| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Institut de Recherche Interdisciplinaire en Biologie Humaine et Moléculaire (J.-C.G., L.J., H.M., C.L., J.-E.D., B.C.), Faculté de Médecine de l’Université Libre de Bruxelles, and Department of Endocrinology (B.C.), Hôpital Universitaire Erasme, 1070 Bruxelles, Belgium; and Microarray Facility (P.V.H.), Katholiek Universiteit van Leuven, 3000 Leuven, Belgium
Address all correspondence and requests for reprints to: Jean-Christophe Goffard, 808 route de Lennik, 1070 Bruxelles, Belgium. E-mail: jcgoffar{at}ulb.ac.be.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
(2). To reproduce the effect of chronic stimulation of the thyrocyte by the cAMP cascade, three transgenic mouse models examining the oncogenic potential of the constitutively elevated cAMP in the thyroid have been developed. These models were all generated using a thyroglobulin promoter as the expression of the natural thyroglobulin gene is tightly restricted to the follicular cell of the thyroid gland (3). A fragment of the bovine gene promoter has been shown to limit the expression of the reporter gene to thyroid cells in transgenic mice. In the first mouse model, ectopic expression of the Gs protein-coupled receptor, adenosine A2a receptor, leads to permanent stimulation of adenylyl cyclase as a result of the continuous release of adenosine by most cell types (4). In the two other models, Gs is constitutively activated, either directly via an activating mutation of Gs R201H (5), or via thyroid specific expression of cholera toxin A1 subunit (6). In mice, these transgenes promote both cell proliferation and function, similar to the effect of the cAMP cascade in primary cultures of dog thyrocytes. This results in the appearance of a differentiated goiter and severe hyperthyroidism. The mice, with a thyroid-specific chronic activation of the cAMP cascade, are the animal corollary of human nonautoimmune familial hyperthyroidism, which is caused by a constitutive stimulation of the thyroid cAMP cascade affecting all thyrocytes from the embryological stage (7). Autonomous adenoma, arising in response to an activating mutation of the cAMP pathway, share the same pathogenic mechanism but affect only one thyrocyte that grows by clonal expansion. Thus, we can study the precise role of the cAMP cascade in the sequence of events promoting cell proliferation in in vivo conditions and during a long life span. This model is certainly a more physiological model than primary cultures which last a few weeks at the most. Among the three transgenic mouse models developed to study the role of the cAMP cascade in the thyroid, we decided to use the Tg-A2aR transgene, created in our lab, as it has the most severe phenotype.
To examine the transcriptional effects of a thyroid permanent activation of the cAMP cascade, we analyzed the changes in global gene expression in the thyroid of transgenic mice using DNA microarrays (8). This approach enabled the quantification of the effects of chronic cAMP stimulation on approximately 13,000 genes. Results of interest were validated by a real-time PCR assay on different independent samples.
The possible role of these genes in the pathogenesis of these benign tumors is also discussed
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
) (r2 = 0.88; P < 0.05). After correction of the background, the expression level of 360 genes or expressed sequence tags (ESTs) of 13,187 genes represented on the arrays were either decreased or increased more than 2-fold and, therefore, considered as significantly regulated. By using these criteria, we identified 167 spots representing genes that were significantly up-regulated, of which 94 were ESTs, and 193 spots that were significantly down-regulated, of which 89 were ESTs. Modulation level, Unigene accession numbers, and Gene ontology (9) proposed function are shown in Table 1 for the 73 up-regulated genes and in Table 2 for the 104 down-regulated genes (EST excluded). When considering only the 2151 genes with a known function, according to the gene ontology database for the level 3 biological process, 54.7% of the genes spotted on the array are involved in metabolism, 14.7% in signal transduction, 13.5% in transport, 7.9% in response to external stimulus, 7.8% in cell cycle, 7.16% in cell organization and biogenesis, 5.8% in embryogenesis and morphogenesis, 5.1% in cell adhesion, 3.4% in cell death, 1.6% in cell motility, 1.1% in cell proliferation, 1% in cell-cell signaling. The other categories represent less than 1% (Fig. 2). The distribution of regulated genes among the various categories was quite different from the total gene distribution, indicating that the regulation was not randomly distributed. Up-regulated genes are more frequent in the following functional categories: transport, cell adhesion, cell-to-cell signaling, cell differentiation, cell growth, and cell proliferation. Down-regulated genes are more frequent in the functional categories related to: response to external stimuli, cell adhesion, cell death and cell-to-cell signaling. Several categories stand out on closer examination of biological function:
|
|
|
|
|
|
|
|
|
|
|
|
|
For the down-regulated genes, we identified:
A high proportion of genes are not regulated. Among these are the genes of ribosomal proteins and translation factors, mitochondrial proteins, channels, histones, and heat shock proteins, i.e. genes of the basic housekeeping machinery.
