help button home button Endocrine Society Endocrinology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
This Article
Right arrow Abstract Freely available
Right arrow Full Text (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
Right arrow Request Copyright Permission
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Toda, S.
Right arrow Articles by Sugihara, H.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Toda, S.
Right arrow Articles by Sugihara, H.
Endocrinology Vol. 138, No. 12 5561-5575
Copyright © 1997 by The Endocrine Society

ARTICLES

Transforming Growth Factor-ß1 Induces a Mesenchyme-Like Cell Shape without Epithelial Polarization in Thyrocytes and Inhibits Thyroid Folliculogenesis in Collagen Gel Culture

Shuji Toda, Sueo Matsumura1, Noboru Fujitani, Tomohisa Nishimura, Nobuhisa Yonemitsu and Hajime Sugihara

Departments of Pathology (S.T., T.N., N.Y., H.S.) and Biochemistry (S.M.), Saga Medical School, Saga 849; and the Department of Forensic Medicine and Human Genetics, Kurume University School of Medicine (N.F.), Kurume 830, Japan

Address all correspondence and requests for reprints to: Dr. Shuji Toda, Department of Pathology, Saga Medical School, Nabeshima 5–1-1, Saga 849, Japan. E-mail: toda-s{at}post.saga-med.ac.jp


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
Transforming growth factor-ß1 (TGFß1) induces a mesenchyme-like cell shape in some epithelial cell types. To clarify the role of TGFß1 in the morphological regulation of thyrocytes, we performed collagen gel culture of porcine thyrocytes with serum-free medium. TGFß1-nontreated cells organized follicles. In contrast, the cells treated with 10 ng/ml TGFß1 became spindle shaped, i.e. they resembled mesenchymal fibroblasts, and did not form follicles. To characterize the spindle-shaped cells, we examined the fine structures and expression of thyroglobulin (Tg) and cytoskeletal proteins using electron microscopy, immunohistochemistry, and immunoblotting. TGFß1-nontreated cells had microvilli at the apical side facing follicle lumen and had basal lamina at the basal side in contact with collagen gel. TGFß1-treated cells showed both microvilli and basal lamina at the basal side. TGFß1-nontreated cells expressed Tg, whereas TGFß1-treated cells showed no expression. TGFß1-nontreated cells barely expressed vimentin, but they expressed enough cytokeratin. TGFß1-treated cells extensively displayed vimentin along with the change in shape to become spindle-like and retained a decreased expression of cytokeratin. TSH (10 mU/ml) did not essentially influence any TGFß1 effects on the cells. These results indicate that TGFß1 induces a mesenchyme-like cell shape accompanied by cytoskeletal molecular change and the loss of both epithelial polarization and a function in thyrocytes, and that it results in inhibiting thyroid folliculogenesis with or without TSH.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
THYROID follicles, an essential unit of the thyroid, are embedded in extracellular matrix (ECM) (1). In three-dimensional collagen gel culture, thyrocytes (follicular epithelial cells) easily and stably organize follicle structures with physiological cellular polarity of their component cells; the apical pole with microvilli faces the follicle lumen and the basal pole with basal lamina confronts ECM (2, 3, 4, 5, 6, 7). This culture system is, therefore, suitable for studying the proliferation and differentiation of thyrocytes.

The multifunctional polypeptide transforming growth factor-ß (TGFß) plays crucial roles in morphogenesis at the embryonic and adult stages, wound healing, and immune functions through its regulations of growth, differentiation, apoptosis, ECM formation, and cytokine cross-talk in many cell types (8, 9, 10, 11). In the morphogenesis of the breast, lung, and kidney, several studies suggest that TGFß induces a mesenchyme-like cell shape in epithelial cell types of these organs and results in modulating the formation of their luminal structures (12, 13, 14, 15). In fact, ample exogenous or endogenous TGFß inhibits the formation of ductal structures in the breast (16, 17, 18). Furthermore, thyrocytes as well as the epithelial cells of these organs organize luminal structures both in vivo and in vitro, especially in collagen gel culture (5, 19, 20, 21). We have, therefore, hypothesized about the possibility that TGFß may regulate the morphology of thyrocytes and affect thyroid follicle formation. To address this hypothesis, we performed three-dimensional collagen gel culture of isolated porcine thyrocytes with or without TGFß1 as a representative of the TGFß family (9). We also examined the effects of TSH, a main regulator for thyrocytes, on the cells, either alone or in combination with TGFß1.

In this study, TGFß1-treated thyrocytes drastically became spindle shaped and did not form follicles. We also characterized these spindle-shaped cells using electron microscopy, immunohistochemistry, and immunoblotting. We herein describe for the first time that TGFß1 induces a mesenchyme-like cell shape without epithelial polarization in thyrocytes and that it results in inhibiting thyroid folliculogenesis with or without TSH.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Preparation of thyrocytes without follicle structures
Single thyrocytes without follicle structures were prepared from porcine thyroid as described previously (5, 22). Briefly, the cells dissociated with dispase I solution (bacterial neutral protease; 1000 protease U/ml MEM; Goudoh-Shusei Co., Tokyo, Japan) were first cultured in monolayer for 2–4 days in Ham’s F-12 medium supplemented with 10% FCS and 50 µg/ml gentamicin. Single cells without follicle structures were obtained from the confluent monolayer with 0.15% trypsin treatment. Most of the thyrocytes expressed cytokeratin and were clearly distinguished from fibroblasts or endothelial cells, of which a small population might contaminate the primary isolated thyrocytes. Because fibroblasts and endothelial cells did not display cytokeratin (23, 24), the isolated single thyrocytes were embedded in collagen gel.

Three-dimensional collagen gel culture
This culture was carried out as described previously (5, 22). A total of 5 x 105 cells were embedded in 0.5 ml type I collagen gel (Nitta Gelatin Co., Osaka, Japan). To avoid the effects of serum-containing factors on culture cells, we used the following serum-free medium for culture of thyrocytes (25, 26). The cells were cultured in a 1-ml 24-well plastic dish of defined serum-free Ham’s F-12 medium supplemented with ITS premix (5 µg/ml insulin, 5 µg/ml transferrin, and 5 ng/ml selenious acid; Becton Dickinson Labware, Bedford, MA), 10 µg/ml hydrocortisone, 10 ng/ml somatostatin (Peninsula Laboratories, Belmont, CA), 10 ng/ml glycyl-L-histidyl-L-lysine acetate (Biomedical Technologies, Stoughton, MD), 6 ng/ml NaI (Katayama Chemical, Osaka, Japan), and 50 µg/ml gentamicin. Culture medium was exchanged for fresh medium every 2 days. In this serum-free medium, we used 6 ng/ml NaI, because our previous studies on thyroid folliculogenesis (5, 6, 7) were performed in 10% FCS-added medium that contained about 6 ng/ml NaI.

Stimulation of culture cells with TGFß1
Thyrocytes in collagen gel culture were stimulated by 10 ng/ml purified TGFß1 (R&D Systems, Minneapolis, MN), either alone or in combination with 10 mU/ml TSH (Sigma Chemcial Co., St. Louis, MO). At the initiation of the culture, TGFß1 was added to the medium with or without TSH; thereafter, the cells were stimulated with TGFß1 every 2 days. We also used recombinant TGFß1 (King Brewing Co., Kobe, Japan) in the manner described above. No differences were found between purified and recombinant types of TGFß1 in their effects on thyrocytes.

