| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Division of Endocrinology, Department of Medicine, St. Hedwig Hospital and Humboldt University, D-10115 Berlin, Germany
Address all correspondence and requests for reprints to: Professor Dr. Michael Derwahl, Endocrine Research Laboratory, Department of Medicine, St. Hedwig Hospital, Grosse Hamburger Str. 5-11, D-10115 Berlin, Germany. E-mail: m.derwahl{at}alexius.de.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Design: Stem cells were isolated by fluorescence-activated cell sorting as a side population by the Hoechst 33342 efflux technique. Growth pattern of stem cells and cocultures of stem cells with thyrocytes grown as monolayer and in Matrigel was investigated. Expression of stem cell markers, endodermal markers, and thyroid-specific markers was analyzed by RT-PCR. In stem cell-derived thyrocytes, embedded in collagen to form follicles, TSH-dependent 125iodide uptake was measured.
Results: Stem cells were isolated as a side population from a non-side population fraction that consisted of endodermal marker-positive cells and thyroid cells. Intense growth stimulation of stem cells in coculture with thyrocytes resulted in formation of nonadherent, three-dimensional spheres that consisted of highly proliferating stem cells with their characteristic expression profiles. In response to TSH and serum, sphere-derived progenitor cells differentiated into thyrocytes that expressed paired box gene 8, thyroglobulin, sodium iodide symporter, thyroid-stimulating hormone receptor, and thyroperoxidase mRNA and showed TSH-dependent 125iodide uptake.
Conclusion: Quiescent stem cells derived from goiters can be propagated to form spheres that consist of highly proliferating stem cells that are able to differentiate TSH dependently into thyroid cells. Compared with thyrocytes, stem cells display a much higher proliferation rate on acute growth stimulation, which may suggest a putative role of the offspring of stem cells in the chronic growth factor-stimulated nodular transformation of the thyroid.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
At first glance, the high frequency of tumor formation in the thyroid gland is surprising because, compared with highly proliferating tissues such as the colon, breast, skin, or prostate, the growth rate of human thyroid cells is considerably lower. It has been estimated that human thyrocytes divide only about five times during adulthood, which corresponds to a turnover time of about 8.5 yr for the follicular thyroid cell (5). Furthermore, tissues with high cell turnover such as the colon are more sensitive to mutagenesis and other molecular mechanisms that initiate tumor formation, whereas in resting tissues such as the thyroid, such mechanisms should be less operative (6). To explain this discrepancy, it has been suggested that free radicals resulting from reactive oxygen species in the thyroid generate mutations more frequently (6).
As an alternative source of benign and malignant tumor formation, adult stem cells are suggested (7). In fact, stem cells have been detected in both low and high proliferation tissues such as colon, skin, pancreas, liver, and brain (8, 9, 10, 11, 12). Very recently we identified stem cells in the human thyroid gland (13), and stem-like cells derived from thyroid cancer cell lines have been described by Mitsutake et al. (14). Adult stem cells are capable of self-renewal and differentiation throughout the organism’s lifetime. Due to their pluripotency and undifferentiated state, they are widely believed to be involved in the pathogenesis of tumors (15, 16). This concept has also recently been hypothesized for thyroid carcinogenesis (17, 18).
In the present work, we show that adult stem cells can be separated by fluorescence-activated cell sorting (FACS) from differentiated thyrocytes and endodermal marker-positive cells as a side population due to their ability to express ABCG2 transporters, a characteristic of embryonic and adult stem cells (19, 20). Intensive growth stimulation of stem cells in coculture with normal thyrocytes resulted in the formation of nonadherent, three-dimensional spheres that comprised highly proliferating stem cells. Under the influence of serum and TSH, stem cells differentiated into thyrocytes that expressed paired box gene 8 (PAX8), thyroglobulin (TG), sodium iodide symporter (NIS), thyroid-stimulating hormone receptor (TSHr), and thyroperoxidase (TPO) mRNA. Embedded in collagen, cells showed the ability to uptake 125iodide in response to TSH. The propagation of highly proliferating stem cells in response to acute intense growth stimulation resembles in some respects the chronic growth stimulation in nodular goiters, which may point to a pathogenetic role of the offspring of stem cells in this disease.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Monolayer cultures of human thyrocytes isolated from nodular goiters of 30 patients after thyroidectomy were established as described previously (21). Of these patients, who were aged 50–80 yr, 17 were females and 13 males. In all cases, informed consent was obtained; the study was approved by the ethical committee of Charité, Humboldt University. Extreme care was taken in removing connective tissues, and a short-term preplating step was used to get rid of the earliest attached fibroblasts to minimize the probability of fibroblast contamination in the primary thyroid cell cultures. Then human thyrocytes were cultured in Ham’s F-12 medium (Invitrogen, Carlsbad, CA), supplemented with 10% fetal calf serum (FCS; Invitrogen), 1% MEM, 5 mU/ml TSH (Sigma, St. Louis, MO), and five hormones as described previously (22). Monolayer cultures of Fischer rat thyroid cell line (FRTL)-5, obtained from American Type Culture Collection (Bethesda, MD), were grown in the same medium but supplemented with 5% FCS.
