclin D1 antibody (Biosource International, Camarillo, CA), DCS-28 monoclonal anti-cyclin D3 antibody (NeoMarkers), C-19 polyclonal anti-p21 antibody (Santa Cruz), a mixture of the K25020 monoclonal anti-p27 antibody from Transduction Laboratories (BD Biosciences, Erembodegem, Belgium) and the C-15 polyclonal anti-p27 antibody from Santa Cruz, or DCS-50 monoclonal anti-p16 antibody (NeoMarkers)].

Alternatively, for analyses of the 2D gel electrophoresis pattern of CDK4, cultures of thyrocytes in 9-cm Petri dishes were washed with PBS, scraped in 200 µl of denaturing lysis buffer [50 mM Tris-HCl (pH 7.5), 0.6% SDS, 10 mM dithiothreitol, 50 mM NaF, 100 µM vanadate, and protease inhibitors], boiled for 5 min, and frozen. Cell lysates containing 150–300 µg of proteins were cleared by centrifugation and diluted by adding 5 vol of RIPA buffer without SDS [10 mM Tris-HCl (pH 7.5), 1% Na deoxycholate, 1% NP-40, 150 mM NaCl, 50 mM NaF, 100 µM vanadate and protease inhibitors]. They were then subjected to immunoprecipitation as above with the C-22 polyclonal anti-CDK4 antibody.

pRb-Kinase Assay (51)
Washed complexes were resuspended in 40 µl of kinase reaction buffer containing 2 mM ATP, 0.3 µg of a 56-kDa fragment (amino acids 379–928) of pRb (QED Bioscience, San Diego, CA), 10 mM ß-glycerophosphate, 0.1 mM orthovanadate, 1 mM NaF, 60 µg/ml Pefabloc, and 1 µg/ml leupeptine, and incubated for 30 min at 30 C with occasional gentle agitation. Reactions were stopped by adding 60 µl of twice-concentrated Laemmli buffer and boiling for 5 min. Proteins were resolved by SDS-PAGE and transferred on polyvinylidene difluoride membranes. The phosphorylation on Ser780 of the pRb fragment was detected using a phospho-specific-pRb antibody from Cell Signaling Technology. Membranes were then immediately reprobed using the DCS-156 CDK4 antibody, which allowed a direct internal comparison of the pRb-kinase activity with the amount of coimmunoprecitated CDK4. D-type cyclins and p21 were detected on separate membranes from the same pRb-kinase immunoprecipitation sample, using the antibodies previously described for the Western blotting.

2D Gel Electrophoresis
Immunoprecipitated proteins were denatured in a buffer containing 7 M urea and 2 M thiourea. Proteins were separated by isoelectric focusing as described (49, 94) on immobilized linear pH gradient (pH 3–10) IEF strips (Amersham Biosciences). After loading onto SDS-polyacrylamide slab gels (12.5%) for separation according to molecular mass, and transfer on polyvinylidene difluoride membranes, CDK4 was detected with the C-22 polyclonal antibody (Santa Cruz).

Enhanced chemiluminescence detections of Western blots were quantitated using a GS-800 densitometer and the Quantity One software (Bio-Rad Laboratories, Hercules CA).


   ACKNOWLEDGMENTS
 
We thank A. Van Keymeulen for her contribution in the initial part of this study, K. Coulonval and L. Bockstaele for helpful discussion, Drs. M. Cappello and P. Rocmans (Hôpital Erasme, Brussels, Belgium) for providing thyroid tissue samples, and Dr. J. Bartek (Danish Cancer Society, Denmark) for kindly providing several antibodies.


   FOOTNOTES
 
This work was supported by grants from the Fonds National de la Recherche Scientifique (FNRS), Fonds de la Recherche Scientifique Médicale (FRSM), Opération Télévie, Fedération Belge Contre le Cancer, and Actions de Recherche Concertées (ARC) de la Communauté Française de Belgique. S.P. was a fellow of the Fonds pour la formation à la Recherche dans l’Industrie et l’Agriculture (FRIA). P.P.R. is a Research Associate of the FNRS.

Disclosure statement: The authors have nothing to declare.

First Published Online August 17, 2006

Abbreviations: BrdU, Bromodeoxyuridine; CAK, CDK activating kinase; 2D, two-dimensional; EGF, epidermal growth factor; FCS, fetal calf serum; INK4, CDK4 inhibitory; NP-40, Nonidet P-40; PI3-kinase, phosphatidylinositol 3-kinase; SDS, sodium dodecyl sulfate; TSAbs, TSH receptor-stimulating autoantibodies.