All of the differentially regulated genes implicated directly or indirectly in the process of apoptosis were modulated in an antiapoptotic way, the proapoptotic genes (cell death-inducting DNA fragmentation factor, Fsp27, Lsp1, Lga11, Gelsolin) were all down-regulated, whereas the antiapoptotic genes were up-regulated (fructose phosphatase 1, Clusterin). In previous Northern blots of the RNAs of A2R thyroids, cyclin D1 expression was increased, whereas cyclin D2 and D3 were slightly decreased (10). The same modulations were also observed in the microarray but not at a level considered to be significant. In the microarrays cyclin D1 mRNA was increased (ratio 1.9), but not significantly, and cyclin D2 and D3 were barely decreased (ratio 0.9).
Validation of Microarray Data for Tg-A2aR Transgenic Mice by Using Quantitative Real-Time PCR Assay
To confirm the data obtained by microarray, we used quantitative real-time PCR assay (Taqman) as described in Materials and Methods. The differential expression of five up-regulated genes and one down-regulated gene identified by microarray analysis were validated by quantitative RT-PCR (QRT-PCR). We chose to confirm genes that showed the strongest up-regulation [mesoderm-specific transcript (MEST), TGF-ß I, cathepsin-L] or down-regulation on microarray (TM4SF6) or genes implicated in a common pathway of signal transduction [IGF-I, IGF binding protein (IGFBP) 5]. Porphobilinogen deaminase was used as normalizing control. QRT-PCR analysis confirmed that the six genes were statistically differentially regulated. The fold change of genes, as determined by QRT-PCR analysis, qualitatively correlated with the fold change reported by microarray analysis with no major discrepancies. We also used the thyroperoxidase (TPO) gene, known to be positively regulated by the cAMP cascade, for the validation of our Taqman experiment (Fig. 3). A 3-fold up-regulation of TPO was observed with QRT-PCR analysis. This result confirms the marked increase of TPO mRNA observed by Northern blot analysis in this transgenic line (4).
|
Western Blot and Immunohistochemistry
Western blot for caveolin 1 on protein extracted by Trizol confirmed a decrease in the expression of this gene at the protein level in most samples studied. The samples that did not show a decrease were the ones with a high cellular content, demonstrated by the high yield of RNA during the same Trizol extraction. The immunostaining showed that the down-regulation was mainly due to a decrease in the vascular expression of caveolin (Fig. 4).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
activation (12). In both cases, the stimulated thyroid overfunctions, causing continuous hyperthyroidism and growth, persisting into adulthood. Continuous goitrogenesis leads to repetitive surgery and possibly cancer. In both congenital hyperthyroidism and A2R mice all the thyroid cells are affected, i.e. the lesion presents a 100% phenotype. Autonomous adenomas have the same cause as congenital hyperthyroidism but occur by somatic mutation later in development and, in general, do not grow indefinitely. In both diseases as well as in the A2R model, the thyroid cells are only slightly different from normal thyroid cells. In fact, their normal function and cell proliferation process are activated but not qualitatively modified. They thus correspond to the definition of minimal deviation tumors as applied to hepatomas (13). This explains the first striking observation of this gene expression study that shows that only a limited number of genes are actually modulated. Indeed, around 2% of the genes represented on the arrays were significantly up- or down-regulated in Tg-A2aR transgene. The percentage of genes significantly up- or down-regulated is significantly lower than observed those in thyroid papillary cancer obtained from Tg-Ret/PTC3 transgenic mice (
5%, our unpublished results). In so far as gene expression may reflect protein expression, these lesions can therefore be qualified as minimal deviation thyroid tumors. Given that most of the general housekeeping genes, i.e. of the general protein synthesis machinery, of mitochondria and histones were not differentially regulated in these thyroids, is in agreement with the concept that A2R thyroid cells are normal cells functioning at a higher level and exhibiting a more rapid growth. It is possible, however, that the heterogeneity of the A2R thyroid, with areas more or less active, may have blunted to some extent the overall pattern of gene expression (10). cAMP cascade modulation of the vast majority of those genes has never been described before in the thyroid cell or other cell types. The precise regulation of these genes remains to be elucidated. Some of them may be directly activated by the cAMP cascade and play a role in the pathogenesis of the hyperactive goiter, others may act as a part of a negative feedback of such an activation as has been described for the induction of cAMP phosphodiesterase PDE4 in hot nodules and in in vitro-stimulated cells (14). It is therefore difficult to speculate on the implication of genes, with a still partially unknown function, in the pathogenesis of the hyperactive goiter. Conversely, when looking at the significantly up- or down-regulated genes with a well-characterized function, some hypotheses can be proposed.