Immunohistochemistry
Deparaffinized sections of 4% formalin-fixed paraffin-embedded gel or thyroid tissue were immunostained by the avidin-biotin complex immunoperoxidase (ABC) method, as described previously (5). The visualization of each antigen was performed for 5 min with aminoethylcarbazole (AEC substrate kit, Nichirei Co., Tokyo, Japan). To estimate a differentiating property of thyrocytes, thyroglobulin (Tg; polyclonal antibody, Medac Gesellschaft fur Klinishe Spezialpraparate, Munich, Germany) was immunostained. To elucidate the expression of cytoskeletal proteins in the cells, we also immunostained cytokeratin (monoclonal antibody, which covered a spectrum of molecular masses of 40, 45, 46, and 56 kDa; Nichirei Co., Tokyo, Japan) and vimentin (monoclonal antibody; Dako Japan Co., Tokyo, Japan). Cytokeratin and vimentin are well expressed in epithelial and mesenchymal cell types, respectively, of thyroid tissue (23, 24). To examine an expression of TGFß type I receptor that plays a central role in the signal transduction of TGFß (27), the receptor (polyclonal antibody; Santa Cruz Biotechnology, Santa Cruz, CA) was immunostained. As a positive control for Tg, cytokeratin, or vimentin, immunohistochemistry was performed on thyroid tissue (23, 24, 28). Formalin-fixed paraffin-embedded skin tissue was immunostained as a positive control for TGFß type I receptor in the manner described above (27). These controls always gave positive results. As a negative control for Tg, cytokeratin, or vimentin, PBS was used instead of each primary antibody, and normal rabbit and mouse IgGs were used in place of the primary antibodies for Tg and cytokeratin or vimentin, respectively. As a negative control for TGFß type I receptor, the receptor antibody (1 µg) neutralized with the TGFß receptor protein (10 µg; Santa Cruz Biotechnology) was used. These controls always gave negative results. In addition, to obtain the rate of cytokeratin- or vimentin-positive staining in culture thyrocytes, 1000 cells were counted, and the percentage of positive cells was calculated. To confirm colocalization of cytokeratin and vimentin in the cells, we performed double immunostaining: cytokeratin was immunostained by the ABC method and was visualized using the AEC kit, whereas vimentin was immunostained by the avidin-biotin complex immunoalkaline phosphatase and was visualized using the fast blue substrate kit (Nichirei).

Western blotting of cytokeratin and vimentin
To examine the effects of TGFß1 on the expression of intermediate filaments in thyrocytes, 70 x 105 cells were embedded in 7 ml collagen gel in 100-mm diameter plastic dishes and cultured for 7 days in 14 ml serum-free medium under various conditions as described above. After the media were aspirated, cell layer gels were washed three times with 5 ml cold PBS and scraped from the dishes. Each cell layer gel was homogenized in 7 ml 0.1 M Tris-HCl (pH 6.8) containing 0.2% SDS and 5% 2-mercaptoethanol and then centrifuged for 30 min at 20,000 x g. The supernatants were made 1% with respect to SDS and boiled for 15 min The samples were lyophilized and thereafter dissolved in 1 ml distilled water. Ten microliters of each sample were subjected to 10% SDS-PAGE and then transferred to a nitrocellulose membrane sheet (Bio-Rad, Richmond, CA). The sheet was incubated with anticytokeratin or antivimentin antibody. The antigen on the membrane was visualized by the ABC method described in the manual supplied by Bio-Rad. The density of the bands was assessed by densitometry. The results were presented as a percentage of the control values derived from cultures with neither TSH nor TGFß1 stimulation.

Detection of Tg in culture supernatant
To estimate the effects of TGFß1 on Tg synthesis and secretion of thyrocytes, we tried to detect Tg in culture supernatants at 7 days under various conditions as described above, using dot blotting with the same anti-Tg antibody as that used for immunohistochemistry. Ten milliliters of culture supernatants were lyophilized and thereafter dissolved in 0.1 ml distilled water. Two microliters of each sample were dotted onto a nitrocellulose membrane. The antigenicity was detected by the ABC method. As a standard of Tg, 0.001–100 µg/ml porcine Tg (Fluka Chemie, Buchs, Switzerland) were spotted onto the sheet. As a negative control, culture medium in which cells were not cultured was used.

Cell proliferation
At 3 and 7 days in collagen gel culture, cell proliferation was examined by immunohistochemistry for bromodeoxyuridine (BrdU; Cell Proliferation Kit, Amersham, Arlington Heights, IL) after 24-h incubation with 30 mg/ml BrdU, as described previously (5). To obtain the rate of nuclear BrdU intake, 1000 cells were counted, and percentage of BrdU-positive nuclei was calculated.

Morphology and morphometric analysis
Culture cells were observed by phase contrast microscopy. We further examined the cells with hematoxylin-eosin (H-E) staining, using deparaffinized sections of the cell layer gel that were fixed with 4% formalin, routinely processed, and embedded in paraffin (5). To examine fine structures of the cells, we also performed transmission electron microscopy by the standard method, using materials fixed with 2.5% glutaraldehyde (5).

In this study, follicle formation at 7 days in culture was estimated as follows. On H-E-staining sections of cell layer gel obtained from 10 blocks in each of various conditions, we performed the morphological analyses of culture cells by light microscopy. Structures that consisted of 2 or more cells and had clearly luminal spaces were judged as reconstructed follicles (see Fig. 2CGo). A total of 1000 follicles were counted in at least 20 randomly chosen noncontiguous and nonoverlapping fields (at high power view, x20 objective) of the H-E-staining sections. The sizes of the follicles were determined by measuring the largest diameter, using an objective micrometer. The follicles were separated into less than 30, 30–50, and more than 50 µm size groups, and the percentages of these grouped sizes were calculated. In addition, to obtain the rate of Tg-positive staining in the follicles at 2 and 7 days in culture, 1000 follicles were counted using the method described above, and the percentage of Tg-positive follicles was calculated.



View larger version (113K):
[in this window]
[in a new window]
  		Figure 2. Collagen gel culture of TGFß1-nontreated thyrocytes in the absence (-) or presence (+) of TSH. The cells just after being embedded in collagen gel are spherical (A), and they organize small follicle structures at 2 days in culture (B and C). The follicles do not grow to larger entities even after 7 days in culture (F and G). TSH does not appear to essentially affect follicle formation at 2 (D and E) and 7 days (H and I) in culture. A: TSH (-), 0 h; B and C: TSH (-), 2 days; D and E: TSH (+), 2 days; F and G: TSH (-), 7 days; H and I: TSH (+), 7 days. A, B, D, F, and H, Phase contrast microscopy; C, E, G, and I, H-E staining. Arrow, Follicle lumen; *, gel.

 
Statistical analyses
The data obtained through five independent experiments were statistically examined. The immunohistochemical data for Tg, cytokeratin, vimentin, and BrdU were analyzed by two-way ANOVA. The data for both follicle size and density of bands in electrophoresis and Western blotting were analyzed by paired Student’s t test. Values represent the mean ± SD. P < 0.05 was considered significant.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
Expression of TGFß type I receptor
We examined an expression of TGFß type I receptor in thyrocytes, using immunohistochemistry. In thyroid tissue, many thyrocytes clearly expressed the receptor (Fig. 1AGo). An absorption test resulted in negative staining for the receptor (Fig. 1BGo). Thyrocytes just after being embedded in collagen gel also displayed TGFß type I receptor (Fig. 1CGo). These results indicated that thyrocytes would be expected to respond to TGFß1, although TGFß type II receptor remained to be elucidated.