Received for publication December 15, 2005. Accepted for publication July 7, 2006.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 

  1. Dumont JE, Maenhaut C, Christophe D, Vassart G, Roger PP 2005 Thyroid regulatory factors. In: DeGroot LJ, Jameson JL, eds. Endocrinology. Philadelphia: Saunders; 1837–1860
  2. Roger PP, Servais P, Dumont JE 1983 Stimulation by thyrotropin and cyclic AMP of the proliferation of quiescent canine thyroid cells cultured in a defined medium containing insulin. FEBS Lett 157:323–329[CrossRef][Medline]
  3. Roger P, Taton M, Van Sande J, Dumont JE 1988 Mitogenic effects of thyrotropin and adenosine 3',5'-monophosphate in differentiated normal human thyroid cells in vitro. J Clin Endocrinol Metab 66:1158–1165[Abstract]
  4. Huber GK, Safirstein R, Neufeld D, Davies TF 1991 Thyrotropin receptor autoantibodies induce human thyroid cell growth and c-fos activation. J Clin Endocrinol Metab 72:1142–1147[Abstract]
  5. Parma J, Duprez L, Van Sande J, Cochaux P, Gervy C, Mockel J, Dumont J, Vassart G 1993 Somatic mutations in the thyrotropin receptor gene cause hyperfunctioning thyroid adenomas. Nature 365:649–651[CrossRef][Medline]
  6. Duprez L, Parma J, Van Sande J, Allgeier A, Leclere J, Schvartz C, Delisle MJ, Decoulx M, Orgiazzi J, Dumont J 1994 Germline mutations in the thyrotropin receptor gene cause non-autoimmune autosomal dominant hyperthyroidism. Nat Genet 7:396–401[CrossRef][Medline]
  7. Williams ED 1994 Thyroid tumorigenesis. Horm Res 42:31–34[Medline]
  8. Mazzaferri EL 1990 Thyroid cancer and Graves’ disease. J Clin Endocrinol Metab 70:826–829[Medline]
  9. Pohl V, Roger PP, Christophe D, Pattyn G, Vassart G, Dumont JE 1990 Differentiation expression during proliferative activity induced through different pathways: in situ hybridization study of thyroglobulin gene expression in thyroid epithelial cells. J Cell Biol 111:663–672[Abstract/Free Full Text]
  10. Roger PP, Dumont JE 1984 Factors controlling proliferation and differentiation of canine thyroid cells cultured in reduced serum conditions: effects of thyrotropin, cyclic AMP and growth factors. Mol Cell Endocrinol 36:79–93[CrossRef][Medline]
  11. Paschke R, Eck T, Herfurth J, Usadel KH 1995 Stimulation of proliferation and inhibition of function of xenotransplanted human thyroid tissue by epidermal growth factor. J Endocrinol Invest 18:359–363[Medline]
  12. Bidey SP, Hill DJ, Eggo MC 1999 Growth factors and goitrogenesis. J Endocrinol 160:321–332[Abstract]
  13. Melillo RM, Castellone MD, Guarino V, De F, V, Cirafici AM, Salvatore G, Caiazzo F, Basolo F, Giannini R, Kruhoffer M, Orntoft T, Fusco A, Santoro M 2005 The RET/PTC-RAS-BRAF linear signaling cascade mediates the motile and mitogenic phenotype of thyroid cancer cells. J Clin Invest 115:1068–1081[CrossRef][Medline]
  14. Mitsutake N, Miyagishi M, Mitsutake S, Akeno N, Mesa JC, Knauf JA, Zhang L, Taira K, Fagin JA 2005 BRAF mediates RET/PTC-induced MAPK activation in thyroid cells: functional support for requirement of the RET/PTC-RAS-BRAF pathway in papillary thyroid carcinogenesis. Endocrinology 147:1014–1019
  15. Greco A, Roccato E, Pierotti MA 2004 TRK oncogenes in papillary thyroid carcinoma. Cancer Treat Res 122:207–219[Medline]
  16. Kimura ET, Nikiforova MN, Zhu Z, Knauf JA, Nikiforov YE, Fagin JA 2003 High prevalence of BRAF mutations in thyroid cancer: genetic evidence for constitutive activation of the RET/PTC-RAS-BRAF signaling pathway in papillary thyroid carcinoma. Cancer Res 63:1454–1457[Abstract/Free Full Text]