The induction of IGF-I in the A2R mice fills a gap in our knowledge of growth control in the thyroid. We and others have shown that IGF-I, or high concentrations of insulin, acting through the IGF-I receptor and phosphatidylinositol 3-kinase, is required for the in vitro mitogenic action of TSH and the cAMP pathway in dog and human cells (12, 14, 15, 16, 17). This raises the question of the in vivo origin of the necessary IGF-I. The follicular cells are the main source of IGF-I in the thyroid (18, 19), suggesting an autocrine role for IGF-I in follicular cell growth. The data obtained in this study would indicate that TSH and cAMP indeed induce the IGF-I that is necessary for their proliferation effect.
The role of IGF-I in the pathogenesis of hot thyroid nodule is still controversial (20). It is interesting that in FRTL5 cells, as in cells from thyroid hot nodules, the requirement for exogenous IGF-I may disappear as the cells secrete their own somatomedins and thus become autonomous with regard to these hormones (15, 21). Our experiment, thus, seems to confirm the essential role of IGF-I in our model of congenital hyperthyroidism. The importance of this role in the thyroid hyperplasia of A2R mice is supported by the mild thyroid hyperplasia observed in mice with a thyroid specific overexpression of IGF-I and IGF-I receptor, the potentiation of the goitrogenic effect of methimazole in these mice (22) and by the generation of thyroid tumors in IGF-II overexpressing mice (23).
The mechanism of these effects of IGF-I and TSH at the level of cyclin and cyclin-dependent protein kinases involves the induction of cyclin D by IGF-I and the cAMP-elicited generation and activation of the cyclin D-CDK4 complex. In dog thyroid, the cyclin D involved in this process is cyclin D3 (17). By Northern blot analysis, cyclin D1, but not cyclin D2 or D3 mRNAs, were shown to be overexpressed in A2R thyroids (10). A similar but not significant effect is observed with the microarrays.
The role of IGFBPs are still controversial. They modulate IGF bioactivity positively or negatively and they also exert IGF-independent effects. IGFBPs possess both growth promoting and inhibitory effects on cells that are independent of IGF action and which are mediated through specific IGFBP receptors located at the cell membrane, cytosol, or nuclear compartments and in the extracellular matrix (24). On the other hand, the predominant effect of IGFBP overexpression in vivo is an inhibition of IGF-I action (25). IGF-I and TSH have been shown to modulate IGFBPs expression in vitro, presumably as a negative feedback (26). TSH acutely stimulates the expression of IGFBP3 in dog thyroid cells (27). Overactivity of the IGF pathway in mice transgenics expressing IGF-I and IGF-I receptor in their thyroids decreases IGFBP3 protein expression (22). Increased expression of IGFBP3 in the A2R thyroids thus probably results from the action of the cAMP pathway. Stimulation of IGFBP5 synthesis by the IGF-I in thyroid cells appears to be regulated at the mRNA level because it stimulates similar increases in both IGFBP5 mRNA and media protein at maximally effective concentrations. TSH, on the other hand, decreased basal levels of IGFBP5 mRNA and attenuated the increase in IGFBP5 mRNA stimulated by IGF-I (26). Our results thus show the up-regulation of IGFBP3 and IGFBP5 in the development of the hyperactive goiter secondary to a cAMP chronic stimulation. The overall growth effect of constitutive A2R in thyroid may therefore result from the dominant effect of IGF-I over the effects of the IGFBPs.
Ornithine decarboxylase is also significantly up-regulated. TSH and cAMP are known to enhance ornithine decarboxylase activity, the rate-limiting enzyme in polyamine synthesis (28). Formation of polyamines is closely linked to cell growth, although the mechanism is not known.