View larger version (103K):
[in this window]
[in a new window]
  		Figure 1. Immunohistochemistry for TGFß type I receptor. Thyrocytes in vivo clearly express the receptor (A). An absorption test results in negative staining for the receptor (B). The receptor is detected in spherical cells just after being embedded in collagen gel (C). F, Follicle lumen; *, gel.

 
Three-dimensional collagen gel culture
Isolated single thyrocytes without follicle structures were cultured in collagen gel with serum-free medium. Just after being embedded in collagen gel, the cells were spherical in shape, and singly and uniformly distributed (Fig. 2AGo). TGFß1-nontreated cells clearly organized follicle structures after 48–72 h in culture (Fig. 2Go, B and C). The reconstructed follicles thereafter did not grow to larger entities, and the size of most follicles (87.5 ± 5.5%) was less than 30 µm even after 7 days in culture (Fig. 2Go, F and G). The findings reported above were not essentially affected by 10 mU/ml TSH (percentage of follicles <30 µm, 92.2 ± 3.0%; Fig. 2Go, D, E, H, and I). In contrast, 10 ng/ml TGFß1-treated cells retracted and showed a multipolar appearance after 6–18 h in culture (Fig. 3Go, A and B). At 48–72 h in culture, over 90% of the cells became spindle shaped and did not organize follicle structure (Fig. 3Go, E and F). After 7 days in culture, the cells remained spindle shaped and failed to form follicles (Fig. 3IGo). Some of the cells (30–50%) had cellular linkage. This suggested that some TGFß1-affected cells retained an epithelial nature. TSH did not essentially affect these morphological changes in TGFß1-treated cells (Fig. 3Go, C, D, G, H, and J). Finally, 1 ng/ml TGFß1 induced the morphological changes described above in only a small population of the cells, although 0.1 ng/ml TGFß1 did not induce those changes (data not shown). To clarify the effects of TGFß1 on the cells in more detail, we, therefore, examined 10 ng/ml TGFß1-affected cells.



View larger version (161K):
[in this window]
[in a new window]
  		Figure 3. Collagen gel culture of TGFß1-treated thyrocytes in the absence (-) or presence (+) of TSH. At 12 h in culture, spherical cells just after being embedded in collagen gel are retracted in multipolar appearance (A and B), and they become spindle-shaped after 48 h in culture (E and F). Even after 7 days in culture, the cells remain spindle shaped and do not form follicle structures (I). At 12 h (C and D), 48 h (G and H), and 7 days (J) in culture, TGFß1 induces the morphological changes described above in TSH-treated cells as well as in TSH-nontreated cells. A and B: TSH (-), 12 h; C and D: TSH (+), 12 h; E and F: TSH (-), 48 h; G and H: TSH (+), 48 h; I: TSH (-), 7 days; J: TSH (+), 7 days. A, C, E, G, I, and J, Phase contrast microscopy; B, D, F, and H, H-E staining. *, Gel.

 
Effect of TGFß1 on Tg expression
To elucidate an effect of TGFß1 on functional differentiation of thyrocytes, we examined Tg expression as a representative of differentiating properties of the cells. Tg was clearly detected in the lumen of follicles organized by TGFß1-nontreated cells as well as in the lumen of follicles in vivo, although the staining intensity was stronger in follicles in vitro than in their counterparts in vivo (Fig. 4Go, A and B). Tg-positive rates of the reconstructed follicles after 2 and 7 days in culture were 22.3 ± 4.0% and 90.8 ± 7.3%, respectively. TSH (10 mU/ml) did not affect the staining intensity (Fig. 4CGo) or the positive rates on both culture days (day 2, 21.6 ± 3.8%; day 7, 90.1 ± 6.4%). In contrast, TGFß1-treated cells with or without TSH did not organize follicles or express Tg on the same culture days (Fig. 4Go, D and E). Data for Tg detection in culture supernatants clearly supported all of the immunohistochemical results reported above (Fig. 5Go).



View larger version (139K):
[in this window]
[in a new window]
  		Figure 4. Immunohistochemistry for Tg (A–E), cytokeratin (F–J), and vimentin (K–O) in thyroid tissue and thyrocytes at 4 days in culture. Tg is detected mainly in the lumens of thyroid follicles in vivo (A). In TGFß1-nontreated thyrocytes with TSH (C) or without TSH (B), Tg is strongly stained in lumens of organized follicles. In TGFß1-treated cells with TSH (E) or without TSH (D), Tg is not detected. Cytokeratin is clearly detected in thyrocytes in vivo (F). In TGFß1-nontreated thyrocytes with TSH (H) or without TSH (G), cytokeratin is expressed in the component cells of organized follicles, although the expression seems to be better in the cells without TSH (G) than in those with TSH (H). TGFß1-treated cells with TSH (J) or without TSH (I) retain the expression of cytokeratin. Vimentin is detected in endothelial cells of blood vessels (K), although it is not expressed in thyrocytes in vivo. In TGFß1-nontreated thyrocytes with TSH (M) or without TSH (L), vimentin is weakly stained in component cells of organized follicles. In contrast, TGFß1-treated cells with TSH (O) or without TSH (N) are strongly stained with vimentin along with the spindle-shaped change. A, F, and K, Thyroid tissue; B, G, and L, without TSH or TGFß1; C, H, and M, with TSH, without TGFß1; D, I, and N, without TSH, with TGFß1; E, J, and O, with TSH and TGFß1. Large F, Follicle lumen; *, gel.

 


View larger version (23K):
[in this window]
[in a new window]
  		Figure 5. Detection of Tg in supernatants of collagen gel culture at 7 days by dot blotting. The 100-fold concentrated samples were subjected to dot blotting as described in Materials and Methods. In TGFß1-nontreated culture, Tg is detected in both supernatants with and without TSH. In contrast, Tg is not detected in TGFß1-treated culture supernatants with or without TSH. The concentrated supernatants of TGFß1-nontreated culture appear to contain 0.1–1 µg/ml Tg on the basis of their color yields by comparison with those of standard porcine Tg (positive control). NC, Negative control (culture medium with which cells are not cultured). Arrowhead, Faint staining of 0.1 µg/ml Tg.

 
Effects of TGFß1 on the expression of cytokeratin and vimentin
To evaluate the effects of TGFß1 on the cytoskeleton of thyrocytes, we examined the expression of cytokeratin and vimentin, which are easily detected in epithelial and mesenchymal cell types, respectively, of the thyroid (23, 24, 26). In thyroid tissue, cytokeratin was detected in most thyrocytes (Fig. 4FGo). In contrast, vimentin was not expressed in thyrocytes, although it was detected in endothelial cells of blood vessels (Fig. 4KGo). In collagen gel culture, cytokeratin was expressed in both TGFß1-treated and -nontreated cells, although its expression was slightly decreased in TGFß1-treated cells (Fig. 4Go, G and I, and Fig. 6Go, B and D). Vimentin was strongly detected in TGFß1-treated cells along with a change in shape to become spindle-like, whereas TGFß1-nontreated cells minimally expressed vimentin (Fig. 4Go, L and N, and Fig. 6Go, C and D). TSH (10 mU/ml) did not essentially affect the expression of cytokeratin and vimentin in TGFß1-treated cells, whereas TSH decreased the expression of their intermediate filament types in TGFß1-nontreated cells (Fig. 4Go, H, J, M, and O, and Fig. 6Go, B–D). Cytokeratin-positive rates of the cells during the time tested under all conditions showed no significant change, and cytokeratin was constantly detected in more than 85% of the cells. The vimentin-positive rates of the cells exhibited various changes in response to culture time or conditions (Fig. 7Go). Finally, these results definitely confirmed that TGFß1-induced spindle-shaped cells originated from epithelial thyrocytes themselves, but not from fibroblasts or endothelial cells, of which a small population might contaminate the primary-isolated thyrocytes, because TGFß1-induced spindle-shaped cells displayed cytokeratin, whereas fibroblasts or endothelial cells did not express cytokeratin (23, 24, 26).



