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 Tonoli, H.
Right arrow Articles by Munari-Silem, Y.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Tonoli, H.
Right arrow Articles by Munari-Silem, Y.
Endocrinology Vol. 141, No. 4 1403-1413
Copyright © 2000 by The Endocrine Society

ARTICLES

Formation of Three-Dimensional Thyroid Follicle-Like Structures by Polarized FRT Cells Made Communication Competent by Transfection and Stable Expression of the Connexin-32 Gene1

Hélène Tonoli2, Virginie Flachon2, Christine Audebet, Aleth Callé, Therese Jarry-Guichard, Massimo Statuto3, Bernard Rousset and Yvonne Munari-Silem

INSERM, U-369, Faculté de Médecine Lyon-RTH Laennec (H.T., V.F., C.A., A.C., B.R., Y.M.S.), 69372 Lyon; Laboratoire de Génétique et Physiologie du Développement, Institut de Biologie du Développement de Marseille, Campus de Luminy (T.J.G.), 13288 Marseille, France

Address all correspondence and requests for reprints to: Dr. Yvonne Munari-Silem, INSERM U-369, Faculté de Medecine, RTH Laennec, 7 rue Guillaume Paradin, 69372 Lyon Cedex 08, France. E-mail: u369{at}laennec.univ-lyon1.fr


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
Pig thyrocytes, either in the intact gland or cultured under conditions leading to thyroid follicle reconstitution, coexpress two gap junction proteins, connexin-32 (Cx32) and connexin-43 (Cx43). As thyrocytes cultured in the form of a monolayer only express Cx43, we hypothesized that Cx32 could play a role in thyroid folliculogenesis. In the present work, we analyzed the ability of polarized FRT cells (that are gap junction deficient) to form follicle-like structures after stable transfection with either Cx32 or Cx43 genes.

Wild-type and transfected FRT cells, while growing, showed the capacity to form three-dimensional structures corresponding to domes that result from the accumulation of fluid underneath limited areas of the cell layer. The number of domes formed by FRT cells expressing Cx32 (FRT-Cx32) was 2- to 3-fold higher than that obtained with either wild-type or Cx43-transfected FRT cells (FRT-Cx43). Domes generated by FRT-Cx32 cells were stable (beyond 3 weeks of culture), whereas those formed from wild-type or FRT-Cx43 cells were transient, disappearing when cells reached confluence. Inspection of the cell organization within domes formed from FRT-Cx32 cells by phase contrast and confocal microscopy revealed a progressive transition from domes toward closed structures with a lumen. The tightness of the lumen was demonstrated by the retention of a fluorescent probe, lucifer yellow, introduced by microinjection. Electron microscope examinations showed that the neoformed follicle-like structures had an inside-out polarity. Analyses of cell motion and division with time, by fluorescence video microscopy, indicated that the transformation of domes into inside-out follicles brings into play the migration of cells and, to a lesser extent, cell multiplication underneath the domes.

In conclusion, FRT cells forced to express Cx32 give rise to domes that transform into closed inside-out follicles. This gain of function appears Cx specific, as FRT-Cx43 cells did not form similar structures. Our data suggest that the formation and/or functioning of Cx32 gap junctions might represent a key event in thyroid epithelium morphogenesis, i.e. formation of a lumen from a tight epithelial cell layer.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
EPITHELIA EXHIBIT a morphology adapted to their specific functions (1). Although much has been learned about the general principles underlying the biogenesis of polarized monolayers from individual cells (2, 3, 4, 5, 6), the mechanisms that determine how each epithelium acquires and keeps its tissue-specific morphology throughout its lifespan are much less understood. In the thyroid gland, the epithelial cell layer surrounds a lumenal compartment and forms tight spherical structures, the follicles. Such a three-dimensional cell organization, which is unique to the thyroid, is required for tissue function, i.e. the synthesis and secretion of thyroid hormones. Indeed, the lumenal compartment is both the site of storage of the thyroid prohormone, thyroglobulin, and the site of thyroglobulin iodination by thyroid peroxidase bound to the apical plasma membrane of cells delimiting the follicle lumen (7, 8).

Folliculogenesis is a complex phenomenon that involves the coordination of many different cellular activities for the positioning of cells with respect to one another and for the setting of inside-in polarity (8, 9). Although the influence of cell-substrate contacts in the formation and stability of follicles has been studied in detail (8, 10, 11, 12), the contribution of cell-cell contacts is less documented (13). The present study aimed at investigating the role of gap junctions (GJ) on folliculogenesis. GJ are composed of integral membrane proteins or connexins (Cx) that oligomerize in the plasma membrane of one cell to form connexons, delineating a central aqueous pore. Connexons of two adjacent cells dock in a mirror symmetry through the extracellular domains of Cx molecules, creating a continuous channel allowing the direct cell to cell transfer of cytoplasmic molecules with a molecular mass lower than 1 kDa. The complementary DNA (cDNA) for 14 different Cx have been cloned in mammals (14, 15, 16). In a previous work we reported that normal pig thyrocytes in situ coexpress two Cx, Cx32 and Cx43, that form distinct GJ channels; Cx43 GJ is located within tight junctions and colocalized with ZO1 (17). Rodent thyrocytes also express Cx26 (18, 19, 20). Interestingly, the Cx expression profile of pig thyroid epithelial cells depends on their morphological phenotype; thyroid cells organized in follicles in situ or in primary culture coexpress Cx32 and Cx43. By contrast, thyroid cells cultured in the form of a monolayer no longer express Cx32 but still express Cx43 (17, 21). Green et al. (19, 20, 22) found a reduction of Cx gene expression in the gland of mice spontaneously developing autoimmune thyroid disease and rats in which autoimmune thyroiditis was induced by immunization with thyroglobulin. Interestingly, thyroid cells from the diseased animals form fewer follicles and expressed reduced amounts of Cx32 in primary culture. These observations, suggesting that Cx expression is somewhat linked to folliculogenesis, prompted us to examine the respective roles of Cx43 GJ and Cx32 GJ in thyroid cell morphological differentiation using the thyroid-derived cell line FRT as a model system.

