| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Department of Surgery (S.H., V.S., A.W., I.H., S.L., A.Z.), and Division of Gastroenterology and Endocrinology, Department of Medicine (L.C.H.), Philipps University, D-35032 Marburg, Germany
Address all correspondence and requests for reprints to: Dr. Sebastian Hoffmann, Baldingerstrasse, 35043 Marburg, Germany. E-mail: hoffmans{at}mailer.uni-marburg.de.
| Abstract |
|---|
|
|
|---|
, TGF-ß1, fibroblast growth factor, platelet-derived growth factor, and hepatocyte growth factor] cells were analyzed for VEGF gene expression by Northern blotting and for VEGF protein by enzyme immunoassay. TSHr signal transduction was evaluated by analyzing the effect of stimulators (cholera toxin, 8-bromo-cAMP, forskolin, and 12-O-tetradecanoyl-phorbol-13-acetate) and inhibitors (2',5'-dideoxyadenosine and staurosporine) on VEGF protein levels under basal and TSH-stimulated conditions. TSH increased VEGF mRNA and protein in a dose-dependent manner in HTC TSHr and XTC cells by up to 40%. The effects of TSH were mediated by protein kinase C (PKC), rather than protein kinase A (PKA), stimulation, because inhibition of PKC by staurosporine resulted in a decrease in VEGF production of up to 65%, whereas inhibition of the PKA signal transduction pathway (2',5'-dideoxyadenosine) resulted in only a minor decrease. TSH was not the most powerful stimulator of VEGF production. TGF-ß1 and EGF were 1.5- to 2-fold more potent. Placenta growth factor and TGF-
did not induce VEGF production in TSHr-positive HTC cells, whereas they did induce VEGF production in TSHr-negative HTC cells. In thyroid cancer cell lines, TSH induces VEGF production involving the PKC, rather than the PKA, pathway. However, EGF and TGF-ß increase the capacity of thyroid cancer cells to provide VEGF more effectively than TSH. In the absence of a functioning TSHr, additional growth factors, such as TGF-
, increase capacity for VEGF stimulation. | Introduction |
|---|
|
|
|---|
TSH is the major regulating hormone of the thyroid, and as such, it determines numerous biological functions of thyrocytes. Commonly, TSH is regarded as a thyroid-specific growth factor, inducing differentiation and, at concentrations considerably higher than those necessary to induce differentiated function, the growth of thyroid cells in vitro (9). It has previously been reported that human thyroid follicles increase VEGF mRNA in response to TSH and that TSH induced VEGF secretion in cultures of human thyroid cells as well as thyroid cancer cells in vitro (10, 11). However, the detailed mechanisms of TSH-induced VEGF production have not yet been studied. Several other growth factors have also been shown to modify the growth and function of thyroid cancer cells in vitro. That these factors may also modulate VEGF production is suggested by circumstantial evidence. For instance, it has recently been reported that VEGF secretion increased upon stimulation of primary cultures of papillary thyroid carcinoma as well as cell lines of leiomyosarcoma and breast cancer with hepatocyte growth factor (HGF), and that epidermal growth factor (EGF) induced VEGF production in human endometrial cells (12, 13, 14). In addition, TGF-ß has been associated with VEGF expression and microvessel density in colorectal cancer (15).
Because endothelial cell proliferation is essential for tumor growth, it is of considerable interest to understand the signaling pathways involved in the regulation of VEGF production by growth factors and to characterize inhibitors of endothelial mitogens. Such knowledge offers the opportunity to apply novel diagnostic and therapeutic modalities (16, 17). Therefore, the aims of the current study were to establish the effect of TSH on VEGF production in thyroid cancer cell lines in vitro and to evaluate the TSH postreceptor signal transduction of this effect. We also wanted to screen other growth factors that have previously been shown to modulate the function of thyroid cells in vitro, with regard to their ability to induce VEGF in thyroid cancer cells.