  17. Ciampi R, Knauf JA, Kerler R, Gandhi M, Zhu Z, Nikiforova MN, Rabes HM, Fagin JA, Nikiforov YE 2005 Oncogenic AKAP9-BRAF fusion is a novel mechanism of MAPK pathway activation in thyroid cancer. J Clin Invest 115:94–101[CrossRef][Medline]
  18. Trovato M, Villari D, Bartolone L, Spinella S, Simone A, Violi MA, Trimarchi F, Batolo D, Benvenga S 1998 Expression of the hepatocyte growth factor and c-met in normal thyroid, non-neoplastic, and neoplastic nodules. Thyroid 8:125–131[Medline]
  19. Aasland R, Akslen LA, Varhaug JE, Lillehaug JR 1990 Co-expression of the genes encoding transforming growth factor-{alpha} and its receptor in papillary carcinomas of the thyroid. Int J Cancer 46:382–387[Medline]
  20. Medina DL, Santisteban P 2000 Thyrotropin-dependent proliferation of in vitro rat thyroid cell systems. Eur J Endocrinol 143:161–178[Abstract]
  21. Kimura T, Van Keymeulen A, Golstein J, Fusco A, Dumont JE, Roger PP 2001 Regulation of thyroid cell proliferation by thyrotropin and other factors: a critical evaluation of in vitro models. Endocr Rev 22:631–656[Abstract/Free Full Text]
  22. Rivas M, Santisteban P 2003 TSH-activated signaling pathways in thyroid tumorigenesis. Mol Cell Endocrinol 213:31–45[CrossRef][Medline]
  23. Tsygankova OM, Kupperman E, Wen W, Meinkoth JL 2000 Cyclic AMP activates Ras. Oncogene 19:3609–3615[CrossRef][Medline]
  24. Cass LA, Summers SA, Prendergast GV, Backer JM, Birnbaum MJ, Meinkoth JL 1999 Protein kinase A-dependent and -independent signaling pathways contribute to cyclic AMP-stimulated proliferation. Mol Cell Biol 19:5882–5891[Abstract/Free Full Text]
  25. Iacovelli L, Capobianco L, Salvatore L, Sallese M, D’Ancona GM, De Blasi A 2001 Thyrotropin activates mitogen-activated protein kinase pathway in FRTL-5 by a cAMP-dependent protein kinase A-independent mechanism. Mol Pharmacol 60:924–933[Abstract/Free Full Text]
  26. Pomerance M, Abdullah HB, Kamerji S, Correze C, Blondeau JP 2000 Thyroid-stimulating hormone and cyclic AMP activate p38 mitogen-activated protein kinase cascade. Involvement of protein kinase A, Rac 1 and reactive oxygen species. J Biol Chem 275:40539–40546[Abstract/Free Full Text]
  27. Kambe F, Miyazaki T, Seo H 1996 Differential induction of fos and jun family genes by thyrotropin in rat thyroid FRTL-5 cells. Thyroid 6:123–128[Medline]
  28. Tell G, Pines A, Arturi F, Cesaratto L, Adamson E, Puppin C, Presta I, Russo D, Filetti S, Damante G 2004 Control of phosphatase and tensin homolog (PTEN) gene expression in normal and neoplastic thyroid cells. Endocrinology 145:4660–4666[Abstract/Free Full Text]
  29. Tramontano D, Chin WW, Moses AC, Ingbar SH 1986 Thyrotropin and dibutyryl cyclic AMP increase levels of c-myc and c-fos mRNAs in cultured rat thyroid cells. J Biol Chem 261:3919–3922[Abstract/Free Full Text]
  30. Yamamoto K, Hirai A, Ban T, Saito J, Tahara K, Terano T, Tamura Y, Saito Y, Kitagawa M 1996 Thyrotropin induces G1 cyclin expression and accelerates G1 phase after insulin-like growth factor I stimulation in FRTL-5 cells. Endocrinology 137:2036–2042[Abstract]
  31. Medina DL, Toro MJ, Santisteban P 2000 Somatostatin interferes with thyrotropin-induced G1-S transition mediated by cAMP-dependent protein kinase and phosphatidylinositol 3-kinase. Involvement of RhoA and cyclin E x cyclin-dependent kinase 2 complexes. J Biol Chem 275:15549–15556[Abstract/Free Full Text]
  32. Correze C, Blondeau JP, Pomerance M 2005 p38 Mitogen-activated protein kinase contributes to cell cycle regulation by cAMP in FRTL-5 thyroid cells. Eur J Endocrinol 153:123–133[Abstract/Free Full Text]