FRT cells are metabolically dedifferentiated, but still express a thyroid-specific transcription factor, Pax8 (23) and are polarized and connected by a continuous belt of tight junctions. In culture, FRT cells have the ability to form domes that correspond to domains of the epithelial layer where cells detached from the culture dish due to transepithelial transport of ions and water and accumulation of fluid underneath the cell layer (24, 25). These cells are communication deficient and no longer express or express at a very low level the Cx normally present in the rat thyroid gland (26). Our approach has been to restore the expression of Cx32 or Cx43 in FRT cells by transfection and stable expression of Cx32 or Cx43 cDNA and to study whether the formation of Cx32-GJ or Cx43-GJ modifies the phenotype of FRT cells, paying special attention to the capacity of the cell monolayer to form three-dimensional structures with thyroid follicle morphological characteristics. We report that FRT-Cx32 cells, but not FRT-Cx43 cells, acquired the ability to form follicle-like structures from domes.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Cell culture
The FRT cell line, provided by Prof. L. Nitsch (Dipartimento di Biologia e Patologia Cellulare e Moleculare L. Califano, Università Degli Studi di Napoli Federico II, Naples, Italy), was originally established and characterized by Ambesi-Impiombato et al. (27) and Nitsch et al. (24). It derives from the spontaneous transformation of normal thyroid cells from Fisher rats. FRT cells were routinely grown in Coon’s modified Ham’s F-12 medium (Seromed, Berlin, Germany) containing 100 U/ml penicillin and 0.1 mg/ml streptomycin and supplemented with 5% FCS and a four-hormone mixture (10 µg/ml insulin, 10 nM hydrocortisone, 5 µg/ml transferrin, and 10 ng/ml glycyl-L-histidyl-L-lysine acetate). All products, including (Bu)2cAMP and 8-bromo-cAMP (8Br-cAMP) used in some experiments, were obtained from Sigma (St. Louis, MO). FRT-Cx32 cell clones had been previously obtained in the laboratory by stable transfection of rat Cx32 cDNA (26). Cells of the three FRT-Cx32 clones used in the present study expressed Cx32 protein and were coupled to similar level as previously described (26). FRT-Cx32 cells and FRT-Cx43 cells (obtained as described below) were cultured under the same conditions as wild-type FRT cells. In all experiments, FRT, FRT-Cx32, or FRT-Cx43 cells were seeded at a density of 50,000 cells/cm2 and maintained in the standard culture medium throughout the experiments. Medium was changed every 2–3 days.

Constructs and transfection
The expression vector encoding full-length Cx43 (pSVK3-Cx43) was made by insertion of the 1.8-kb human Cx43 cDNA fragment isolated by BamHI and XhoI digestion from the Bluescript M13 plasmid (provided by Dr. Glenn I. Fishman, Albert Einstein College of Medicine, Bronx, NY) (28), into the pSVK3 expression vector (Amersham Pharmacia Biotech, Orsay, France). pSVK3 was opened at its unique SmaI site, and Cx43 cDNA was inserted after a conversion into blunt ends. The proper orientation of the cDNA insert was controlled by electrophoretic analysis of EcoRI restriction enzyme digestion fragments and sequencing. FRT cells expressing Cx43 were obtained following the same experimental protocol as that previously used to obtain FRT-Cx32 transfectants (26). Briefly, FRT cells were cotransfected with the plasmids pSVK3-Cx43 and pCMV-neo in a 1:15 ratio, using the calcium phosphate precipitation procedure. Transfected cells were selected using the neomycin analog G418 and were screened for a high level of intercellular communication visualized by the cell to cell transfer of the fluorescent probe, lucifer yellow, microinjected into one of the cells. Clones showing cell to cell transfer of lucifer yellow were amplified and further characterized for Cx43 messenger RNA (mRNA) and protein expression levels.

Microinjection
Microinjection experiments were performed as previously reported (29). Briefly, the fluorescent probes, lucifer yellow (50 mg/ml) and fluorescein isothiocyanate-dextran (FITC-dextran; 10 mg/ml) were injected into the cell cytoplasm or into the lumen of three-dimensional structures using a micromanipulator 5170 and a microinjector 5242 from Eppendorf (Hamburg, Germany) installed on an Axiovert 35M inverted microscope from Carl Zeiss (Oberkochen, Germany). Microinjections were performed under nitrogen gas pressure using glass micropipette (Femtotips) from Eppendorf. For cell microinjection, the applied pressure was adjusted to between 40 and 80 hectopascal, and the time of injection varied from 0.1–0.3 sec, depending on the micropipette. For injections into the lumen of three-dimensional structures, the working pressure and the time of microinjection were increased 5- to 10-fold depending on the size of the lumen.

Indirect immunofluorescence labeling
Cells were fixed and permeabilized for 30 min at room temperature in 4% paraformaldehyde in 10 mM PBS containing 0.25% (vol/vol) Triton-X100 and 0.25% (vol/vol) Tween-20. After a 30-min treatment in PBS containing 1 mg/ml BSA (PBS-BSA), cells were incubated with purified anti-Cx43 antibodies (1 µg/ml) diluted in PBS-BSA for 90 min at room temperature. Rabbit anti-Cx43 antibodies had been previously characterized (30). Goat antirabbit IgG, F(ab')2 fragment specific, conjugated to fluorescein (from Sigma) were used as secondary antibody. Fluorescence images were obtained as previously described (26).

Tubulin labeling was performed using an anti-{alpha}-tubulin mouse monoclonal antibody from Amersham Pharmacia Biotech (Aylesbury, UK). Immune complexes were revealed using FITC-labeled goat antimouse IgG, F(ab')2 fragment specific, from Jackson ImmunoResearch Laboratories, Inc. (West Grove, PA).

Cell proliferation
Cell proliferation was analyzed by 5-bromo-2'deoxyuridine (BrdU) incorporation into DNA. Cells incubated for 4 or 12 h in the presence of 3 µg/ml BrdU (Amersham Pharmacia Biotech) in RPMI 1640 medium (Seromed, Berlin, Germany) were fixed in 70% (vol/vol) ethanol (precooled at -20 C) for 30 min at 4 C and subjected to a 10-min treatment in 2 N HCl at 37 C. After extensive washing in 0.1 M borate, pH 8.5, cells were incubated for 1 h at room temperature with an anti-BrdU mouse monoclonal antibody (ICN Biomedicals, Inc., Costa Mesa, CA). Immune complexes were visualized using the secondary antibody mentioned above.

Photon and electron microscopy
Phase contrast and fluorescence images were made on an Axiovert 35M inverted microscope from Carl Zeiss using a Contax camera. To follow cell migration within or in the vicinity of domes, FITC-dextran (150 kDa) was microinjected into the cytoplasm (31). Guide marks made on the bottom of the culture dish served to locate and follow the track of individual labeled cells for up to 2 or 3 days. Confocal microscope examinations were performed using a laser scanning confocal microscope (LSM 510) from Carl Zeiss (Laboratoire de Biologie Moleculaire et Cellulaire, Ecole Normale Superieure de Lyon, Lyon, France). Transmission electron microscopy was performed after classical steps of fixation of cells with glutaraldehyde, postfixation with OsO4, and embedding in Epon. Ultrathin sections (50–80 nm) were contrasted with uranyl acetate and lead citrate, and then examined on a JEOL 1200EX transmission electron microscope (Centre Commun d’Imagerie de la Faculté de Médecine Laennec, Lyon, France).