| Materials and Methods |
|---|
|
|
|---|
Culture flasks and dishes were obtained from Corning (Corning, NY), cell culture medium was purchased from Sigma-Aldrich Corp. (St. Louis, MO). The random primer labeling kit (Decaprime II) was obtained from Ambion, Inc. (Austin, TX), and [
-32P]deoxy-CTP was purchased from DuPont NEN (Boston, MA). The human ß-actin cDNA insert, ExpressHyb solution, was purchased from Clontech (Palo Alto, CA). All cytokines and bovine TSH as well as all other chemicals were obtained from Sigma-Aldrich Corp.
Thyroid cancer cell lines and culture conditions
Early passages of three established human thyroid cancer cell lines [HTC TSH receptor (TSHr), naive HTC, and XTC] were maintained in DMEM-h21/Hams F-12 (1:1, vol/vol) supplemented with 25 mM HEPES, 0.055 g/liter sodium pyruvate, 0.365 g/liter glutamine, 10% fetal bovine serum, 10,000 U/liter penicillin, and 100 mg/liter streptomycin at 37 C in a 100% humidified, 5% CO2 atmosphere as previously described (18). The HTC cell line is a follicular thyroid cancer cell line subcultured from the well described FTC 133 cell line (19). HTC cells have no endogenous expression of the TSHr, as documented by functional assays as well as PCR (19, 20). These cells were transfected with TSHr cDNA and isolated after G418 selection. The resultant cell line was designated HTC TSHr and has been shown to express functioning TSHr, which were in the order of nonneoplastic thyroid cells (21). These cell lines were provided by Dr. Derwahl (Berlin, Germany). The XTC cell line is a highly differentiated thyroid cancer cell line of Huerthle cell origin and has been described in detail previously (22). Before experiments, transfected HTC cells were expanded in the presence of G418 (100 µg/ml). Cells planned for use in experiments were switched to serum-free H5 medium for 48 h (23). H5 medium contains bovine insulin (10 µg/ml), human transferrin (5 µg/ml), somatostatin (10 ng/ml), glycyl-L-histidyl-L-lysine acetate (2 ng/ml), and hydrocortisone (108 M). For the experiments, thyroid cancer cells in exponential growth were harvested by brief incubation with cold trypsin-EDTA (Sigma-Aldrich Corp.; 0.025%/M) and resuspended in serum-free H5 medium, and vitality was assessed by trypan blue exclusion.
VEGF immunohistochemistry
Immunohistochemical staining for VEGF was performed on naive preparations and on 3-µm sections of specimens obtained from experimental tumors of the employed cancer cell lines. In brief, after melting and incubation in citric acid buffer for 18 h at 60 C, slides were deparaffinized and rehydrated in graded alcohol. After blocking endogenous peroxidases and unspecific binding, incubation with primary antibody (polyclonal mouse anti-VEGF antibody; 1:75 to 1:250; sc152, Santa Cruz Biotechnology, Inc., Santa Cruz, CA) was carried out overnight at 4 C and additionally enhanced using the streptavidin-based Link&Label Detection System (BioGenex, Munich, Germany). Slides were stained with diaminobenzidine-chromogen (Dako-Chemicals, Hamburg, Germany), counterstained with hematoxylin according to Mayer, and mounted. Naive preparations were treated similarly without deparaffination.
Northern blot analysis
Total RNA of thyroid cancer cells was isolated using the RNeasy kit and QiaShredder (Qiagen, Hilden, Germany). Five to 10 µg total RNA were separated on 1.5% (wt/vol) agarose/formaldehyde (2.2 M) gel and transferred to Hybond N+ nylon membranes (Amersham Biosciences, Arlington Heights, IL) by capillary blotting. Random prime labeling and hybridization procedures were carried out as previously described (24). A 320-bp [32P]VEGF probe hybridizing to three mRNA species of VEGF was provided by Dr. Sander (Regensburg, Germany) (25). Control hybridizations with human ß-actin cDNA verified that equal amounts of RNA were loaded. Experiments were repeated three times with similar results.