View larger version (58K):
[in this window]
[in a new window]
  		Figure 6. Electrophoresis and Western blotting of cytokeratin and vimentin. Thyrocytes cultured in collagen gel in the absence (lanes 1 and 2) or presence (lanes 3 and 4) of TGFß1, either alone (lanes 1 and 3) or in combination with TSH (lanes 2 and 4), are extracted and subjected to SDS-PAGE (A) and immunoblotting with anticytokeratin (B) or antivimentin (C) antibody. A, TGFß1-treated cells (lanes 3 and 4) show more

increased expression of actin filaments (arrow; 42 kDa) with or without TSH than TGFß1-nontreated cells (lanes 1 and 2). B, TGFß1-nontreated cells without TSH (lane 1) and with TSH (lane 2) express the highest and lowest levels, respectively, of a low molecular cytokeratin (40 kDa). TGFß1-treated cells with TSH (lane 4) or without TSH (lane 3) express a lower level of the protein than TGFß1-nontreated cells without TSH (lane 1). C, TGFß1-treated cells with TSH (lane 4) or without TSH (lane 3) express the highest level of vimentin (54 kDa), followed in order by TGFß1-nontreated cells without TSH (lane 1) and with TSH (lane 2). D, Densitometric analysis of the density of the bands. In actin expression, there is a statistical significance only between TGFß1-nontreated cells with or without TSH and TGFß1-treated cells with or without TSH (P < 0.002). In cytokeratin expression, there is a statistical significance only between TGFß1-nontreated cells without TSH and TGFß1-nontreated cells with TSH (P < 0.001) or between TGFß1-treated cells with and those without TSH (P < 0.002). In vimentin expression, there is a statistical significance between TGFß1-nontreated and -treated cells (P < 0.0001). There is also a statistical significance between TGFß1-nontreated cells with TSH and those without TSH (P < 0.0005).

 


View larger version (64K):
[in this window]
[in a new window]
  		Figure 7. Time course of vimentin expression and colocalization of cytokeratin and vimentin in culture thyrocytes. Vimentin-positive rates of TGFß1-nontreated cells with TSH and without TSH are increased to 8.4 ± 2.9% and 15.2 ± 3.3%, respectively, in a time-dependent manner, although the rates show no significant change after 48 h in culture. There is a statistical significance between TSH-treated and TSH-nontreated cells after 48 h in culture (P < 0.02). This indicates that TSH decreases vimentin expression of thyrocytes. In TGFß1-treated culture, the vimentin-positive rate of the cells with or without TSH is extensively increased to over 90% in a time-dependent manner, although the rate shows no significant change after 48 h in culture. There is a statistical significance between TGFß1-treated cells and TGFß1-nontreated cells with or without TSH after 12 h in culture (P < 0.001). This indicates that TGFß1 prominently increases vimentin expression of thyrocytes with or without TSH. Colocalization of cytokeratin (CK) and vimentin (V) in the same section is examined with double immunostaining (C). A, CK is stained in light red with AEC. After the result was photographed (A), the staining color was removed with xylene. Then, anti-CK antibody was inactivated in 0.01 M citrate buffer for 10 min at 90 C. B, V is stained in blue with fast blue. After the result was photographed (B), the staining color was removed with 95% methanol. Anti-V antibody was inactivated with the method described above. Lastly, double staining of CK and V was performed using the same procedures without the color removal of AEC and fast blue. C, Colocalization of CK and V is presented in color mixed with light red and blue.

 
Effects of TGFß1 on fine structures of thyrocytes
To examine the effects of TGFß1 on fine structures of thyrocytes, we performed electron microscopy. In TGFß1-nontreated culture, the component cells of organized follicles had physiological cellular polarity; the apical pole with microvilli faces the follicle lumen, and the basal pole with basal lamina confronts the ECM, although microvilli and basal lamina were observed after 2 and 10 days of culture, respectively (Fig. 8Go, A, D, and F). The component cells of the reconstructed follicles had foot processes at the basal side (Fig. 8Go, A and G). Colloid substances were prominently and densely seen in the follicle lumen (Fig. 8AGo). This finding was clearly consistent with the immunohistochemical results for Tg (Fig. 4BGo). The findings presented above were not essentially affected by 10 mU/ml TSH (Fig. 8Go, B, C, and E). In contrast, TGFß1-treated cells became spindle shaped and did not organize follicles (Fig. 9AGo). Rough endoplasmic reticulae (RER) and Golgi apparatuses were well developed, and there was a spindle-shaped change in the cells at 2–4 days in culture (Fig. 9Go, A and B). Thereafter, RER and Golgi apparatuses gradually decreased along with the development of actin and intermediate filaments (Fig. 10Go, A and B). The cells formed fragmented basal lamina at the contact side with collagen gel at 4–7 days in culture (Fig. 10BGo). The cells also had several microvilli at the basal side contacting with collagen gel and had no apical lumen with microvilli (Fig. 9Go, A and B). Colloid droplets were not observed in the cells (Fig. 9Go, A and B). This finding clearly supported the immunohistochemical result for Tg (Fig. 4DGo). The cells also had several lysosomes (Fig. 9BGo). Linked cells had junctional complexes at the contact point of the cells (Fig. 11Go, A and B). This indicated that some TGFß1-affected cells retained an epithelial nature in their fine structure. TSH did not essentially affect any of the TGFß1-induced fine structures in the cells (Figs. 9Go and 10Go, C and D, and Fig. 11CGo). These results indicated that TGFß1 drastically inhibited epithelial polarization of thyrocytes.



View larger version (128K):
[in this window]
[in a new window]
  		Figure 8. Electron micrograph of organized follicles in collagen gel culture of TGFß1-nontreated thyrocytes in the absence (A, D, F, and G) or presence (B, C, and E) of TSH. Follicle structure at 2 days in culture with TSH (B) or without TSH (A) has dense colloid substances (*) in its lumen. After 7 days in culture with TSH (C), dense colloid materials (*) are also observed in the lumen of organized follicle. Microvilli (MV) that show filamentous appearance (F) are clearly observed at the apical surface of follicle lumen (A and C). Foot processes (FP) that have no filamentous appearance (G) are also seen at the basal side contacting with collagen gel (A, B, and C). After 12 days in culture with TSH (E) or without TSH (D), basal lamina (arrowhead) is demonstrated at the contact side with collagen gel. Arrow, Junctional complex; L, lysosome; CD, colloid droplet.

 


View larger version (148K):
[in this window]
[in a new window]
  		Figure 9. Electron micrograph of TGFß1-treated thyrocytes in the absence (A and B) or presence (C and D) of TSH. A, TGFß1 clearly induces the spindle-shaped phenotype of the cells after 3 days in culture without TSH. B, In a higher magnification of a part of the cell in A, many RER (ER) and Golgi apparatuses (G) are clearly observed in the cytoplasm, in which lysosomes (*) are also seen. The cells have a few microvilli (arrowheads) at the contact side with collagen gel. C and D, After 3 days in culture, TSH-treated cells show structures similar to those in TSH-nontreated cells. Arrowheads, Microvilli.