FACS
To isolate the thyroid side population, FACS was performed using Hoechst staining as outlined by Goodell et al. (23). Briefly, thyrocytes of second passage were harvested and suspended at 106 cells/ml in DMEM containing 2% FCS and 10 mM HEPES and preincubated at 37 C for 10 min. Then the cells were labeled in the same medium with 5 µg/ml Hoechst 33342 at 37 C for 120 min with periodic agitation, either alone or in combination with 50 µM verapamil (Sigma), an inhibitor of ABCG2 transporter (23). Finally, the cells were centrifuged, resuspended in cold Hanks balanced salt solution containing 2% FCS and 10 mM HEPES, counterstained with 2 µg/ml propidium iodide (PI) to exclude dead cells, and kept at 4 C until sorting. To remove cellular aggregates, cells were filtered through a 30-µm strainer (Falcon, Oxnard, CA) before analysis.
A 350-nm UV laser was used to excite Hoechst 33342 and PI. Analysis was performed on fluorescence-activated cell sorter (BD Biosciences, Franklin Lakes, NJ) by using a dual-wavelength analysis (blue, 424–444 nm; red, 675 nm) in the facilities of the German Rheumatology Research Center (Berlin, Germany). PI-positive dead cells (<15%) were excluded from the analysis. The side population was identified and selected by gating on the characteristic fluorescence emission profile. Equal numbers of side population and non-side population cells were collected for culture and pelleted for RNA isolation or cytospun onto slides for cytological analysis.
Growth and differentiation of nonadherent spheres from primary thyroid cultures
Human thyroid cell cultures were prepared as described above. The cell suspension was sieved through a 30-µm strainer, centrifuged, and resuspended in growth factor-enriched medium: serum-free DMEM/Ham’s F-12 (1:1) containing B-27 (1:50), 20 ng/ml epidermal growth factor (EGF; Invitrogen), and 20 ng/ml basic fibroblast growth factor (bFGF; Invitrogen). Single cellularity was confirmed under microscope. Cells were cultured at 10,000 viable cells/ml. Under these conditions, most primary thyrocytes did not grow. Only a small number of cells survived and generated floating spheres after 5 d of culture, which we termed thyrospheres.
To explore the differentiation potential of sphere-forming cells, all thyrospheres were collected and dissociated enzymatically (15 min in 0.05% trypsin, 0.53 mM EDTA-4Na at 37 C). The sphere numbers in every dish and the average cell counts in every sphere were analyzed. Thyrosphere differentiation induction was carried out using a two-step protocol. In the first step, cell differentiation was initiated in Ham’s F-12 medium supplemented with 10% FCS at a density of 1000 cells/ml for 3 d. Within 24 h after plating, almost all cells attached. In the second step, 5 mU/ml TSH were added to the medium and the cells were maintained as monolayer in culture for additional 18 d. During the experiment, cells were harvested at d 0, 3, 6, 9, 15, and 21 for gene expression analysis.
To test the correlation between sphere-forming cells and side population, sphere-derived cells were analyzed by FACS for the capacity of Hoechst dye exclusion. Sorted side population and non-side population cells from spheres were analyzed for gene expression.
5-Bromo-2'-deoxyuridine (BrdU) incorporation in thyrosphere cells
To test the proliferative potential of thyrosphere cells, BrdU was added at a final concentration of 10 µM to the medium 1 d after seeding. On d 5, thyrospheres were fixed with ethanol fixative and further analyzed for BrdU incorporation using BrdU Labeling and Detection Kit I (Roche, Stockholm, Sweden) according to the manufacturer’s instructions. Samples were then mounted in Vectashield (Vector Laboratories, Burlingame, CA) and observed under fluorescence microscope.
The proliferation rates of sphere cells that underwent intense growth stimulation and differentiation were analyzed as follows: after 5 d of sphere formation (d 1–5), spheres were transferred to differentiating conditions with both serum and TSH and cultured for another 5 d (d 6–10). During the sphere formation and early differentiation, BrdU was added every day in the respective media 12 h before each assay. Proliferation rates were then estimated by counting the percentage of BrdU-labeled cells. For each experiment at least 300 nuclei were counted.