VEGF protein measurement
Conditioned medium was harvested from cultured cells and centrifuged (15,000 x g, 20 min, 4 C) to remove debris, and aliquots were stored at 80 C until analyzed. VEGF protein concentrations were quantified using a commercially available competitive VEGF enzyme immunoassay (R&D Systems, Wiesbaden, Germany). Experiments were repeated three times with similar results.
Growth factor stimulation and evaluation of TSH signal transduction
Thyroid cancer cells were seeded at a density of 105 cells into six-well multiwell plates and allowed to resume growth for 48 h in regular growth medium, then switched to serum-free H5 medium for 24 h and incubated with bovine TSH at 0.1100 mU/ml for another 48 h. Incubation was also carried out with effectors of the TSH signal transduction cascade and various other growth factors. Assessment of the TSH signal transduction pathway involved G protein effectors (cholera toxin), stimulators (8-bromo-cAMP, forskolin, and 12-O-tetradecanoyl-phorbol-13-acetate, at 10 and 100 ng/ml each), and inhibitors [2',5'-dideoxyadenosine (ddA) and staurosporine, at 10 and 100 ng/ml, respectively] of the protein kinase A/protein kinase C (PKA/PKC) cascade under unstimulated (basal) and TSH-stimulated conditions. The commercially available growth factors [EGF, IGF, placenta growth factor (PIG), TGF-
, TGF-ß, basic fibroblast growth factor (bFGF), platelet-derived growth factor (PDGF), HGF, and bovine TSH] were titrated in pilot experiments (0.01100 ng/ml), and comparative incubation studies were performed at their maximum effective concentrations.
Statistical analysis
Unless otherwise stated, values are expressed as the mean ± SD. A paired t test was used to evaluate differences in continuous variables from samples of interest and their respective controls during TSH stimulation and evaluation of TSH postreceptor signal transduction. All tests were two-tailed. P < 0.05 was considered to indicate significance.
| Results |
|---|
|
|
|---|
Northern blotting, enzyme immunoassay, and immunohistochemical analysis confirmed VEGF gene expression and the presence of VEGF protein in all of the thyroid cancer cell lines (HTC TSHr, naive HTC, and XTC). To confirm that VEGF expression was not limited to the in vitro environment, we obtained immunostains from tumors of the three cell lines, which had previously been grown in nude mice in another study (Fig. 1
). The potential of these cell lines to generate VEGF in an in vivo environment was evidenced by abundant VEGF expression. Individual differences in the amount of steady state VEGF protein secreted into the culture medium were repeatedly documented. When adjusted to the number of viable cells, the average basal VEGF protein accumulation into serum-free, chemically defined medium (H5) was registered as follows: HTC, 388 ± 55 pg/ml; HTC TSHr, 1143 ± 361 pg/ml; and XTC, 469 ± 93 pg/ml.
|
TSH induced VEGF mRNA steady state levels in XTC and HTC TSHr cells from low baseline levels in a dose-dependent fashion at supraphysiological doses of TSH (100 mU/ml), respectively (Fig. 2
). A similar dose-dependent induction of VEGF protein secretion by TSH was observed in the conditioned medium harvested from the two TSHr-positive cell lines (1630% increase at 10 mU/ml and 2037% at 100 mU/ml; P < 0.05, by t test; Fig. 3
). As expected, mRNA levels as well as VEGF protein accumulation were unaffected by TSH regardless of the dose in naive HTC cells devoid of the TSHr. This was interpreted as proof of the specificity of the TSH effects (Fig. 3
).