The proapoptotic genes [cell-inducing dff45-like effector (CIDE)-A, gelsolin, Galectin 1] were down-regulated. CIDE-a is a member of a family (CIDE) of cell death activators. CIDE-a was found to activate apoptosis downstream of caspase 3 activation probably through complex interactions with DNA fragmentation factors (DFF45 and DFF40). The expression of gelsolin, a caspase 3 substrate during apoptosis, is also down-regulated. The caspase cleaved gelsolin fragment can depolymerize actin filaments and can cause apoptosis at least in mouse cells. Galectin 1 is an endogenous mammalian S-type lectin with highly pleiotropic effects on different tissues and is implicated in apoptosis of T cells (29). The antiapoptotic gene Clusterin was up-regulated. The precise role of clusterin is not clear. The protein has been reportedly implicated in several diverse physiological processes such as sperm maturation, transportation, complement inhibition, tissue remodeling, membrane recycling, cell-to-cell and cell-substratum interactions, stabilization of stressed proteins in a folding competent state and promotion or more frequently inhibition of apoptosis (30, 31). Introduction of clusterin gene into human renal cell carcinoma cells enhances their resistance to cytotoxic chemotherapy through inhibition of apoptosis and overexpression of clusterin in these cells makes them resistant to Fas-mediated apoptosis (32).
Up-regulation of genes of antiapoptotic proteins and down-regulation of genes of proapoptotic proteins suggest a shift toward less apoptosis which fits the growth rate of these thyroids.
Continuous growth of A2R thyroid would fit in well with a shift of metabolism from catabolic to anabolic pathways. This is what is found at the level of gene expression with overexpression of the mRNA of proteins involved in anabolism and gluconeogenesis and down-regulation of genes of proteins involved in glycolysis and lipid metabolism. While increased expression of anabolic enzymes is similar to what is found in tumor cells this does not apply to gluconeogenesis (33). Fructose 1–6 biphosphatase, which is overexpressed in the A2R thyroid is a major enzyme implicated in gluconeogenesis. Measurements of the activity of these pathways would be required to confirm this conclusion.
Cathepsin B and L are two endopeptidases belonging to a family of lysosomial proteases responsible for the hydrolysis of thyroglobulin, leading to the secretion of T4 and T3. In thyroids with nonneoplastic diseases, including multinodular goiter and hot thyroid nodule, there was a significant increase in cathepsin B and cathepsin L enzyme activities and cathepsin B mRNA levels compared with normal thyroids from noncancerous cases. The major functional effect of the activation of the TSH cAMP pathway is the stimulation of thyroid secretion. Chronic stimulation leads to endocytosis of thyroglobulin by micropinocytosis and to an increased capacity of the system, hydrolyzing thyroglobulin, i.e. the lysosomes (4, 34). It is therefore quite fitting that the expression of the mRNAs of proteases is increased in A2R thyroids. Increased activity of cathepsins and other hydrolytic enzymes, as well as increased thyroglobulin hydrolysis and thyroid hormone secretion, have been demonstrated in autonomous adenomas (35). The down-regulation of caveolin-1 is of particular interest. This protein is a component of the specialized structure of the membrane called caveolae. Interaction between caveolin-1 and endothelial nitric oxide synthase has been described with the caveolin-1 scaffolding domain. This interaction appears to result in the inhibition of NOS activity (36). In our model of transgenic mice, we observed approximately three times more blood vessels in the Tg-A2aR thyroid compared with control. Caveolin-1 may be a very good candidate to explain this promotion of angiogenesis. In fact, the down-regulation of caveolin-1 observed in our model was mainly due to a decrease in expression of caveolin-1 by the endothelial cells as demonstrated by Western blot and immunostaining (Fig. 4
). This down-regulation can promote the activity of endothelial nitric oxide synthase and the generation of new capillary tubules (37).
More surprising is the down-regulation of the immune system-related proteins, suggesting a decrease of the corresponding population of cells, which had not been previously reported. Certainly absence of any immune reaction in the organ would benefit the observed growth. Thyroids affected by human congenital hyperthyroidism are remarkably free of any lymphocytic infiltration (38). Some of the proteins may be expressed in thyroid cells (e.g. MHC2). In that case, one could relate the decrease of expression to the down-regulation of MHC1 proteins by TSH in FRTL5 cells (39).