 


View larger version (133K):
[in this window]
[in a new window]
  		Figure 10. Electron micrograph of TGFß1-treated thyrocytes in the absence (A and B) or presence (C and D) of TSH. A and B, After 7 days in culture without TSH, the spindle-shaped cells have well developed intermediate filaments (IF) in the cytoplasm, and the filaments appear to be consistent with vimentin. The cells also have many actin filaments (arrow) along with cell membrane and clearly form basal lamina (arrowhead) at the contact side with collagen gel. C and D, After 7 days in culture with TSH, the cells are similar to TSH-nontreated cells in the development of intermediate (IF) and actin filaments (arrow), and basal lamina (arrowhead).

 


View larger version (153K):
[in this window]
[in a new window]
  		Figure 11. Electron micrograph of TGFß1-treated thyrocytes in the absence (A and B) or presence (C) of TSH. Low magnification (A) or high magnification of serial sections of the area indicated by an arrow in A (B) is shown. After 3 days in culture without TSH, a junctional complex (arrow) is clearly formed at the contact points (arrowhead) between two linked cells. C, On the same day of culture with TSH, the structures (arrow) are also observed. E, RER; *, lysosome.

 
Effects of TGFß1 on proliferation of thyrocytes
Cell multiplication was evaluated by nuclear BrdU incorporation of thyrocytes after 3 and 7 days in culture (Fig. 12Go, A and B). The BrdU uptake of the cells under all conditions showed no significant differences between 3 and 7 days in culture (Fig. 12Go). The rates of BrdU intake in TGFß1-nontreated and -treated cells were about 13% and 5%, respectively (Fig. 12Go). TSH (10 mU/ml) decreased BrdU intake of the cells under all of the conditions described above, although there was no statistical significance between TGFß1-treated cells with TSH and their counterparts without TSH (Fig. 12Go). These results support the findings of other studies that both TGFß1 and TSH inhibit the proliferation of porcine thyrocytes (29, 30, 31, 32, 33, 34), although their coaction does not appear to elicit prominent inhibition of cell proliferation.



View larger version (38K):
[in this window]
[in a new window]
  		Figure 12. Effects of TGFß1 on proliferation of thyrocytes at 3 and 7 days in culture. Nuclear BrdU intake (arrowhead) in TGFß1-treated (B) and -nontreated cells (A) is clearly detected. The BrdU intakes of the cells in all conditions have no statistical significance between 3 and 7 days in culture. In TGFß1-nontreated cultures, rates of BrdU intake in the cells with and without TSH are about 7% and 13%, respectively (P < 0.02). In TGFß1-treated cultures, the rates in the cells with and without TSH are about 4% and 5%, respectively, and there is no statistical significance between the two conditions. There is statistical significance between TGFß1-treated cells with or without TSH and TGFß1-nontreated cells without TSH (P < 0.01), although there is no significance between TGFß1-treated cells with or without TSH and TGFß1-nontreated cells with TSH. *, Gel.

 

   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
We have shown in this study that in collagen gel culture TGFß1 drastically induces spindle cell shape in thyrocytes and inhibits thyroid follicle formation. In contrast, thyrocytes not treated with TGFß1 organize follicle structures. TGFß1-affected thyrocytes have the following characteristics. 1) The shape of the cells resembles that of mesenchymal fibroblasts (35). 2) The cells acquire prominent expression of vimentin and retain expression of cytokeratin that is epithelial cell specific, although some smooth muscle cells exceptionally display cytokeratin, although slightly, in development and atherosclerosis of the aorta (36). 3) Fifty to 70% of the cells have no cellular linkage, whereas the remnant linked cells have junctional complexes that are generally observed in epithelial cell types (1). 4) The cells do not express colloid droplets or Tg. 5) The cells have both microvilli and basal lamina at the basal side contacting with collagen gel and have no apical lumen. That is, the cells lose the epithelial apical-basal polarity specific for thyrocytes. 6) The cells well develop RER and Golgi apparatuses at an early culture stage, and then they abundantly acquire both actin filaments along with the cell membrane and intermediate filaments in the cytoplasm. This means that the cells show mesenchymal fibroblast-like reorganization of the fine structures (35). 7) All of the findings above are unaffected by TSH. The TGFß1-affected thyrocytes seem to be in a dedifferentiated state and to undergo an epithelial to mesenchymal semitransdifferentiation, in that those cells coexpress some phenotypes of both epithelial and mesenchymal fibroblast-like differentiation (37). To our knowledge, this is the first instance where TGFß1 is a potent morphological regulator for thyrocytes and an effective inhibitor of thyroid folliculogenesis.

Our results support an interesting finding by Greenburg and Hay (38) that with 10% FCS-added medium, some thyrocytes can change into mesenchyme-like cells in type I collagen gel culture of follicles, but not of isolated single thyrocytes. It is unclear, however, whether TGFß1 is involved in their phenomenon. In our previous (5, 7) and present studies, isolated single thyrocytes reconstruct follicles and do not change into mesenchyme-like cells in type I collagen gel culture with 10% FCS-added or serum-free medium. In their study, therefore, it seems to be essential that the cells with follicle structures are embedded in collagen gel. Unknown factors involved in the follicle structure itself may play a crucial role in the mesenchyme-like transdifferentiation of some thyrocytes in cooperation with collagen gel or serum factors. Considering the structure of thyroid follicles that consist of both thyrocytes and parafollicular cells (1) or may contain their bipotential precursor cells in endodermal origin (39, 40, 41), it is also conceivable that thyrocyte-parafollicular or -precursor cell interaction in the isolated follicles may be involved in the phenomenon reported by Greenburg and Hay. Finally, in our unpublished data, TGFß1 induces mesenchyme-like cell shape in isolated porcine or human single thyrocytes cultured in Matrigel (Becton Dickinson Labware, Bedford, MA) that consists mainly of type IV collagen, laminin, and fibronectin. This suggests that TGFß1-induced mesenchyme-like cell morphology is not inhibited by at least these ECM types, although concentrations of these ECM components remain to be elucidated.

In thyroid folliculogenesis under collagen gel culture of follicles, Westermark et al. (42, 43) have shown that epidermal growth factor (EGF) alone or EGF and TGFß1 (0.1–1 ng/ml) together promote both migration of thyrocytes from mother (primarily embedded) follicles and rupture of the follicle walls, and result in an increase in new microfollicle formation. They also report that microfollicles from mother follicles are formed even with a low dose of 0.1–1 ng/ml TGFß1 alone (43). In our study, a high dose of 10 ng/ml TGFß1 abolishes reorganization of follicle structures from isolated single cells. These results suggest that TGFß1 may have dose-dependent differential effects on folliculogenesis; a high dose of TGFß1 inhibits follicle formation of thyrocytes, whereas a low dose of TGFß1 does not inhibit it.