Northern blot
Northern blot analysis of Cx43 was performed on total RNA as previously described (26). The probe used for the hybridization was the Cx43 cDNA used for the pSVK3-Cx43 plasmid construction.

Western blot
Cell membrane extracts were prepared as previously described (26). Cx43 was immunodetected with the purified anti-Cx43 antibodies used for indirect immunofluorescence labeling (final concentration, 1 µg/ml).


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
Cx43 expression and restoration of cell to cell communication in FRT cells transfected with the Cx43 cDNA
Three independent FRT clones have been selected on the basis of sustained and homogeneous expression of Cx43 mRNA and protein and maintenance of a high intercellular communication capacity over several passages. These characteristics are presented in Figs. 1Go and 2Go. Wild-type FRT cells (Fig. 1AGo, lane 2) and FRT cells cotransfected with the neomycin resistance gene and the empty pSVK3 vector (Fig. 1AGo, lane 3) were devoid of Cx43 transcript. Cx43 transcripts were detected in high amounts in FRT cells transfected with the pSVK3-Cx43 construct (Fig. 1AGo, lanes 4–6). The Cx43 mRNA detected in the three clones had a somewhat reduced size compared with that of the normal Cx43 mRNA from rat heart used as control (Fig. 1AGo, lane 1). This is probably due to the lack of a large part of the 3'-untranslated region and to a shorter polyadenylase tail in the transcript generated from the construct. The Cx43 protein detected by Western blot exhibited the expected molecular mass that corresponded to the 41-kDa unphosphorylated isoform of Cx43. The same isoform and similar amounts of Cx43 protein were detected in the three different clones (Fig. 1BGo, lanes 4–6). Cx43 detected by indirect immunofluorescence labeling appeared mainly as discontinuous lines delineating regions of cell-cell contacts (Fig. 1CGo, b–d). Bright dots were also found inside the cells. The intracellular labeling could correspond to Cx43 oligomers in the course of assembly and trafficking toward the plasma membrane. The level of expression of Cx43 remained stable beyond 20 passages.



View larger version (73K):
[in this window]
[in a new window]
  		Figure 1. Cx43 expression by transfected FRT cells. A, Northern blot analysis. Twenty-five micrograms of total RNA were analyzed in each lane. B, Western blot analysis of 100,000 x g membrane fractions (50 µg protein/lane) separated on 10% acrylamide gels. In A and B: Lane 1, rat heart (control); lane 2, wild-type FRT cells; lane 3, FRT cells expressing the neomycin resistance gene; lane 4, FRT-Cx43 cells (clone 2); lane 5, FRT-Cx43 cells (clone 3); lane 6, FRT-Cx43 cells (clone 5). C, Indirect immunofluorescence labeling of Cx43. Cells attached to petri dishes were fixed, permeabilized, and labeled using purified anti-Cx43 antibodies. a, Wild-type FRT cells; b, FRT-Cx43 cells (clone 2); c, FRT-Cx43 cells (clone 3); d, FRT-Cx43 cells (clone 5). Bar, 20 µm.

 


View larger version (57K):
[in this window]
[in a new window]
  		Figure 2. Restoration of intercellular communication by stable expression of Cx43 in FRT cells. A, Cell to cell communication in wild-type and transfected FRT cells. Lucifer yellow was microinjected into the cytoplasm of 1 cell (identified by an arrow), and the distribution of the fluorescent probe was examined 5 min after microinjection. a and b, Wild-type FRT cells; c and d, Cx43-transfected FRT cells (clone 2). a and c, Phase contrast images; b and d, fluorescence images of the corresponding fields. B, Quantitative analysis of the cell to cell transfer of lucifer yellow. a, Wild-type FRT cells; b, FRT cells expressing only the neomycin resistance gene; c–e, independent FRT clones expressing Cx43. Cells were microinjected at random with lucifer yellow. The number of fluorescent cells around the microinjected cell was determined 5 min after each series of injections. The number of dye-coupled cells varied from 0 (uncoupled cells) to more than 20 cells. Values were grouped into 3 classes (1–10, 11–20, and >20) and used to calculate the frequency of each level of dye coupling. For each cell type, the histogram of frequency was constructed from data obtained in 2–5 experiments. n, Total number of microinjected cells. Bar, 50 µm.

 
The three selected FRT-Cx43 clones exhibited a high level of GJ-mediated cell to cell communication compared with wild-type FRT cells (Fig. 2Go). When lucifer yellow was injected into a FRT-Cx43 cell, it was detected within seconds in a large number of cells surrounding the injected one (Fig. 2AGo, d). The level of intercellular communication assessed by the number of dye-coupled cells after a series of single cell microinjections is shown in Fig. 2BGo. The histogram of frequency shows that 98–100% of FRT-Cx43 cells were capable of transferring the fluorescent probe to adjacent cells and that at least 50% of them transferred the dye to more than 20 neighboring cells. Statistical analyses of data using the {chi}2 test showed that the communication level of Cx43-transfected cells was significantly higher than that of wild-type or neo-FRT cells (P < 0.01). These data establish that the Cx43 protein synthesized from Cx43 cDNA was competent to form functional GJ channels in FRT cells.

Increased propensity of FRT-Cx32 cells to form domes
While growing, wild-type FRT cells, as well as FRT-Cx32 and FRT-Cx43 cells, formed a monolayer and generated within the cell layer particular three-dimensional structures, previously termed domes (32). By varying the microscope focus, it could be seen that domes corresponded to domains where cells detached from the culture dish. Typical domes obtained with wild-type and Cx-transfected FRT cells are shown in Fig. 3Go. In a first series of experiments, it was found that the ability of cells to form domes and the stability of domes during a 2-week period were markedly different for FRT-Cx32 cells compared with those for either wild-type or FRT-Cx43 cells (Fig. 3Go). This led us to perform a detailed comparative analysis of the dynamic of formation and disappearance of domes with the 3 cell types. The data in Fig. 4Go show that the number of domes formed from wild-type FRT cells was maximum on day 3, then rapidly declined; domes were no longer observed beyond day 6 of culture. Control cells only expressing the neomycin resistance gene behaved as wild-type FRT cells. The formation of domes from cells expressing Cx43 also peaked on day 3 to reach a density of about 10 structures/mm2. Domes generated by FRT-Cx43 cells had a somewhat longer lifespan than those formed from wild-type cells. Some three-dimensional structures were still visible on day 11.



View larger version (219K):
[in this window]
[in a new window]
  		Figure 3. Morphological characteristics of wild-type and Cx-transfected FRT cells. Phase contrast micrographs of the three-dimensional structures or domes formed by wild-type FRT (a and d), FRT-Cx43 (b and e), and FRT-Cx32 (c and f) cells. Images were taken 3 days (a–c) or 8 days (d–f) after the outset of culture. Bar, 100 µm.