TGF-ß induced binds to type I, II, and IV collagens. This adhesion molecule has been implicated in human pathology. Missense mutations lead to granular corneal dystrophy without thyroid phenotype. The overexpression of TGF-I in the thyroid of our model is accompanied by over-expression of other adhesion molecules (Fibronectin1, procolloagen type XVIII) and down-regulation of other extracellular matrix molecules (procollagen type VI, chondroitin sulfate proteoglycan 2), illustrating the reorganization of the histological architecture of the transgenic thyroid.
The functions of the mesoderm-specific transcript and the tetraspanin 6 are far less characterized. The tetraspanin superfamily is implicated in a diverse range of biological phenomena, including cell motility, metastasis, cell proliferation, and differentiation (40). Little is known about tetraspanin 6, including its function. The mesoderm-specific transcript is a maternally imprinted gene whose expression was first described in the forming metanephros. A loss of imprinting has been described in various malignant conditions (41, 42) and in the processes of angiogenesis involving MEST in the control of cell proliferation.
The reproducibility of the pattern of gene expression in thyroids of mice 4, 6, and 9 months old indicates the constancy of the physiopathological process and of its mechanisms in these transgenics. It does not suggest further evolution of the lesion such as mutagenic or epigenetic events and spreading to a new population of cells. Indeed, in the absence of a supplementary oncogene (e.g. HPV16, E7), no secondary tumors or metastases appear before 18 months.
Conclusion
As shown in other studies of various tumors, microarrays may provide information difficult to obtain by other techniques. Our data demonstrate that in Tg-A2R thyroid, 360 genes or EST were modulated more than 2-fold when compared with normal mice. The regulation of the genes confirms our hypothesis on the consequences of a chronic cAMP stimulation in the thyroid. However, the regulation of the vast majority of the genes was not expected, and additional data are needed for a better understanding of their precise role in the pathogenesis of thyroid hyperplasia in Tg-A2R mice.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
The Tg-A2aR mice were treated with Methimazole (0.01%) (Sigma, St. Louis, MO; M-8506) added to the drinking water for a period extending to 1 wk before they were killed so as to avoid death related to hyperthyroidism.
Thyroid Dissection
The mice were killed under CO2 asphyxia. The thyroids were carefully removed, snap frozen in liquid nitrogen and stored at -80 C until use.
RNA Extraction
To allow for optimal RNA extraction, several thyroids were pooled to obtain approximately 100 mg of tissue. Total RNA extraction was performed using the Trizol reagent (Invitrogen, Carlsbad, CA). We used two different pools of normal thyroids obtained from wild-type mice aged 2–9 months old. Three additional different pools of Tg-A2aR thyroids were established from Tg-A2aR mice aged 4, 6, and 9 months old, respectively. The quality of RNA was tested on a 1% agarose gel by checking the integrity of the 18S and 28S bands. Quality of the RNA used for microarray hybridization was further analyzed by the bioanalyzer (Agilent Technologies, Palo Alto, CA). Only the RNA showing a greater peak for the 28S band was retained for microarray hybridizations.
RNA Amplification
Before microarray hybridization, antisense RNA amplification was performed using a modified protocol of in vitro transcription as published by Barry and Eberwine (43). For first-strand cDNA, 3 µg of total RNA was mixed with 2 µg of a HPLC-purified anchored oligo-deoxythymidine + T7 promoter (5'-GGCCAGTGAATTGTAATACGACTCACTATAGGGAGGCGGT24(ACG)-3') (Eurogentec, Seraing, Belgium), 40 U RNaseOUT (Invitrogen) and 0.9 M D(+) trehalose (Sigma) in a total volume of 11 µl, and heated to 75 C for 5 min. Two microliters of 0.1 M dithiothreitol (DTT), 1 µl 10 mM deoxynucleotide triphosphate (dNTP) mix, 1 µl 1,7 M D(+) trehalose (Sigma) and 1 µl (200 U) of SuperScript II (Invitrogen). The samples were then incubated in a Biometra-Unoll (Goettingen, Germany) thermocycler at 37 C for 5 min and 45 C for 10 min, followed by 10 cycles at 60 C for 2 min and at 55 C for 2 min. To the first-strand reaction mix, 103.8 µl water, 33.4 µl 5x second strand synthesis buffer (Life Technologies, Inc., Gaithersburg, MD), 3.4 µl 10 mM dNTP mix, 1 µl of 10 U/µl Escherichia coli DNA ligase (Invitrogen), 4 µl 10 U/µl E. coli DNA polymerase I (Invitrogen) and 1 µl 2 U/µl E. coli ribonuclease H (Invitrogen) were added. The samples were then incubated for 2 h at 16 C. The synthesized double-stranded cDNA was purified using a Qiaquick kit (QIAGEN, Chatsworth, CA). Antisense RNA was synthesized in a total volume of 20 µl using an AmpliScribe T7 high yield transcription kit (Epicenter Technologies, Madison, WI) according to the manufacturer’s instructions. The RNA was purified with a Rneasy purification kit (QIAGEN).