Gene expression analysis by semiquantitative RT-PCR
Total RNA was extracted from cultured cells using RNeasy minikit (QIAGEN, Valencia, CA) according to the manufacturer’s specifications. RT-PCR was performed as follows: 1 µg of total RNA was added to a 25-µl reaction mixture containing 5 µl reverse transcription buffer, 1.25 µl 10 mmol/liter deoxynucleotide triphosphates, 1 µl RNAGuard, 2 µl primer p(dT)15, and 1 µl reverse transcriptase. For PCR amplification, 4 µl cDNA were added to a 50-µl reaction containing 5 µl 10x reaction buffer, 1.5 mM MgCl2, 1 liter deoxynucleotide triphosphates, 30 mol sense and antisense primers, and 2.5 U Taq polymerase. Reactions were carried out at 95 C for 10 min; 30–35 cycles at 95 C for 30 sec, 53–59 C (primer specific) for 30 sec, and 72 C for 1 min, followed by an extension at 72 C for 10 min and termination at 4 C. All cDNAs were amplified in the log-linear phase of PCR by 30–35 cycles.
After FACS, comparable numbers of side population and non-side population cells were collected for total RNA extraction using RNeasy microkit (QIAGEN), and RT-PCR was performed as described above.
In all PCR analyses, the amount of cDNA in each sample was normalized using ß-actin as an internal control. Primer sequences, product sizes and annealing temperatures are listed in Table 1
. The PCR products were separated by electrophoresis on 1.5% agarose gels stained with ethidium bromide. The signal intensities of gene bands were analyzed with Image J software (version 1.34s; National Institutes of Health, Bethesda, MD).
|
During differentiation of thyrosphere cells into thyrocytes, sphere-derived cells were grown as monolayer in the presence of serum and TSH as described above and then embedded in collagen at a concentration of 2 x 105 cells per 20 µl in 24-well plates as described by Chambard et al. (24). To measure the iodide uptake activity of follicles developed in collagen after 3–4 d of culture, cells were washed with Hanks balanced salt solution and incubated in the same buffer containing 125I (0.1 µCi) and NaI (0.5 µmol/liter) for 40 min at 37 C in the absence or presence of sodium perchlorate. Then the collagen was washed and transferred to Ep tubes. The radioactivity was measured in a 
-counter as previously described (25). 125I-uptake of thyrosphere cells that were maintained in the absence of TSH and serum for 21 d and normal cultured thyrocytes from the same patients was also measured. FRTL-5 cells were used as a positive control. 125I-uptake was expressed as counts per minute per well. The values represented the mean ± SE of three independent experiments performed in triplicate.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Adult stem cells derived from human goiters were separated by FACS from differentiated thyrocytes and endodermal marker-positive cells as a side population due to their ability to express ABCG2 transporters. As depicted in Fig. 1A, side population-gated cells accounted for about 0.1% of total cells. Incubation with verapamil almost completely abolished the side population profile (Fig. 1B).
|
, C–F). Both hallmarks are typical for quiescent adult stem cells (26, 27).
Analysis of expression profiles by semi quantitative RT-PCR (Fig. 1G) revealed that side population cells strongly expressed ATP binding cassette transporter subfamily G member 2 (ABCG2) and octamer transcription factor (Oct)-4, a major transcription factor for embryonic and some adult stem cells (9, 13, 28, 29, 30). However, ABCG2 and Oct4 were almost undetectable in non-side population cells. In contrast, side population cells were negative for thyroid differentiation markers. As expected, the non-side population of thyroid cells expressed the typical differentiation markers (TG, NIS, TSHr and TPO). PAX8, which is crucial for early thyroid development, was detectable only in non-side population cells. Interestingly, some non-side population cells also expressed endodermal markers, the GATA binding protein 4 (GATA4) and hepatocyte nuclear factor (HNF)-4
transcription factors (31, 32, 33) that are not expressed in differentiated thyroid cells (13). This indicates that the side population fraction of cells consists of a very small percentage of undifferentiated stem cells, whereas the major fraction of cells contains a heterogeneous cell population of differentiated thyrocytes and endodermal marker-positive cells that have already lost expression of ABCG2 transporters.
Viable adult stem cells isolated as a side population were maintained in culture for up to 14 d. However, neither cell attachment nor growth was observed in monolayer culture or in Matrigel. A significant proliferation rate was also absent when side population cells were cocultured with normal thyrocytes in a two-chamber culture system, even under intense growth stimulation with EGF and bFGF (data not shown).