 


View larger version (22K):
[in this window]
[in a new window]
  		Figure 4. Quantitative analyses of the formation of three-dimensional structures by wild-type and transfected FRT cells. A, Evidence that FRT cells expressing Cx32 had a higher capacity of forming domes than either wild-type or FRT-Cx43 cells. Wild-type FRT ({blacktriangleup}), FRT-Cx43 (•), and FRT-Cx32 ({blacksquare}) cells were cultured for up to 16 days. The formation of three-dimensional structures was followed in the same 2–3 petri dishes throughout the culture period. B, Effect of (Bu)2cAMP on the formation of domes. (Bu)2cAMP (0.3 mM) was added to the culture medium 24 h after seeding (as indicated by the arrow), and domes were counted 4, 24, and 48 h later. Closed symbols, Untreated cells; open symbols; (Bu)2cAMP-treated cells. In both A and B, three-dimensional structures were counted in 10 different microscope fields/dish using phase contrast optics and the x32 objective. Symbols and vertical bars represent the mean ± SEM values obtained from 3 clones of FRT-Cx32 cells, 3 clones of FRT-Cx43 cells, and 3 independent cultures of wild-type FRT cells.

 
Cells expressing Cx32 exhibited a much higher capacity to form domes than either wild-type or Cx43-transfected cells. Domes appeared earlier, and their number progressively increased up to day 8 to reach a density about 3-fold higher than that observed within wild-type FRT or FRT-Cx43 cells monolayers. The three-dimensional structures generated from FRT-Cx32 cells presented a remarkable stability, as they were observed beyond 2 months of culture (data not shown). Noteworthy, the three independent clones exhibited similar properties. As it had been reported that the formation of domes (from polarized pig thyrocytes in primary culture) was increased in response to the activation of the cAMP cascade (32), we examined whether the cell-permeant dibutyryl derivative of cAMP could alter the generation of domes from FRT cells. Addition of 0.3 mM (Bu)2cAMP to the culture medium on day 1 did not modify the formation of domes by wild-type and FRT-Cx43 cells. By contrast, (Bu)2cAMP markedly activated the rate of formation of domes in the FRT-Cx32 cell monolayers (Fig. 4BGo). After 2 days of treatment, the number of domes was increased more than 2-fold (mean ± SEM, 44 ± 6 vs. 16 ± 2; P < 0.001). Similar results were obtained when 8Br-cAMP (1 mM) was used instead of (Bu)2cAMP (data not shown). Subsequent observations revealed that the (Bu)2cAMP-induced domes had the same stability as those spontaneously formed by FRT-Cx32 cells. Knowing that FRT cells were responsive to transforming growth factor-ß (TGFß) (our unpublished results), we analyzed the action of this cytokine on dome formation. Introduced in the culture medium 24 h after cell seeding, TGFß (1 ng/ml) totally inhibited the generation of domes by wild-type and Cx-transfected cells (data not shown).

Evidence for transformation of domes into closed spherical structures with a lumen
Careful phase contrast microscope observations of the domes developed in the FRT-Cx32 cell monolayers revealed that these three-dimensional structures were possibly undergoing transformations. Besides true domes under which there was no cell, we identified structures with an underlying cell layer in the plane of the cell monolayer (see a and b of Fig. 5AGo). Optical sections obtained by fluorescence confocal microscopy after immunofluorescence labeling of a cytoplasmic protein, tubulin, demonstrated that part of the three-dimensional structures were composed of a continuous cell layer delimiting a lumen. This is illustrated in Fig. 5BGo showing xy (a) and xz (b) optical sections through two neighboring structures; the left one had the morphological characteristics of a dome, whereas the right one was a follicle-like structure.



View larger version (129K):
[in this window]
[in a new window]
  		Figure 5. Cell organization within the three-dimensional structures formed by FRT-Cx32 cells after 8 days of culture. A, Phase contrast images of a given microscope field at two different focuses; focus was set either at the level of the epithelial cell layer attached to the culture dish (a) or at the top of the three-dimensional structures (b). B, Laser scan confocal microscope observations after tubulin immunolabeling. Optical sections were made through two neighboring three-dimensional structures: a, xy section (parallel to the surface of the dish); and b, xz section (perpendicular to the surface of the dish). The area under the three-dimensional structures shown on the left of a (A) and b (B) was devoid of cells, whereas there was a continuous cell layer under the three-dimensional structures shown on the right of the same panels. Bars, 100 µm.

 
The tightness of the lumen of the apparently closed structures was checked by microinjection of the membrane-impermeant fluorescent probe, lucifer yellow (Fig. 6Go). When a structure looking like a dome (under phase contrast optics) was microinjected, lucifer yellow immediately spread under the cell monolayer underlining intercellular spaces (a and b of Fig. 6Go). By contrast, lucifer yellow injected into the internal cavity of a structure with an underlying cell layer remained strictly located in this lumenal compartment. As illustrated in c and d of Fig. 6Go, lucifer yellow neither escaped under the cell monolayer nor diffused to the dome adjacent to the injected follicle. From a large series of microinjections, it was found that structures exhibiting an underlying cell layer (visualized by phase contrast optics) corresponded to closed follicle-like structures in more than 90% of the cases. Domes and closed structures were counted over a 17-day culture period of FRT-Cx32 cells. Figure 7AGo shows that the formation of closed follicles started 2 days after the initiation of dome formation, a fact in keeping with a dome to follicle transformation. After 2 weeks, follicles represented 50% of the total number of three-dimensional structures. Follicle-like structures have never been observed in wild-type or FRT-Cx43 cell cultures. The transformation of domes into follicles was not modified by activation of the cAMP cascade. Addition of (Bu)2cAMP to the medium of FRT-Cx32 cells cultured for 14 days caused an increase in dome formation within 24 h (28 ± 6 vs. 10 ± 5 structures/mm2 in controls; P < 0.05), but did not alter the number of follicles. TGFß, which is known to prevent the reconstitution of follicles from pig thyrocytes (33, 34), induced a 60% reduction of the total number of structures (4 ± 1 vs. 10 ± 1 in controls; P < 0.01).



View larger version (138K):
[in this window]
[in a new window]
  		Figure 6. Analysis of the tightness of the internal compartment of three-dimensional structures. Lucifer yellow was microinjected into the cavity of three-dimensional structures formed by FRT-Cx32 cells on day 8 of culture. a and c, Phase contrast images; b and d, corresponding fluorescence images taken 15 min after microinjection. The top panels show an open structure corresponding to a dome; the fluorescent probe diffused under the cell monolayer. The bottom panels identify a closed structure; the fluorescent probe remained strictly located inside the lumenal compartment. Arrows indicate the site of microinjection. Bar, 100 µm.