|
|
Up-regulation of unstimulated (basal) VEGF protein was observed with the Gs inductor cholera toxin, the PKC agonist 12-O-tetradecanoyl-phorbol-13-acetate, and, in the HTC cell lines only, apparently also with forskolin, a potent stimulator of cAMP, at 100 ng/ml each. In contrast, the adenylate cyclase (AC) stimulator 8-bromo-cAMP did not induce VEGF protein accumulation in any of the cell lines (Table 1
). From these studies it appeared that induction of VEGF secretion most likely followed the phospholipase C-ß (PLC-ß)/PKC pathway in TSHr-positive cell lines. Accordingly, secretion of VEGF was most effectively blocked by inhibition of PKC. Blocking AC/PKA with the cAMP antagonist ddA (up to 100 ng/ml) had no discernible effect on basal VEGF secretion, whereas the PKC antagonist staurosporine produced a decrease in basal VEGF secretion of 3365%. The inhibitory action of staurosporine was also registered in the TSHr-negative HTC cell line, where it resulted in the most pronounced decrease in basal VEGF secretion, exceeding that of TSHr-positive cell lines by 2535% (Table 1
). It is therefore suggested that regulation of basal VEGF secretion follows the PLC-ß/PKC, rather than the AC/PKA, signal transduction pathway (Fig. 4
).
|
|
Growth factor stimulated VEGF production
At their maximal effective concentrations, EGF (10/100 ng/ml), TGF-ß (10/100 ng/ml), and TSH (100 mU/ml) induced VEGF mRNA steady state levels in TSHr-positive XTC and HTC cells. No changes were seen with IGF, PIG, TGF-
, bFGF, PDGF, and HGF. In the TSHr-negative HTC cell line, EGF, and TGF-ß caused a similar induction of VEGF mRNA. In addition, TGF-
was found to induce VEGF mRNA (Fig. 5
). No appreciable changes occurred with HGF, PIG, TSH, IGF, bFGF, or PDGF (data not shown).
|
did not induce VEGF gene expression or stimulate VEGF secretion in TSHr-positive HTC and XTC cells, whereas they did in the TSHr-negative HTC cell line. In these cells, an increase in VEGF secretion up to 35% occurred in response to TGF-
and PIG, suggesting that the absence of a functioning TSHr may allow alternate pathways of VEGF up-regulation. Finally, we found that coincubation of the different growth factors (1100 ng/ml) with TSH (1100 mU/ml) in the TSHr-positive XTC cell line had no synergistic effect on VEGF production, suggesting a dominant regulation of VEGF secretion by EGF and TGF-ß.
|
| Discussion |
|---|
|
|
|---|
VEGF is a dominant regulator of angiogenesis in benign as well as malignant thyroid disease (30, 31, 32). VEGF has been detected in enlarging and recurrent thyroid nodules of patients with nodular goiter (33). VEGF gene expression has been documented in thyroid cancer cell lines (5). It has been demonstrated that up-regulation of VEGF occurs in human thyroid cancer and that this phenomenon is associated with malignancy and progression (34, 35). VEGF expression has been suggested to be a prognostic marker of local and distant metastasis in papillary thyroid cancer and to be correlated with tumor stage (6, 7, 8). In the present report all of the thyroid cancer cell lines had documented VEGF gene expression and accumulation of VEGF protein into conditioned medium. However, we found significant differences in the basal VEGF production of individual cell lines. It is currently not known how the magnitude of in vitro VEGF secretion relates to malignancy.