Thyroid stem cells proliferate as spheres in response to intense growth stimulation and require TSH for differentiation into thyroid cells
In a complementary approach, single cell suspensions of primary human thyrocytes derived from nodular goiters were grown in a special medium that does not allow adherence to a substratum. Most thyrocytes did not grow under these conditions, but a small number of cells grew out as floating spherical colonies after 5 d of culture (Fig. 2A). Interestingly, no sphere formation was observed in the presence of serum or TSH.
|
), indicating that thyrospheres grew in size by stem cell division. During the sphere formation, numbers of spheres in every dish increased slightly. The average cell counts in each sphere increased substantially during the first 3–4 d in culture and then reached a plateau from d 5–10 (Fig. 2, C and D). No spheres increased in size upon extension of the culture period beyond 5 d. Gene profile of thyrospheres displayed high expression of stem cell markers Oct4 and ABCG2 as well as endodermal markers GATA4 and HNF4,but no expression of thyroid differentiation markers PAX8, TG, NIS, TSHr and TPO (Fig. 3B).
|
). After 3 d of differentiation initiation with serum (d 3), sphere-derived cells began to express TSHr gene at very low level. Then 5 mU/ml TSH were added to the medium and cells were cultured for 18 additional days. Notably, TSHr expression gradually increased from d 3 to 21 of culture in the presence of both TSH and serum. These data demonstrated that TSH can stimulate TSHr expression in thyrosphere cells in the presence of serum. When differentiation in response to serum and TSH was completed at d 21, sphere-derived cells differentiated into thyrocytes that expressed PAX8, TG, NIS, TSHr, and TPO mRNA, respectively (Fig. 3B) but almost no stem cell or endodermal markers. Importantly, embedded in collagen for 3–4 d, differentiated cells formed thyroid follicle-like structures (Fig. 4A), which displayed TSH-dependent 125iodide uptake (Fig. 4B), a hallmark of differentiated thyroid cells.
|
), demonstrated that sphere cells proliferated rapidly during the first 3–4 d of sphere formation in response to intense growth stimulation, and that proliferation slowed down on d 5; upon stimulation with TSH in serum-enriched medium, sphere-derived cells divided slowly and stably in the same manner as primary cultured thyrocytes. FACS analysis of human thyrosphere cells
Whereas in primary thyroid cultures, stem cells sorted as a side population fraction represented only about 0.1% of the total cell population (Fig. 1, A and B), growth factor-stimulated sphere cells contained about 5% stem cells (Fig. 5A). This demonstrates a 50-fold enrichment of stem cells in thyrospheres. Analysis of gene expression in cells isolated from spheres demonstrated that the proliferating sphere cells consisted of either Oct4- and ABCG2-positive cells (side population fraction) or endodermal marker-positive cells (non-side population) (Fig. 5B).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
, A, B, and G). Because FACS separation of side population cells is limited by the fact that there is a continuous distribution between side population and non-side population cells, the size of the side population gate was reduced to select only verapamil-sensitive, i.e. ABCG2 transporter expressing, cells that represent a stem cell population with the lowest degree of cellular differentiation (34). Isolated side population cells cultured as monolayer or in Matrigel were resistant to growth stimulation. This observation that was also made with other stem cells may be explained by the interaction between stem cells and microenvironmental cells (niche cells) that nurture stem cells and regulate the balance between cell quiescence and activity (35). Niche cells provide a sheltering environment that protects stem cells from uncontrolled differentiation stimuli, apoptotic stimuli and, on the other hand, from excessive proliferation that could lead to cancer (35). The lack of these regulatory cells may explain why adult stem cells isolated as single cells do not proliferate despite intense growth stimulation.
With a complementary approach that was first used to separate neural stem cells (36), proliferating thyroid stem cells were isolated directly from primary thyroid cultures as nonadherent, three-dimensional spheres in a medium enriched with EGF and bFGF. The initial doubling time of stem cells was about 12 h, which is much shorter than that of normal human thyrocytes (22). With increasing size of spheres, proliferation rate slowed down (Figs. 2D and 3C).
The isolation was possible only in a serum-free medium without TSH. When either TSH or serum was added, formation of spheres was not observed. Why do stem cells escape the growth control of putative niche cells only in the absence of serum? In serum-free medium, differentiated cells may undergo apoptosis because they lose anchorage-dependent growth (37). In addition, TSH inhibits Fas antigen-mediated apoptosis of thyroid cells (38), which may explain the preserved function of putative niches in response to TSH-containing but serum-free medium.