 


View larger version (23K):
[in this window]
[in a new window]
  		Figure 7. Quantitative assessment of the transformation of domes into closed follicle-like structures from FRT cells expressing Cx32. A, Time course of domes and follicle-like structures formation. At the indicated time after cell seeding, the number of three-dimensional structures without ({square}) or with ({blacksquare}) an underlying cell layer in the plane of the culture dish was counted in 10 different microscope fields for each of the 3 FRT-Cx32 cell clones. Symbols and vertical bars are the mean ± SEM number of structures expressed per mm2 in the three clones. B, Modulation of dome and follicle formation by (Bu)2cAMP and TGFß. On day 13, FRT-Cx32 cells were treated with (Bu)2cAMP (0.3 mM) and TGFß (1 ng/ml) for 24 h. At the end of treatment, the numbers of open structures or domes (open columns) and closed structures or follicles (closed columns) were determined as described above.

 
Follicles formed by FRT-Cx32 cells exhibit an inside-out polarity
The electron micrograph of Fig. 8aGo shows a semithin section through a follicle structure and the contiguous FRT-Cx32 cell monolayer. At higher magnification (on ultrathin sections), we could see that cells delimiting the lumen were all connected by tight junctions. Microvilli were located on the plasma membrane domains facing the culture medium (Fig. 8bGo), indicating that the closed structures formed from FRT-Cx32 cells corresponded to inside-out follicles. Deposits of extracellular matrix components were visible below the cells anchoring the follicle to the culture substratum. We identified cells insuring a continuity between the follicle structure and surrounding cells in the monolayer. The cell marked by an asterisk in c1 and d of Fig. 8Go, is attached to the culture dish, has established tight junctions with two adjacent cells belonging to the follicle structure, and, in addition, is joined through a tight junction to a contiguous cell of the monolayer. This interconnecting cell only exhibits microvilli on the plasma membrane domain facing the culture medium.



View larger version (115K):
[in this window]
[in a new window]
  		Figure 8. Ultrastructural characteristics of cells involved in the constitution of closed follicle-like structures from FRT cells expressing Cx32. Semithin (a) and ultrathin (b–d) sections were prepared from FRT-Cx32 cells after 8 days of culture. a, Micrograph showing a follicle with its lumen (L). b–d, Enlarged fields corresponding to frames b–d drawn in a. Note the presence of microvilli (closed arrows) on the plasma membrane domains in contact with the culture medium. Open large arrows identify tight junctions. c2 represents an enlargement of a tight junction from panel c1. The asterisk identifies a cell appearing in both c1 and d. Bars, 20 µm in a, 2 µm in b–d, and 0.5 µm in c2.

 
Steps in the transformation of a dome into a follicle structure
The generation of a closed structure from a dome first implies the filling of the free space underneath the dome by cells that will secondarily constitute the bottom part of the follicle. To determine whether cell migration was involved in this process, we choose to label cells at random in the cell monolayer by microinjection of FITC-dextran and to follow their positions over several days with respect to dome and then follicle formation. Among 30 cells that have been tracked, we definitely found 4 labeled cells becoming progressively part of the basal cell layer of follicle structures within 48 h. Such a case is illustrated in Fig. 9Go. Within 48 h after FITC-dextran injection, a cell, first positioned in the immediate vicinity of a neoformed dome, was later found underneath this structure that concurrently transformed into a closed follicle. It was noticed that the morphology, spreading, and position of cells were rapidly changing within the confluent cell monolayer.



View larger version (135K):
[in this window]
[in a new window]
  		Figure 9. Evidence for cell migration along the process of follicle formation. FITC-dextran (150 kDa) was introduced by microinjection into the cytoplasm of FRT-Cx32 cells randomly taken within the cell monolayer after 8 days of culture. Images of the same fluorescent cell were made 2 h (a–c), 24 h (d–f), and 48 h (g–i) after microinjection. a, d and g, Phase contrast images; b, e, and h, combined phase contrast and fluorescence images of the same fields; c, f, and i, corresponding fluorescence images. The fluorescent cell belongs to the underlying cell layer of a neoformed follicle. Bar, 100 µm.

 
Dividing cells within FRT-Cx32 cell cultures, near or at confluence, were visualized and quantified after BrdU incorporation. The proportion of labeled cells after a 4-h incubation with BrdU was the same in the monolayer, in domes, and in follicles (2.1%, 2.0%, and 1.8%, respectively), indicating that domes or follicles were not preferential sites of cell proliferation. After a longer period (12 h) of BrdU incorporation, BrdU-positive cells were found at the top, on the side, and at the bottom of follicle structures (Fig. 10Go). Even if there was no preferential cell division under domes, the space under domes was probably filled in, in part, by cell multiplication.



View larger version (140K):
[in this window]
[in a new window]
  		Figure 10. Location of FRT-Cx32 cells that divide after 8 days of culture. Cells near or at confluence were incubated with BrdU for 12 h. BrdU-positive nuclei were identified by immunofluorescence labeling with anti-BrdU antibodies. Nuclei were stained with the Hoescht reagent. a–e, Images from the same microscope field; a, phase contrast image; b and d, Hoescht fluorescence images taken at two different planes (by varying the microscope focus) at the level of cells attached to the culture dish (b) and at the top of a follicle-like structure (d). c and e, Fluorescence images of BrdU-labeled nuclei taken at the two levels mentioned above. Arrowheads identify BrdU-labeled nuclei at the bottom (c and b) and at the top (e and d) of a three-dimensional structure. Bar, 100 µm.

 

   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
The data presented here extend previous findings showing that Cx32 expression by thyrocytes was closely related to the organization of cells into thyroid follicles (17, 19, 20, 21). Using thyroid-derived FRT cells exhibiting a polarized phenotype forced to express Cx32 by stable transfection, we bring direct experimental evidence for a functional link between the formation of Cx32 GJ and/or Cx32 GJ-mediated intercellular communication and the reorganization of cells into three-dimensional follicle-like structures. The fact that Cx43 GJ formation and/or Cx43 GJ-mediated intercellular communication did not promote similar morphological changes emphasizes differences in the functional impact of these two Cx that are normally expressed by thyroid cells.