That TSH is involved in the regulation of VEGF in the thyroid had already been reported. For instance, an increase in VEGF mRNA in response to TSH and Graves immunoglobulin G has been demonstrated in human thyroid follicles in vitro (10). TSH has been shown to increase VEGF protein in cultures of normal human thyrocytes as well as thyroid cancer cell lines (11). However, how TSH regulates VEGF production in neoplastic thyroid cells had not previously been studied. In particular, the detailed mechanisms of TSHr signal transduction were unknown. In the present study TSH stimulated VEGF gene expression and accumulation of protein of differentiated thyroid cancer cell lines in a dose-dependent manner. TSH binds to specific high affinity receptors and activates two main signal transduction pathways, the AC-PKA pathway and the PLC-PKC pathway. Activation of the AC-PKA system by TSH is commonly thought to initiate differentiated function of follicular thyroid cells, whereas PKC stimulates undifferentiated function and growth of transformed thyrocytes (36). We have previously reported that TSH stimulates the growth and invasion of thyroid cancer cells in vitro and that such activation is also mediated via PKC (18, 37, 38, 39). We have now shown that with respect to the production of VEGF, TSH also acts through the PLC-ß/PKC signal transduction pathway. Taken together, these findings support the concept that several features of the malignant phenotype of thyroid cancer cells, such as in vitro proliferation, migration, invasion, and secretion of angiogenesis-inducing growth factors, involve activation of PKC, rather than AC-PKA. These observations may be yet another explanation for the reported negative association between the level of PKC activity and clinical outcome (40). Moreover, this concept may have therapeutic potential. In a recent study, treatment with selective PKC inhibitors decreased plasma VEGF levels as well as intratumoral vessel density and tumor growth in experimental human lung, renal, and colon cancer (41).
However, we found the increase in VEGF protein in response to TSH to be rather moderate. Given the fact that loss of the TSHr is characteristic of anaplastic thyroid cancer, which, in turn, usually exhibits significantly increased VEGF expression and a high degree of neoangiogenesis compared with differentiated thyroid cancer, it is likely that growth factors and cytokines other than TSH are also involved in regulating VEGF-mediated angiogenesis. To this end, we have documented for the first time that EGF, TGF-ß, and TGF-
are, in fact, potent stimulators of VEGF in thyroid cancer cells in vitro and are more effective than TSH.
The effects of EGF on thyroid cancer cells are quite well studied. We and others have repeatedly documented proliferation, migration, and invasion of thyroid carcinoma cells to be enhanced by EGF in vitro and in vivo (37, 38, 42, 43). Accordingly, in vitro growth was inhibited when either neutralizing anti-TGF or anti-EGF receptor (EGFR) antibodies were applied to thyroid carcinoma cell lines (44, 45). Although proliferation is probably mediated by auto- or paracrine loops, as suggested by the synchronous expression of EGF, TGF-
, and EGFR (46, 47), EGF-induced invasion may be related to matrix metalloprotease-1 gene expression (48).
Although EGF and EGFR are widely expressed in normal as well as neoplastic thyroid tissue, EGFR has been shown to be an independent prognostic indicator of tumor recurrence and recurrence-free survival (49, 50, 51). The association between EGF/EGFR expression and poor clinical outcome has since been included in the concept of malignant progression of thyroid tumors (47). Finally, it has recently been reported that the expression of CD97, a member of the EGF-TM7 family, also correlates with tumor stage and lymph node involvement in thyroid carcinomas (52).
How EGF and TGF are associated with angiogenesis, however, is not fully understood. EGF has been shown to increase VEGF production in human endometrial stromal cells (14). Involvement of EGF in the angiogenetic cascade was also suggested by the ability of EGF to induce thrombospondin-1 and plasminogen activator inhibitor-1 (53). Findings in human retinal pigment epithelial cultures, however, have documented TGF-ß1, TGF-ß2, and TGF-ß3, but not EGF, to stimulate in vitro VEGF secretion (54). Our data demonstrate for the first time that EGF and TGF-ß are major independent regulators of VEGF in thyroid carcinoma cells in vitro, stimulating VEGF secretion in all of the investigated thyroid cancer cell lines. Our observation that TGF-
and PIG stimulate VEGF secretion in thyroid carcinoma cells lacking functioning TSHr supports the hypothesis that the absence of functioning TSHr may give rise to alternate pathways of VEGF up-regulation. This finding is well in line with the hypothesis that undifferentiated thyroid carcinoma cells, as evidenced by the absence of TSHr, escape from the regulatory effect of the differentiating growth factor TSH. Support for this hypothesis came from our previous studies of the growth, invasion, and in vivo growth of growth factor-primed thyroid cancer cells (37, 38, 39). This hypothesis was underscored by the fact that reintroducing functioning TSHr to undifferentiated follicular thyroid carcinoma cells by means of transfection resulted in less proliferative and invasive in vitro potential, diminished in vivo growth (55), and less angiogenesis (Hoffmann, S., A. Zielke, and A. Wunderlich, unpublished observations).