When stem cells isolated from thyrospheres were grown as monolayer or embedded in collagen, cells differentiated under the influence of TSH in a serum-enriched medium. They expressed PAX8, TG, NIS, TSHr, and TPO mRNA (Fig. 3B) and when embedded in collagen showed 125iodide uptake in response to TSH (Fig. 4B). This proves the ability of adult stem cells derived from goiters to differentiate into thyroid cells.
It has been shown that adhesion is a key factor for the differentiation of stem cells. In particular, the extracellular matrix is considered a factor of survival and differentiation for many adherent cells (39). Adhesion generates cell tensional integrity (tensegrity) and repression of apoptotic signals, whereas detachment has the opposite effect. In the present study, in serum-enriched medium, thyroid stem cells regained the ability of adhesion and interaction with the extracellular matrix in the first 3 d of differentiation initiation. Consecutively, they underwent the first steps of differentiation including induction of TSHr gene expression. TSH-dependent differentiation of stem cells was first described for the development of the mice thyroid (40). A directed differentiation of mouse embryonic stem cells into thyroid follicular cells required the involvement of TSH, which stimulated and maintained TSHr expression.
Depending on the diagnostic approach, i.e. histological examination or molecular analysis, thyroid nodules are described as hyperplastic lesions, adenomatous nodules, adenomas, and clonal or polyclonal nodules (3). In most cases the mechanism underlying autonomous growth of thyroid lesions is the spontaneous proliferation of a benign neoplasia (1). The molecular alterations that cause the formation of these benign neoplasias are largely unknown (4). Recently stem cells have been hypothesized to be a source of malignant tumors including thyroid carcinomas (17, 18). In the case of colon tumors, the concept of stem cell transformation has also been extended to adenomas (41). The same concept may also apply to benign thyroid neoplasias.
Although any conclusions drawn from in vitro data have to be treated with caution, there are some findings that may suggest stem cells also as a putative source of benign thyroid neoplasias. The role of growth factors in the formation of thyrospheres bears resemblance to the pathogenetic importance of these factors for the growth of thyroid nodules (42, 43). In fact, chronic stimulation of the thyroid gland by growth factors, highly accelerated under the conditions of iodide deficiency (44), resembles in some ways the acute intense stimulation of quiescent stem cells derived from goiter tissues in vitro. Outgrowth of stem cells as spheres is provoked by the same growth factors that are also involved in the nodular transformation of the thyroid gland (43, 45, 46).
In vitro, the way from (undifferentiated) stem cells to differentiated thyrocytes takes about 3 wk. In vivo, on the way from the adult stem cell to a differentiated thyrocyte, which may take much longer, molecular alterations may occur that confer a growth advantage to the affected cell. Depending on the state of development from an undifferentiated stem cell to a more differentiated progenitor cell, a less or more differentiated thyroid carcinoma may emerge as it has been hypothesized previously (17). In the same way, a more differentiated progenitor cell with a limited proliferation capacity may outgrow to form a benign thyroid neoplasia.
More than 20 yr ago, transplantation of nodular goiter tissues on nude mice exhibited autonomous growth of some thyroid cells with a constitutively higher growth potential (47). Whether these autonomously proliferating cells correspond to the offspring of adult stem cells, which are scattered throughout the human thyroid gland, has to be demonstrated.
Further work has to focus on the molecular and cellular aberrations that may occur on the long way from adult stem cells to differentiated thyroid cells in vivo and may induce thyroid tumor formation.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Disclosure Statement: The authors have nothing to disclose.
First Published Online July 3, 2007
1 L.L. and D.C. contributed equally to this work. 
Abbreviations: bFGF, Basic fibroblast growth factor; BrdU, 5-bromo-2'-deoxyuridine; EGF, epidermal growth factor; FACS, fluorescence-activated cell sorting; FCS, fetal calf serum; FRTL, Fischer rat thyroid cell line; HNF, hepatocyte nuclear factor; NIS, sodium iodide symporter; Oct, octamer transcription factor; PAX8, paired box gene 8; PI, propidium iodide; TG, thyroglobulin; TPO, thyroperoxidase; TSHr, thyroid-stimulating hormone receptor.
Received February 6, 2007.
Accepted June 25, 2007.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  C. Y. Chen, H. Kimura, M. A. Landek-Salgado, J. Hagedorn, M. Kimura, K. Suzuki, W. Westra, N. R. Rose, and P. Caturegli Regenerative Potentials of the Murine Thyroid in Experimental Autoimmune Thyroiditis: Role of CD24 Endocrinology, January 1, 2009; 150(1): 492 - 499. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Thomas, S. Friedman, and R.-Y. Lin Thyroid stem cells: lessons from normal development and thyroid cancer Endocr. Relat. Cancer, March 1, 2008; 15(1): 51 - 58. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