Follicle-like structures generated from FRT cells expressing Cx32 are derived from more simple structures, called domes, that are commonly observed with polarized epithelial cells in culture (24, 35, 36). The formation of domes originates from the accumulation of fluid between the polarized cell layer and the culture support that leads to the local detachment of the continuous tight cell layer from the culture substratum. The fluid accumulation probably results from an imbalance between opposite transepithelial transport of Na+ from the apical to the basolateral side of the epithelial cell layer and of Cl- from the basolateral to the apical compartment (37, 38). As a consequence of the coupling between ion and water transports, there is a net accumulation of water at the basolateral side of the epithelial barrier. In vivo, such a process would probably participate in regulation of the volume of the follicles and in control of the osmotic pressure inside the follicle lumen containing thyroglobulin at a very high concentration. Under culture conditions generally used to propagate FRT cells, both wild-type and Cx-transfected FRT cells, while growing, rapidly formed domes. Dome formation was activated in Cx32-expressing FRT cells in response to a treatment by cAMP derivatives. There is no clear reason why a similar effect was not observed in wild-type and Cx43-expressing FRT cells. As the two cAMP derivatives, (Bu)2cAMP and 8Br-cAMP were similarly ineffective, one cannot attribute this lack of effect to differences in plasma membrane permeability between FRT-Cx32 cells and wild-type and Cx43-transfected FRT cells (39). Furthermore, extracellular concentrations of 0.3 mM (Bu)2cAMP or 1 mM 8Br-cAMP are expected to elicit an elevation of the intracellular cAMP concentration, high enough to activate subsequent steps of the cAMP pathway in each cell, independently of GJ-mediated cell to cell transfer of the second messenger. It has been reported that FRT cells respond to 8Br-cAMP by increasing their endocytic activity (40); thus, these cells should possess activatable protein kinase A. The differential response of FRT-Cx32 cells might be related to a change in the responsiveness of cAMP downstream effectors. Alternately, the ability of FRT cells expressing Cx32 to respond to cAMP might indicate that these cells have acquired a new functional property (compared with wild-type FRT cells) subjected to regulation by activation of the cAMP cascade. The stimulation of dome formation via activation of the cAMP cascade by TSH could not be analyzed because FRT cells are devoid of TSH receptors. As Cx43 GJ channels and Cx32 GJ channels exhibit different permeability properties, it is worth considering that cAMP-activated dome formation could involve the production of a factor transferred from cell to cell through Cx32 GJ and not through Cx43 GJ. Dome formation by wild-type, Cx32-transfected, and Cx43-transfected FRT cells was inhibited by TGFß. This is in keeping with the inhibitory effect of TGFß on in vitro reconstitution of follicles from pig thyrocytes in primary culture (33, 34). This cytokine is capable of inhibiting the production of a matricellular protein, thrombospondin-1, that exerts a negative action on folliculogenesis from pig thyrocytes through a possible interference with cell-cell associations. Accordingly, TGFß could inhibit dome formation by FRT cells and stable transfectants by altering cell-cell interactions and the tightness of the cell layer, which would prevent fluid accumulation and, thus, the initial stages of dome constitution.

Wild-type FRT, FRT-Cx43, and FRT-Cx32 cells are distinguishable by the number of domes and the subsequent transformations of the three-dimensional structures to which they give rise. The fact that FRT-Cx32 cells form a higher number of domes compared with either wild-type or FRT-Cx43 cells might be explained by an increase in dome stability. Indeed, domes formed by FRT and FRT-Cx43 cells, disappearing with time, cannot accumulate and cannot transform into follicle-like structures.

The transformation of domes into closed follicle-like structures probably brings into play a complex series of cellular events. Initiation of the process would consist in the expansion of an adherent cell layer underneath the dome. This could occur through cell migration and/or cell multiplication, as tentatively presented in Fig. 11Go. We demonstrate that cells outside a dome can migrate under the dome and participate in the formation of the underlying cell layer. The implication of cell multiplication, although conceptually plausible, was far more difficult to document. Indeed, FRT-Cx32 cells even at confluence for several days had a residual proliferation activity; cells undergoing division were observed at the base or under the domes, but also in the cell wall of the domes and all over the cell population constituting the monolayer. As spaces under the domes were not preferential sites of cell proliferation, it is conceivable that cell multiplication, in the vicinity of a dome or at some distance from it, could indirectly cause cell migration toward the interior of the domes. Closed three-dimensional structures formed from domes exhibited an inverse polarity compared with that of in vivo follicles. Such inside-out follicles have been generated from freshly dispersed thyrocytes cultured in suspension (41). It was reported that these inside-out follicles can be converted into inside-in follicles, i.e. normal follicles, by an additional culture in a type I collagen matrix (10, 11, 12, 42). Such a reversal of cell polarity was not obtained when FRT-Cx32 inside-out follicles were cultured in the presence of type I collagen; instead, we observed a disruption of follicles (data not shown), probably due to the opening of tight junctions. Garbi et al. (25) reported that treatment of FRT cells with type I collagen induces a dramatic loss of transepithelial resistance due to the interaction of type I collagen fibers with ß1 integrin present on the apical plasma membrane, which causes major modifications of the actin filament network.



View larger version (120K):
[in this window]
[in a new window]
  		Figure 11. Schematic representation of the formation of an inside-out follicle from a dome. Two processes, cell proliferation and cell migration, might be involved in the constitution of the bottom cell layer, leading to closure of the three-dimensional structure. Arrowheads identify cells that could migrate from the side of a dome to either the upper or the lower part of the structure in the course of closure. Thin arrows identify possible sites of cell proliferation.

 
Among the two Cx studied in this work, only Cx32 made FRT cells competent to form follicles. Similarly, stable expression of Cx32 by another thyroid-derived cell line, the FRTL5 cell line, has been found to promote expression of differentiation. The expression of the gene encoding thyroglobulin was increased in FRTL5 cells expressing Cx32 (26). Unlike FRT cells, FRTL5 cells are not polarized, and stable expression of Cx32 did not confer upon these cells the capacity to form three-dimensional structures. The reason why stable expression of Cx43 does not reproduced the phenotypic changes induced by stable expression of Cx32 in FRT cells remains uncertain. This could be linked to the fact that Cx32 GJ channels and Cx43 GJ channels are not permeable to the same molecules (43). In addition to their basic function, i.e. cell to cell communication, GJ might have other roles. As intercellular junctions, they might be involved in the setting and/or the strengthening of cell-cell contacts (44, 45). Cx32 GJ and Cx43 GJ could have here a differential impact, because we know that the two types of GJ have a distinct location in the regions of contact between thyrocytes (17). Finally, although highly homologous in their main functional domains, i.e. transmembrane segments and extracellular loops, Cx32 and Cx43 differ in their C-terminal part. Consequently, the two Cx might interact with distinct cytoplasmic proteins and initiate separate cell responses. It has recently been reported that Cx43 interacts with the tight junction-associated protein, ZO1 (46, 47). Progress in the knowledge of Cx functions, other than those related to GJ-mediated cell to cell communication, in different cell systems should provide explanations for the selective role of Cx32 in the three-dimensional organization of thyrocytes that emerges from previous studies (17, 19, 20, 21) and the present findings.


   Acknowledgments
 
We thank Stéphane Ory, Ecole Normale Superieure de Lyon (Lyon, France), for his help with the confocal microscope studies, and Prof. G. I. Fishman, Albert Einstein College of Medecine (Bronx, NY), for providing us with the hCx43 cDNA.