In summary, we have shown that TSH induces VEGF gene expression and protein secretion in thyroid carcinoma cell lines in vitro. Our data confirm VEGF production of transformed thyrocytes to be stimulated by physiological concentrations of TSH and to involve the PKC signal transduction pathway. However, induction of VEGF by TSH was moderate compared with that by other cytokines with established importance in thyroid neoplasia, such as EGF, TGF-ß, and TGF-
. These growth factors, which involve tyrosine kinase-mediated pathways, are more effective than TSH to induce VEGF in thyroid cancer cell lines.
It also appeared that the absence of a functioning TSHr increases the panel of VEGF-stimulating factors. It is suggested that tyrosine kinase-mediated growth factors may significantly contribute to VEGF-mediated angiogenesis in undifferentiated thyroid carcinomas. It should be worthwhile to perform studies on antiangiogenesis using specific inhibitors of PKC or inhibitors of tyrosine kinases in undifferentiated thyroid cancers.
| Footnotes |
|---|
Abbreviations: AC, Adenylate cyclase; bFGF, basic fibroblast growth factor; ddA, 2',5'-dideoxyadenosine; EGF, epidermal growth factor; EGFR, epidermal growth factor receptor; HGF, hepatocyte growth factor; PDGF, platelet-derived growth factor; PIG, placenta growth factor; PLC, phospholipase C; TSHr, TSH receptor; VEGF, vascular endothelial growth factor.
Received June 30, 2004.
Accepted August 31, 2004.
| References |
|---|
|
|
|---|
and epidermal growth factor receptor in thyroid tumors. In Vivo 6:291296[Medline]
This article has been cited by other articles:
![]() |
J. H. D. Bassett, A. J. Williams, E. Murphy, A. Boyde, P. G. T. Howell, R. Swinhoe, M. Archanco, F. Flamant, J. Samarut, S. Costagliola, et al. A Lack of Thyroid Hormones Rather than Excess Thyrotropin Causes Abnormal Skeletal Development in Hypothyroidism Mol. Endocrinol., February 1, 2008; 22(2): 501 - 512. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. Klubo-Gwiezdzinska, R. Junik, E. Kopczynska, O. Juraniec, and H. Kardymowicz The comparison of serum vascular endothelial growth factor levels between patients with metastatic and non-metastatic thyroid cancer, and patients with nontoxic multinodular goiter Eur. J. Endocrinol., October 1, 2007; 157(4): 521 - 527. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Lonn, P. Bhatti, B. H. Alexander, M. A. Pineda, M. M. Doody, J. P. Struewing, and A. J. Sigurdson Papillary Thyroid Cancer and Polymorphic Variants in TSHR- and RET-Related Genes: a Nested Case-Control Study within a Cohort of U.S. Radiologic Technologists Cancer Epidemiol. Biomarkers Prev., January 1, 2007; 16(1): 174 - 177. [Abstract] [Full Text] [PDF] |
||||
![]() |
X.-M. Yu, C.-Y. Lo, W.-F. Chan, K.-Y. Lam, P. Leung, and J. M. Luk Increased Expression of Vascular Endothelial Growth Factor C in Papillary Thyroid Carcinoma Correlates with Cervical Lymph Node Metastases Clin. Cancer Res., November 15, 2005; 11(22): 8063 - 8069. [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 |