The mechanistic basis for TGFß1-induced inhibition of thyroid folliculogenesis is unclear. In our present study, TGFß1 drastically inhibits epithelial polarization of thyrocytes. Considering the role of E-cadherin, which regulates the organization of cellular polarity in thyrocytes as well as other epithelial cell types and results in modulating their organomorphogenesis or differentiated states (44, 45, 46, 47), it is conceivable that TGFß1 may inhibit thyroid folliculogenesis through down-regulation of E-cadherin expression of thyrocytes. This possibility is supported by an interesting study by Nilsson et al. (43), which showed that down-regulation of E-cadherin expression of thyrocytes may be involved in follicle disruption generated by cooperation of TGFß1 and EGF. In addition, loss of cellular polarity of thyrocytes seems to closely relate to their spindle-shaped change at least in collagen gel culture. In fact, in the transition of some thyrocytes to mesenchyme-like cells, Greenburg and Hay (38) show that the spindle-shaped change in the cells is accompanied by their loss of epithelial polarization. In other epithelial cell types of the breast, lung, and kidney, Miettinen et al. (15) show that TGFß1-induced down-regulation of their E-cadherin expression is involved in the spindle-shaped change in those cell types. These studies and ours suggest that the loss of polarization through TGFß1-mediated down-regulation of E-cadherin expression of thyrocytes or other epithelial cells may play a crucial role in an induction of the mesenchyme-like cell shape in them.

Cytokeratin is constantly expressed in thyrocytes under in vivo or in vitro conditions. In contrast, vimentin expression of the cells seems to depend on various situations or the species of the cell (23, 24, 26, 48, 49). Thus, coexpression of cytokeratin and vimentin is expected to be observed in normal or abnormal thyrocytes under various conditions, although in our current study coexpression of the two intermediate filament types is not seen, at least in porcine thyrocytes, in vivo. In thyroid tumors, including hyalinizing trabecular adenoma and papillary, follicular, or anaplastic carcinoma, the spindle or nonspindle tumor cells coexpress both cytokeratin and vimentin (23, 24, 26, 48). Also, in spindle cell carcinoma of skin, esophagus, gallbladder, or larynx, the sarcomatoid tumor cells coexpress cytokeratin and vimentin (23, 50, 51). Furthermore, TGFß1 induces spindle cell shape with coexpression of cytokeratin and vimentin in thyrocytes, as described in our present study. These results suggest that TGFß1 may be involved in the pathogenesis of coexpression of cytokeratin and vimentin or spindle cell shape in epithelial tumor cells of the thyroid or other organs. In addition, Coclet et al. (49) show that EGF-treated cells thereafter cultured with TSH can regain an epithelial morphology from EGF-induced spindle cell shape despite the persistence of coexpression of cytokeratin and vimentin (49). This suggests that the coexpression of cytokeratin and vimentin may not always be associated with the dedifferentiated fibroblast-like cell shape.

Thyrocytes express Tg in monolayer or collagen gel culture (2, 5). The cells in vitro have apical-basal polarity. The cells in the monolayer have the apical side with microvilli facing culture medium and the attachment (basal) side without basal lamina confronting plastic surface (2, 22), whereas the cells in the collagen gel exhibit physiological polarity, as explained in the introductory section above. However, TGFß1-affected thyrocytes have no apical-basal polarity, as described in this study. In addition, several studies using collagen gel culture of thyrocytes show that single cells without polarization do not express Tg before reorganization of follicle structures (5, 7, 38). These results suggest that polarization of thyrocytes may be a prerequisite for Tg expression of the cells, although we cannot at present rule out the possibility that TGFß1 may directly inhibit Tg expression of the cells, because TGFß1 inhibits iodide intake and its organization in thyrocytes (30, 31). Finally, in TGFß1-nontreated cells we unexpectedly found that colloid substances and Tg in the lumens of the organized follicles with or without TSH are more prominently and densely detected in serum-free culture than in 10% FCS-added culture (5, 6, 7), although both of those culture media have almost the same concentration of iodide. It is conceivable that some factors added to the serum-free medium, although unnameable at present, may play a crucial role in this phenomenon. It is also conceivable that serum-containing factors inhibit Tg expression of the cells.

Many studies show that TGFß extensively promotes ECM production in some cell types (52). In thyrocytes, Garbi et al. (53) have shown that TGFß1 accelerates the production of fibronectin and laminin. Our present study also has shown by electron microscopy that TGFß1-treated thyrocytes in collagen gel culture clearly form fragmented basal lamina even at 4–7 days in culture, although TGFß1-nontreated cells do not organize basal lamina before 10 days in culture, as described herein and previously (5, 6). These results suggest that TGFß1 accelerates ECM production in thyrocytes. Further studies are needed to clarify in more detail what kinds of ECM components TGFß1-treated cells can produce and what role each of the ECM molecules produced in the microenvironment plays in the biological behavior of the cells.

TSH is a main differentiating factor for thyrocytes. However, it has not yet been clearly decided whether thyroid folliculogenesis is TSH independent (54, 55) or TSH dependent (56 57). In collagen gel culture of TGFß1-nontreated thyrocytes with serum-free medium, we have shown in this study that the cells can reconstruct follicle structures in TSH-free medium as well as in TSH-added medium. Furthermore, our present and previous studies have demonstrated that the growth of reorganized follicles is less extensive in serum-free medium than in 10% FCS-added medium (5, 6). These results suggest the following conclusions. 1) Thyroid folliculogenesis itself may be essentially TSH independent, at least in collagen gel culture, although TSH receptor activity of the cells in this culture system remains to be elucidated. 2) Many known or unknown serum-containing growth factors may cooperatively play crucial roles in growth of the follicles, as suggested by Dumont et al. (58). Finally, our present study also disclosed that TSH decreases the expression of cytokeratin and vimentin in TGFß1-nontreated thyrocytes. In our current study, their significance remains to be elucidated, and further studies are in order.

In conclusion, we have shown in collagen gel culture of isolated single thyrocytes that TGFß1 induces a mesenchyme-like cell shape without epithelial polarization in the cells and that it results in inhibiting both differentiation of the cells and thyroid folliculogenesis in a TSH-independent manner. This suggests that TGFß1 is a potent morphological regulator for the cells and may be involved in development and morphogenesis of the thyroid. Further studies using this culture method will probably provide new clues to the mechanism of thyroid folliculogenesis that closely involves the proliferation and differentiation of thyrocytes.


   Acknowledgments
 
We thank Messrs. H. Ideguchi, S. Nakahara, F. Mutoh, K. Tomoda, and S. Takuma for technical assistance, and Messrs. T. Tanamachi and Y. Tateishi for photography. We also thank Prof. H. Kimura, Dr. T. Hashiguchi, and Prof. R. Gärtner for their helpful support, and Prof. K. Tohkaichi for helping to edit the English of the manuscript.


   Footnotes
 
1 Current address: Department of Basic Allied Medicine, Faculty of Health Science, Kobe University School of Medicine, Suma, Kobe 654–01, Japan. Back