   Footnotes
 
1 This work was supported by Grant 9161 from Association pour la Recherche sur le Cancer. Back

2 H.T. and V.F. equally contributed to this work. Back

3 Present address: Unità di Patologia Generale, Facultà di Scienze, Università degli Studi dell’Insubria, 21100 Varese, Italy. Back

Received August 4, 1999.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Gumbiner BM 1992 Epithelial morphogenesis. Cell 69:385–387[Medline]
  2. Rodriguez-Boulan E, Nelson WJ 1989 Morphogenesis of the polarized epithelial cell phenotype. Science 245:718–725[Abstract/Free Full Text]
  3. Wang AZ, Ojakian GK, Nelson WJ 1990 Steps in the morphogenesis of a polarized epithelium. I. Uncoupling the roles of cell-cell and cell-substratum contact in establishing plasma membrane polarity in multicellular epithelial (MDCK) cysts. J Cell Sci 95:137–151[Abstract/Free Full Text]
  4. Wang AZ, Ojakian GK, Nelson WJ 1990 Steps in the morphogenesis of a polarized epithelium. II. Disassembly and assembly of plasma membrane domains during reversal of epithelial cell polarity in multicellular epithelial (MDCK) cysts. J Cell Sci 95:153–165[Abstract/Free Full Text]
  5. Simons K, Wandinger-Ness A 1990 Polarized sorting in epithelia. Cell 62:207–210[CrossRef][Medline]
  6. Zurzolo C, Le Bivic A, Quaroni A, Nitsch L, Rodriguez-Boulan E 1992 Modulation of transcytotic and direct targeting pathways in a polarized thyroid cell line. EMBO J 11:2337–2344[Medline]
  7. Dumont JE 1971 The action of thyrotropin on thyroid metabolism. Vitam Horm 29:287–412[Medline]
  8. Bidey SP, Tomlinson S 1988 The regulation and integration of thyroid follicular differentiation and function. Clin Endocrinol (Oxf) 28:423–444[Medline]
  9. Mauchamp J, Mirrione A, Alquier C, Andre F 1998 Follicle-like structure and polarized monolayer: role of the extracellular matrix on thyroid cell organization in primary culture. Biol Cell 90:369–380[CrossRef][Medline]
  10. Chambard M, Gabrion J, Mauchamp J 1981 Influence of collagen gel on the orientation of epithelial cell polarity: follicle formation from isolated thyroid cells and from preformed monolayers. J Cell Biol 91:157–166[Abstract/Free Full Text]
  11. 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]
  12. 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]
  13. Yap AS, Stevenson BR, Keast JR, Manley SW 1995 Cadherin-mediated adhesion and apical membrane assembly define distinct steps during thyroid epithelial polarization and lumen formation. Endocrinology 136:4672–4680[Abstract]
  14. Bennett MVL, Barrio LC, Bargiello TA, Spray DC, Hertzberg E, Saez JC 1991 Gap junctions: new tools, new answers, new questions. Neuron 6:305–320[CrossRef][Medline]
  15. Bruzzone R, White TW, Paul DL 1996 Connections with connexins: the molecular basis of direct intercellular signaling. Eur J Biochem 238:1–27[Medline]
  16. Simon AM, Goodenough DA 1998 Diverse functions of vertebrate gap junctions. Trends Cell Biol 8:477–483[CrossRef][Medline]
  17. Guerrier A, Fonlupt P, Morand I, Rabilloud R, Audebet C, Krutovskikh V, Gros D, Rousset B, Munari-Silem Y 1995 Gap junctions and cell polarity: connexin 32 and connexin 43 expressed in polarized thyroid epithelial cells assemble into separate gap junctions, which are located in distinct regions of the lateral plasma membrane domain. J Cell Sci 108:2609–2617[Abstract]
  18. Meda, P, Pepper MS, Traub O, Willecke K, Gros D, Beyer E, Nicholson B, Paul D, Orci L 1993 Differential expression of gap junction connexins in endocrine and exocrine glands. Endocrinology 133:2371–2378[Abstract]
  19. Green, LM, Lazarus JP, LaBue M, Shah MM 1995 Reduced cell-cell communication in a spontaneous murine model of autoimmune thyroid disease. Endocrinology 136:3611–3618[Abstract]
  20. Green, LM, LaBue M, Lazarus JP, Jennings JC 1996 Reduced cell-cell communication in experimentally induced autoimmune thyroid disease. Endocrinology 137:2823–2832[Abstract]
  21. Munari-Silem Y, Guerrier A, Fromaget C, Rabilloud R, Gros D, Rousset B 1994 Differential control of connexin-32 and connexin-43 expression in thyroid epithelial cells: evidence for a direct relationship between connexin-32 expression and histiotypic morphogenesis. Endocrinology 135:724–734[Abstract]
  22. Green, LM, Lazarus JP, Song X, Stagg RB, LaBue M, Hilliker S 1997 Elevation of protein kinase C in thyrocytes isolated from a Lewis rat model of autoimmune thyroiditis prevents assembly of immunodetectable connexin43 gap junctions and reduces intercellular communication. Thyroid 7:913–921[Medline]
  23. Mascia, A, De Felice M, Lipardi C, Gentile R, Cali G, Zannini M, Di Lauro R, Nitsch L 1997 Transfection of TTF-1 gene induces thyroglobulin gene expression in undifferentiated FRT cells. Biochim Biophys Acta 1354:171–181[Medline]
  24. Nitsch L, Tramontano D, Ambesi-Impiombato FS 1985 Morphological and functional polarity of an epithelial thyroid cell line. Eur J Cell Biol 38:57–66[Medline]
  25. Garbi C, Negri R, Cali G, Nitsch L 1996 Collagen interaction with apically expressed beta1 integrins: loss of transepithelial resistance and reorganization of cultured thyroid cell monolayer. Eur J Cell Biol 69:64–75[Medline]
  26. Statuto M, Audebet C, Tonoli H, Selmi-Ruby S, Rousset B, Munari-Silem Y 1997 Restoration of cell-to-cell communication in thyroid cell lines by transfection with and stable expression of the connexin-32 gene. J Biol Chem 272:24710–24716[Abstract/Free Full Text]
  27. Ambesi-Impiombato FS, Coon HG 1979 Thyroid cells in culture. Int Rev Cytol 10:163–172
  28. Fishman GI, Spray DC, Leinwand LA 1990 Molecular characterization and functional expression of the human cardiac gap junction channel. J Cell Biol 111:589–598[Abstract/Free Full Text]
  29. Munari-Silem Y, Audebet C, Rousset B 1991 Hormonal control of cell-to-cell communication: regulation by thyrotropin of the gap junction-mediated dye transfer between thyroid cells. Endocrinology 128:3299–3309[Abstract]
  30. Bastide B, Jarry-Guichard T, Briand JP, Deleze J, Gros D 1996 Effect of antipeptide antibodies directed against three domains of connexin43 on the gap junctional permeability of cultured heart cells. J Membr Biol 150:243–253[CrossRef][Medline]
  31. Munari-Silem Y, Mesnil M, Selmi S, Bernier-Valentin F, Rabilloud R, Rousset B 1990 Cell-cell interactions in the process of differentiation of thyroid epithelial cells into follicles: a study by microinjection and fluorescence microscopy on in vitro reconstituted thyroid follicles. J Cell Physiol 145:414–427[CrossRef][Medline]
  32. Bourke JR, Matainaho T, Huxham GJ, Manley SW 1987 Cyclic AMP-stimulated fluid transport in the thyroid: influence of thyroid stimulators, amiloride and acetazolamide on the dynamics of domes in monolayer cultures of porcine thyroid cells. J Endocrinol 115:19–26[Abstract]
  33. Toda S, Matsumura S, Fujitani N, Nishimura T, Yonemitsu N, Sugihara H 1997 Transforming growth factor ß-1 induces a mesenchyme-like cell shape without epithelial polarization in thyrocytes and inhibits thyroid folliculogenesis in collagen gel culture. Endocrinology 138:5561–5575[Abstract/Free Full Text]
  34. Pellerin S, Croizet K, Rabilloud R, Feige JJ, Rousset B 1999 Regulation of the three-dimensional organization of thyroid epithelial cells into follicle structures by the matricellular protein, thrombospondin-1. Endocrinology 140:1094–1103[Abstract/Free Full Text]
  35. Mauchamp J, Margotat A, Chambard M, Charrier B, Remy L, Michel-Bechet M 1979 Polarity of three-dimensional structures derived from isolated hog thyroid cells in primary culture. Cell Tissue Res 204:417–430[Medline]
  36. Tanner C, Frambach DA, Misfeldt DS 1984 Biophysics of domes formed by the renal cell line Madin-Darby canine kidney. Fed Proc 43:2217–2220[Medline]
  37. Yap AS, Armstrong JW, Cragoe EJ, Bourke JR, Huxham GJ, Manley SW 1991 Regulation of thyroid follicular volume by bidirectional transepithelial ion transport. Mol Cell Endocrinol 82:R1–R5
  38. Yap AS, Stevenson BR, Armstrong JW, Keast JR, Manley SW 1994 Thyroid epithelial morphogenesis in vitro: a role for bumetanide-sensitive Cl- secretion during follicular lumen development. Exp Cell Res 213:319–326[CrossRef][Medline]
  39. Sammak, PJ, Adams SR, Harootunian AT, Schliwa M, Tsien RY 1992 Intracellular cyclic AMP not calcium, determines the direction of vesicle movement in melanophores: direct measurement by fluorescence ratio imaging. J Cell Biol 117:57–72[Abstract/Free Full Text]
  40. Hansen, SH, Olsson A, Casanova JE 1995 Wortmannin, an inhibitor of phosphoinositide 3-kinase, inhibits transcytosis in polarized epithelial cells. J Biol Chem 270:28425–28432[Abstract/Free Full Text]
  41. Herzog V, Miller F 1981 Structural and functional polarity of inside-out follicles prepared from pig thyroid gland. Eur J Cell Biol 24:74–84[Medline]
  42. Chambard M, Verrier B, Gabrion J, Mauchamp J 1984 Polarity reversal of inside-out thyroid follicles cultured within collagen gel: reexpression of specific functions. Biol Cell 51:315–326[Medline]
  43. Elfgang C, Eckert R, Lichtenberg-Fraté H, Butterweck A, Traub O, Klein RA, Hülser DF, Willecke K 1995 Specific permeability and selective formation of gap junction channels in connexin-transfected HeLa cells. J Cell Biol 129:805–817[Abstract/Free Full Text]
  44. Yeager M, Nicholson BJ 1996 Structure of gap junction intercellular channels. Curr Opin Struct Biol 6:183–192[CrossRef][Medline]
  45. Perkins G, Goodenough D, Sosinsky G 1997 Three-dimensional structure of the gap junction connexon. Biophys J 72:533–544[Medline]
  46. Giepmans BNG, Moolenaar WH 1998 The gap junction protein connexin43 interacts with the second PDZ domain of the zona occludens-1 protein. Curr Biol 8:931–934[CrossRef][Medline]
  47. Toyofuku T, Yabuki M, Otsu K, Kuzuya T, Hori M, Tada M 1998 Direct association of the gap junction protein connexin-43 with ZO-1 in cardiac myocytes. J Biol Chem 273:12725–12731[Abstract/Free Full Text]