  33. Sherr CJ, McCormick F 2002 The RB and p53 pathways in cancer. Cancer Cell 2:103–112[CrossRef][Medline]
  34. Nedachi T, Akahori M, Ariga M, Sakamoto H, Suzuki N, Umesaki K, Hakuno F, Takahashi SI 2000 Tyrosine kinase and phosphatidylinositol 3-kinase activation are required for cyclic adenosine 3',5'-monophosphate-dependent potentiation of deoxyribonucleic acid synthesis induced by insulin-like growth factor-I in FRTL-5 cells. Endocrinology 141:2429–2438[Abstract/Free Full Text]
  35. Villone G, De Amicis F, Veneziani BM, Salzano S, Di Carlo A, Tramontano D 1997 Sustained versus transient cyclic AMP intracellular levels: effect on thyrotropin-dependent growth of thyroid cells. Cell Growth Differ 8:1181–1188[Abstract]
  36. Burikhanov R, Coulonval K, Pirson I, Lamy F, Dumont JE, Roger PP 1996 Thyrotropin via cyclic AMP induces insulin receptor expression and insulin co-stimulation of growth and amplifies insulin and insulin-like growth factor signaling pathways in dog thyroid epithelial cells. J Biol Chem 271:29400–29406[Abstract/Free Full Text]
  37. Van Keymeulen A, Roger PP, Dumont JE, Dremier S 2000 TSH and cAMP do not signal mitogenesis through Ras activation. Biochem Biophys Res Commun 273:154–158[CrossRef][Medline]
  38. Coulonval K, Vandeput F, Stein R, Kozma S, Lamy F, Dumont JE 2000 Phosphatidylinositol 3-kinase, protein kinase B and ribosomal S6 kinases in the stimulation of thyroid epithelial cell proliferation by cAMP and growth factors in the presence of insulin. Biochem J 348:351–358
  39. Vandeput F, Perpete S, Coulonval K, Lamy F, Dumont JE 2003 Role of the different mitogen-activated protein kinase subfamilies in the stimulation of dog and human thyroid epithelial cell proliferation by cyclic adenosine 5'-monophosphate and growth factors. Endocrinology 144:1341–1349[Abstract/Free Full Text]
  40. Pirson I, Coulonval K, Lamy F, Dumont JE 1996 c-Myc expression is controlled by the mitogenic cAMP-cascade in thyrocytes. J Cell Physiol 168:59–70[CrossRef][Medline]
  41. Reuse S, Pirson I, Dumont JE 1991 Differential regulation of protooncogenes c-jun and jun D expressions by protein tyrosine kinase, protein kinase C, and cyclic-AMP mitogenic pathways in dog primary thyrocytes: TSH and cyclic-AMP induce proliferation but downregulate C-jun expression. Exp Cell Res 196:210–215[CrossRef][Medline]
  42. Deleu S, Pirson I, Clermont F, Nakamura T, Dumont JE, Maenhaut C 1999 Immediate early gene expression in dog thyrocytes in response to growth, proliferation and differentiation stimuli. J Cell Physiol 181:342–354[CrossRef][Medline]
  43. Depoortere F, Van Keymeulen A, Lukas J, Costagliola S, Bartkova J, Dumont JE, Bartek J, Roger PP, Dremier S 1998 A requirement for cyclin D3-cyclin-dependent kinase (cdk)-4 assembly in the cyclic adenosine monophosphate-dependent proliferation of thyrocytes. J Cell Biol 140:1427–1439[Abstract/Free Full Text]
  44. Depoortere F, Dumont JE, Roger PP 1996 Paradoxical accumulation of the cyclin-dependent kinase inhibitor p27kip1 during the cAMP-dependent mitogenic stimulation of thyroid epithelial cells. J Cell Sci 109:1759–1764[Abstract]
  45. Lukas J, Bartkova J, Bartek J 1996 Convergence of mitogenic signalling cascades from diverse classes of receptors at the cyclin D-cyclin-dependent kinase-pRb-controlled G1 checkpoint. Mol Cell Biol 16:6917–6925[Abstract]
  46. Van Keymeulen A, Bartek J, Dumont JE, Roger PP 1999 Cyclin D3 accumulation and activity integrate and rank the comitogenic pathways of thyrotropin and insulin in thyrocytes in primary culture. Oncogene 18:7351–7359[CrossRef][Medline]