Received April 28, 1997.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Capen CC 1991 Anatomy, comparative anatomy, and histology of the thyroid. In: Braverman LE, Utiger RD (eds) The Thyroid. Lippincott, Philadelphia, pp 22–40
  2. Bidey SP, Tomlinson S 1988 The regulation and integration of thyroid follicular differentiation and function. Clin Endocrinol (Oxf) 28:423–444[Medline]
  3. Chambard MB, Gabrion VJ, Maucham J 1984 Polarity reversal of inside-out thyroid follicles cultured within collagen gel: reexpression of specific functions. Biol Cell 51:315–326[Medline]
  4. Garbi C, Nitsch L, Wollman SH 1984 Embedding in a collagen gel stabilizes the polarity of epithelial cells in thyroid follicles in suspension culture. Exp Cell Res 151:458–465[CrossRef][Medline]
  5. Toda S, Sugihara H 1990 Reconstruction of thyroid follicles from isolated porcine follicle cells in three-dimensional collagen gel culture. Endocrinology 126:2027–2034[Abstract]
  6. Toda S, Yonemitsu N, Hikichi Y, Koike N, Sugihara H 1992 Differentiation of human thyroid follicle cells from normal subjects and Basedow’s disease in three-dimensional collagen gel culture. Pathol Res Pract 188:874–882[Medline]
  7. Toda S, Yonemitsu N, Minami Y, Sugihara H 1993 Plural cells organize thyroid follicles through aggregation and linkage in collagen gel culture of porcine follicle cells. Endocrinology 133:914–920[Abstract]
  8. Kingsley DM 1994 The TGF-ß superfamily: new members, new receptors, and new genetic tests of function in different organisms. Genes Dev 8:133–146[Free Full Text]
  9. Massague J 1990 The transforming growth factor-beta family. Annu Rev Cell Biol 6:597–564[CrossRef]
  10. McCartney-Francis NL, Wahl SM 1994 Transforming growth factor-ß: a matter of life and death. J Leukocyte Biol 55:401–409[Abstract]
  11. Sporn MB, Roberts AB 1992 Transforming growth factor-beta: recent progress and new challenges. J Cell Biol 119:1017–1021[Free Full Text]
  12. Silberstein GB, Daniel CW 1987 Reversible inhibition of mammary gland growth by transforming growth factor-ß1. Science 237:291–293[Abstract/Free Full Text]
  13. Heine UI, Munoz EF, Flanders KC, Roberts AB, Sporn MB 1990 Colocalization of TGF-beta 1 and collagen I and III, fibronectin and glycosaminoglycans during lung branching morphogenesis. Development 109:29–36[Abstract]
  14. Roger SA, Ryan G, Purchio AF, Hammerman MR 1993 Metanephric transforming growth factor-ß1 regulates nephrogenesis in vitro. Am J Physiol 264:F996–F1002
  15. Miettinen PJ, Ebner R, Lopez AR, Derynck R 1994 TGF-ß1 induced transdifferentiation of mammary epithelial cells to mesenchymal cells: involvement of type I receptors. J Cell Biol 127:2021–2036[Abstract/Free Full Text]
  16. Daniel GW, Silberstein GB, van Horn K, Strickland P, Robinson S 1989 TGF-ß1-induced inhibition of mouse mammary ductal growth: developmental specificity and characterization. Dev Biol 135:20–30[CrossRef][Medline]
  17. Robinson SD, Silberstein GB, Roberts AB, Flanders KC, Daniel CW 1991 Regulated expression growth inhibitory effects of transforming growth factor-beta isoforms in mouse mammary gland development. Development 113:867–878[Abstract]
  18. Pierce DFJ, Johnson MD, Matsui Y, Robinson SD, Gold LI, Purchio AF, Daniel CW, Hogan BLM, Moses HL 1993 Inhibition of mammary duct development but not alveolar outgrowth during pregnancy in transgenic mice expressing active TGF-ß1. Genes Dev 7:2308–2317[Abstract/Free Full Text]
  19. Yang J, Richards J, Bowman P, Guzman R, Enami J, McCormick K, Hamamoto S, Pitelka D, Nandi S 1979 Sustained growth and three-dimensional organization of primary mammary tumor epithelial cells embedded in collagen gels. Proc Natl Acad Sci USA 76:3401–3405[Abstract/Free Full Text]
  20. Sugihara H, Toda S, Miyabara S, Fujiyama C, Yonemitsu N 1993 Reconstruction of alveolus-like structure from alveolar type II epithelial cells in three-dimensional collagen gel matrix culture. Am J Pathol 142:783–792[Abstract]
  21. Montesano R, Matsumoto K, Nakamura T 1991 Identification of a fibroblast-derived epithelial morphogen as hepatocyte growth factor. Cell 67:901–908[CrossRef][Medline]
  22. Toda S, Sugihara H 1996 Primary culture of the thyroid: three-dimensional culture using extracellular matrix. In: Griffiths JB, Doyle A, Newell DG (eds) Cell and Tissue Culture: Laboratory Procedures. Wiley and Sons, London, pp 17B:2.1–12
  23. Osborn M, Weber K 1983 Tumor diagnosis by intermediate filament typing: a novel tool for surgical pathology. Lab Invest 48:372–394[Medline]
  24. Mittinen M, Franssila K, Lehto VP, Paasivuo R, Virtanen I 1984 Expression of intermediate filament proteins in thyroid gland and thyroid tumors. Lab Invest 50:262–270[Medline]
  25. Errick JE, Ing KW, Eggo M, Burrow G 1986 Growth and differentiation in cultured human thyroid cells: effects of epidermal growth factor and thyrotropin. In Vitro Cell Dev Biol 22:28–36[Medline]
  26. Hamilton WG, Ham RG 1977 Clonal growth of Chinese hamster cell lines in protein-free media. In Vitro 13:537–547[Medline]
  27. Attisano L, Carcamo J, Ventura F, Weis FMB, Massague J, Wrana JL 1993 Identification of human activin and TGF-ß type I receptors that form heterometric kinase complexes with type II receptors. Cell 75:671–680[CrossRef][Medline]
  28. Henzen-Logmans SC, Mullink H, Ramaekers FCS, Tadema T, Meijer CJLM 1987 Expression of cytokeratin and vimentin in epithelial cells of normal and pathologic thyroid tissue. Virchows Arch A 410:347–354[CrossRef]
  29. Grubeck-Loebenstein B, Buchan G, Sadeghi R, Kissonerghis M, Turner M, Pirich K, Roka R, Niederle B, Kassal H, Waldhausl W, Feldmann M 1989 Transforming growth factor beta regulates thyroid growth: role in the pathogenesis of nontoxic goiter. J Clin Invest 83:764–770
  30. Morris III JC, Ranganathan G, Hay ID, Nelson RE, Jiang NJ 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]
  31. Tsushima T, Arai M, Saji M, Ohba Y, Murakami H, Ohmura E, Sato K, Shizume K 1988 Effects of transforming growth factor-ß on deoxyribonucleic acid synthesis and iodine metabolism in porcine thyroid cells in culture. Endocrinology 123:1187–1194[Abstract]
  32. Gärtner R, Garbil W, Demharter P, Horn K 1985 Involvement of cyclic AMP, iodine and metabolites of arachidonic acid in the regulation of cell proliferation of isolated porcine thyroid follicle cells. Mol Cell Endocrinol 42:145–155[CrossRef][Medline]
  33. Toda S, Matsumura S, Yonemitsu N, Fujitani N, Terayama K, Funatsumaru S, Sugihara H 1995 Effects of various types of extracellular matrices on adhesion, proliferation, differentiation, and c-Fos protein expression of porcine thyroid follicle cells. Cell Struct Funct 20:345–354[Medline]
  34. Westermark K, Westermark B, Karlsson FA, Ericson LE 1986 Location of epidermal growth factor receptors on porcine thyroid follicle cells and receptor regulation by thyrotropin. Endocrinology 118:1040–1046[Abstract]
  35. Facett DW 1994 Cells of connective tissue. In: Facett DW (ed) A Textbook of Histology, ed 12. Chapman and Hall, New York, pp 146–155
  36. Bader BL, Jahn L, Franke WW 1988 Low level expression of cytokeratins 8, 18 and 19 in vascular smooth muscle cells of human umbilical cord and in cultured cells derived therefrom, with an analysis of the chromosomal locus containing cytokeratin 19 gene. Eur J Cell Biol 47:300–319[Medline]
  37. Eguchi G, Kodama R 1993 Transdifferentiation. Curr Opin Cell Biol 5:1023–1028[CrossRef][Medline]
  38. Greenburg G, Hay ED 1988 Cytoskeleton and thyroglobulin expression change during transformation of thyroid epithelium to mesenchyme-like cells. Development 102:605–622[Abstract]
  39. Kameda Y, Ikeda A 1978 The identification of a specific fragment to dog thyroglobulin responsible for immunoreactivity to parafollicular cells. Endocrinology 102:1702–1709[Medline]
  40. Kameda Y, Ikeda A 1980 Immunohistochemical study of the C-cell complex of dog thyroid glands with reference to the reactions of calcitonin, C-thyroglobulin and 19S thyroglobulin. Cell Tissue Res 208:405–415[Medline]
  41. Kameda Y, Shigemoto H, Ikeda A 1980 Development and cytodifferentiation of C cell complexes in dog fetal thyroids. Cell Tissue Res 206:403–415[Medline]
  42. Westermark K, Nilsson M, Ebendal T, Westermark B 1991 Thyrocyte migration and histiotypic follicle regeneration are promoted by epidermal growth factor in primary culture of thyroid follicles in collagen. Endocrinology 129:2180–2186[Abstract]
  43. Nilsson M, Dahlman T, Westermark B, Westermark K 1995 Transforming growth factor-ß promotes epidermal growth factor-induced thyroid cell migration and follicle neoformation in collagen gel separable from cell proliferation. Exp Cell Res 220:257–265[CrossRef][Medline]
  44. Hoang-Vu C, Cetin Y, Sheumann GFW, Behrends J, Horn R, Von Zu Muhlen A, Dralle H, Brabant G 1993 Expression of the cellular adhesion molecule E-cadherin in normal human thyrocytes, and in differentiated and undifferentiated thyroid carcinomas. Exp Clin Endocrinol 101:78–82[Medline]
  45. Brabant G, Hoang-Vu C, Behrends J, Cetin Y, Potter E, Dumont JE, Maenhaut C 1995 Regulation of the cell-cell adhesion protein, E-cadherin, in dog and human thyrocytes in vitro. Endocrinology 136:3113–3119[Abstract]
  46. Takeichi M 1991 Cadherin cell adhesion receptors as a morphogenetic regulator. Science 251:1451–1455[Abstract/Free Full Text]
  47. Wollner DA, Krzeminski KA, Nelson WJ 1992 Remodeling the cell surface distribution of membrane proteins during development of epithelial cell polarity. J Cell Biol 116:889–899[Abstract/Free Full Text]
  48. Dockhorn-Dworniczak B, Franke WW, Schroder S, Czernobilsky B, Gould VE, Bocker W 1987 Patterns of expression of cytoskeletal proteins in human thyroid gland and thyroid carcinomas. Differentiation 35:53–71[CrossRef][Medline]
  49. Coclet J, Lamy F, Rickaert F, Dumont JE, Roger PP 1991 Intermediate filaments in normal thyrocytes: modulation of vimentin expression in primary cultures. Mol Cell Endocrinol 76:135–148[CrossRef][Medline]
  50. Nishihara K, Tsuneyoshi M 1993 Undifferentiated spindle cell carcinoma of the gallbladder: a clinicopathologic, immunohistochemical, and flow cytometric study of 11 cases. Hum Pathol 24:1298–1305[CrossRef][Medline]
  51. Toda S, Maehara N, Yonemitsu N, Miyabara S, Koike N, Sugihara H 1989 Polypoid squamous cell carcinoma of the larynx: an immunohistochemical study for ras p21 and cytokeratin. Pathol Res Pract 185:860–866[Medline]
  52. Wahl SM 1991 Transforming growth factor beta (TGF-ß) in inflammation: a cause and a cure. J Clin Immunol 12:61–74
  53. Garbi C, Colletta G, Cirafici AM, Marchisio PC, Nitsch L 1991 Transforming growth factor-beta induced cytoskeleton and extracellular matrix modifications in FRTL-5 thyroid epithelial cells. Eur J Cell Biol 53:281–289
  54. Fayet G, Hovsepian S, Dickson JG, Lissitzky S 1982 Reorganization of porcine thyroid cells into functional follicles in a chemically defined, serum- and thyrotropin-free medium. J Cell Biol 93:479–488[Abstract/Free Full Text]
  55. Hilfer SR, Searis RL 1980 Differentiation of the thyroid in the hypophysectomized chick embryo. Dev Biol 79:107–118[CrossRef][Medline]
  56. Fayet G, Michel-Bechet M, Lissitzky S 1971 Thyrotropin-induced aggregation and reorganization into follicles of isolated porcine-thyroid cells in culture. II. Ultrastructural studies. Eur J Biochem 24:100–111[Medline]
  57. Kerkof PR, Lond PJ, Chaikoff IL 1964 In vitro effects of thyrotropic hormone. I. On the pattern of organization of monolayer cultures of isolated sheep thyroid gland cells. Endocrinology 74:170–179
  58. Dumont JE, Maenhaut C, Prison I, Baptist M, Roger PP 1991 Growth factors controlling the thyroid gland. Baillieres Clin Endocrinol Metab 5:727–754[Medline]