This article has been cited by other articles:


Home page
 Am. J. Physiol. Endocrinol. Metab.Home page
G. Prost, F. Bernier-Valentin, Y. Munari-Silem, S. Selmi-Ruby, and B. Rousset
Connexin-32 acts as a downregulator of growth of thyroid gland
Am J Physiol Endocrinol Metab, February 1, 2008; 294(2): E291 - E299.
[Abstract] [Full Text] [PDF]

Home page
 Endocr Relat CancerHome page
K. Fujarewicz, M. Jarzab, M. Eszlinger, K. Krohn, R. Paschke, M. Oczko-Wojciechowska, M. Wiench, A. Kukulska, B. Jarzab, and A. Swierniak
A multi-gene approach to differentiate papillary thyroid carcinoma from benign lesions: gene selection using support vector machines with bootstrapping
Endocr. Relat. Cancer, September 1, 2007; 14(3): 809 - 826.
[Abstract] [Full Text] [PDF]

Home page
 Mol. Endocrinol.Home page
T. Kusakabe, A. Kawaguchi, N. Hoshi, R. Kawaguchi, S. Hoshi, and S. Kimura
Thyroid-Specific Enhancer-Binding Protein/NKX2.1 Is Required for the Maintenance of Ordered Architecture and Function of the Differentiated Thyroid
Mol. Endocrinol., August 1, 2006; 20(8): 1796 - 1809.
[Abstract] [Full Text] [PDF]

Home page
 J. Biol. Chem.Home page
V. Shlyonsky, A. Goolaerts, R. Van Beneden, and S. Sariban-Sohraby
Differentiation of Epithelial Na+ Channel Function: AN IN VITRO MODEL
J. Biol. Chem., June 24, 2005; 280(25): 24181 - 24187.
[Abstract] [Full Text] [PDF]

Home page
 EndocrinologyHome page
S. J. Winters, A. C. Dalkin, and T. Tsujii
Evidence That Pituitary Adenylate Cyclase Activating Polypeptide Suppresses Follicle-Stimulating Hormone-{beta} Messenger Ribonucleic Acid Levels by Stimulating Follistatin Gene Transcription
Endocrinology, October 1, 1997; 138(10): 4324 - 4329.
[Abstract] [Full Text] [PDF]

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
C