  47. Roger PP, Servais P, Dumont JE 1987 Regulation of dog thyroid epithelial cell cycle by forskolin, an adenylate cyclase activator. Exp Cell Res 172:282–292[CrossRef][Medline]
  48. Depoortere F, Pirson I, Bartek J, Dumont JE, Roger PP 2000 Transforming growth factor ß(1) selectively inhibits the cyclic AMP-dependent proliferation of primary thyroid epithelial cells by preventing the association of cyclin D3-cdk4 with nuclear p27(kip1). Mol Biol Cell 11:1061–1076[Abstract/Free Full Text]
  49. Coulonval K, Bockstaele L, Paternot S, Dumont JE, Roger PP 2003 The cyclin D3-CDK4–p27kip1 holoenzyme in thyroid epithelial cells: activation by TSH, inhibition by TGFß, and phosphorylations of its subunits demonstrated by two-dimensional gel electrophoresis. Exp Cell Res 291:135–149[CrossRef][Medline]
  50. Paternot S, Coulonval K, Dumont JE, Roger PP 2003 Cyclic AMP-dependent phosphorylation of cyclin D3-bound CDK4 determines the passage through the cell cycle restriction point in thyroid epithelial cells. J Biol Chem 278:26533–26540[Abstract/Free Full Text]
  51. Paternot S, Arsenijevic T, Coulonval K, Bockstaele L, Dumont JE, Roger PP 2006 Distinct specificities of pRb phosphorylation by CDK4 activated by cyclin D1 or cyclin D3: differential involvement in the distinct mitogenic modes of thyroid epithelial cells. Cell Cycle 5:61–70[Medline]
  52. Diehl JA 2002 Cycling to cancer with cyclin D1. Cancer Biol Ther 1:226–231[Medline]
  53. Lamy F, Taton M, Dumont JE, Roger PP 1990 Control of protein synthesis by thyrotropin and epidermal growth factor in human thyrocytes: role of morphological changes. Mol Cell Endocrinol 73:195–209[CrossRef][Medline]
  54. Maenhaut C, Brabant G, Vassart G, Dumont JE 1992 In vitro and in vivo regulation of thyrotropin receptor mRNA levels in dog and human thyroid cells. J Biol Chem 267:3000–3007[Abstract/Free Full Text]
  55. Ambesi-Impiombato FS, Villone G 1987 The FRTL-5 thyroid cell strain as a model for studies on thyroid cell growth. Acta Endocrinol Suppl (Copenh) 281:242–245[Medline]
  56. Zarkowska T, Mittnacht S 1997 Differential phosphorylation of the retinoblastoma protein by G1/S cyclin-dependent kinases. J Biol Chem 272:12738–12746[Abstract/Free Full Text]
  57. Matsushime H, Quelle DE, Shurtleff SA, Shibuya M, Sherr CJ, Kato JY 1994 D-type cyclin-dependent kinase activity in mammalian cells. Mol Cell Biol 14:2066–2076[Abstract/Free Full Text]
  58. Morris JC3, Ranganathan G, Hay ID, Nelson RE, Jiang NS 1988 The effects of transforming growth factor-ß on growth and differentiation of the continuous rat thyroid follicular cell line, FRTL-5. Endocrinology 123:1385–1394[Abstract]
  59. Taton M, Lamy F, Roger PP, Dumont JE 1993 General inhibition by transforming growth factor ß 1 of thyrotropin and cAMP responses in human thyroid cells in primary culture. Mol Cell Endocrinol 95:13–21[CrossRef][Medline]
  60. Sherr CJ, Roberts JM 1999 CDK inhibitors: positive and negative regulators of G1-phase progression. Genes Dev 13:1501–1512[Free Full Text]
  61. Hall M, Bates S, Peters G 1995 Evidence for different modes of action of cyclin-dependent kinase inhibitors: p15 and p16 bind to kinases, p21 and p27 bind to cyclins. Oncogene 11:1581–1588[Medline]
  62. Bockstaele L, Kooken H, Libert F, Paternot S, Dumont JE, de Launoit Y, Roger PP, Coulonval K 2006 Regulated activating Thr 172-phosphorylation of CDK4: its relationship with cyclins and CDK "inhibitors." Mol Cell Biol 26:5070–5085[Abstract/Free Full Text]