This article has been cited by other articles:


Home page
 EndocrinologyHome page
E. Sonoda, S. Aoki, K. Uchihashi, H. Soejima, S. Kanaji, K. Izuhara, S. Satoh, N. Fujitani, H. Sugihara, and S. Toda
A New Organotypic Culture of Adipose Tissue Fragments Maintains Viable Mature Adipocytes for a Long Term, Together with Development of Immature Adipocytes and Mesenchymal Stem Cell-Like Cells
Endocrinology, October 1, 2008; 149(10): 4794 - 4798.
[Abstract] [Full Text] [PDF]

Home page
 J. Cell Sci.Home page
M. Grande, A. Franzen, J.-O. Karlsson, L. E. Ericson, N.-E. Heldin, and M. Nilsson
Transforming growth factor-{beta} and epidermal growth factor synergistically stimulate epithelial to mesenchymal transition (EMT) through a MEK-dependent mechanism in primary cultured pig thyrocytes
J. Cell Sci., November 15, 2002; 115(22): 4227 - 4236.
[Abstract] [Full Text] [PDF]

Home page
 Endocr. Rev.Home page
N. Auersperg, A. S. T. Wong, K.-C. Choi, S. K. Kang, and P. C. K. Leung
Ovarian Surface Epithelium: Biology, Endocrinology, and Pathology
Endocr. Rev., April 1, 2001; 22(2): 255 - 288.
[Abstract] [Full Text]

Home page
 J. Leukoc. Biol.Home page
K. Chen, Y. Wei, G. C. Sharp, and H. Braley-Mullen
Characterization of thyroid fibrosis in a murine model of granulomatous experimental autoimmune thyroiditis
J. Leukoc. Biol., December 1, 2000; 68(6): 828 - 835.
[Abstract] [Full Text]

Home page
 EndocrinologyHome page
H. Tonoli, V. Flachon, C. Audebet, A. Calle, T. Jarry-Guichard, M. Statuto, B. Rousset, and Y. Munari-Silem
Formation of Three-Dimensional Thyroid Follicle-Like Structures by Polarized FRT Cells Made Communication Competent by Transfection and Stable Expression of the Connexin-32 Gene
Endocrinology, April 1, 2000; 141(4): 1403 - 1413.
[Ab