The labeling of RNA was performed during reverse transcription. First-strand cDNA probes were generated by reverse transcription of 2 µg poly(A) RNA using an anchored oligo-deoxythymidine (d-T25-dA/C/G) primer (0.4 µM; Genset, Evry, France), 0.1 mM deoxy (d)-(G/T/A)TPs, 0.05 mM dCTP (Amersham Pharmacia Biotech, Buckinghamshire, UK), 0.05 mM Cy3-dCTP or Cy5-dCTP (Amersham Pharmacia Biotech) 1x first-strand buffer, 10 mM DTT, and 200 U SuperScript II (Invitrogen) in a 20 µl total volume. The RNA and primers were denatured at 75 C for 5 min and cooled on ice before adding the other reaction components. After 2 h incubation at 42 C, mRNA was hydrolyzed in 250 mM NaOH for 15 min at 37 C. The sample was neutralized with 10 µl of 2 M 3-morpholinopropane sulfonic acid and purified using Qiaquick columns (QIAGEN).
Microarray Construction and Hybridization
A total of 4608 PCR amplified cDNA fragments from sequenced verified IMAGE clones (Mouse Gem I, Incyte) were spotted twice, one on each side of the slide. The cDNA inserts were amplified with M13 primers, purified with MultiScreen-PCR plate (Millipore, Eugene, OR) and resuspended in 20 µl 50% dimethylsulfoxide in an average concentration of 100 ng/µl. PCR amplicons were arrayed on type V silane-coated slides (Amersham Pharmacia Biotech) using a Molecular Dynamics Generation III printer (Amersham Pharmacia Biotech). Slides were blocked just before hybridization in 3.5% saline sodium citrate (SSC), 0.2% sodium dodecyl sulfate (SDS), 1% BSA for 10 min at 60 C. A set of five different slides, each containing 4596 different cDNA spotted in duplicate, were used.
The probes were resuspended in 30 µl hybridization solution (50% formamide, 5x SSC, 0.1% SDS, 100 mg/ml salmon sperm DNA) and prehybridized at 42 C for 30 min to block hybridizations of polydeoxythymidine tails of the cDNA on the arrays. Mouse COT DNA (1 mg/ml) (Invitrogen) was added to the mixture and placed on the array under a glass coverslip. Slides were incubated for 18 h at 42 C in a humid hybridization cabinet (Amersham Pharmacia Biotech). Posthybridization washing was performed for 10 min at 56 C in 1x SSC, 0.1% SDS, twice for 10 min at 56 C in 0.1x SSC, 0.1% SDS and for 2 min at 37 C in 0.1x SSC. The hybridizations were performed twice. First with the control samples labeled with Cy3 and the Tg-A2aR labeled with Cy5. The second hybridization was conducted with fluorophore inversion between the samples.
Scanning and Data Analysis
Arrays were scanned at 532 nm and 635 nm using a Generation III scanner (Amersham Pharmacia Biotech). Image analysis was performed using ArrayVision (Imaging Research, Inc. St. Catharines, Ontario, Canada). Spot intensities were background corrected and filtered based on 2 SD above background. Ratios were normalized by a linear regression between log10 ratio(cy5/Cy3) and log10 total intensity of Cy5 x Cy3. From duplicate spots, average ratios Cy5/Cy3 were used for further analysis. Only the spots with an average ratio greater than two in both hybridizations were retained for further analysis.
Validation of Gene Expression Data
Real-time quantitative PCR.
Total RNA (10 µg) was treated with deoxyribonuclease (DNase) to remove genomic DNA contamination in a total volume of 100 µl using 10 IU DNase in a 1x DNase buffer (Invitrogen) for 15 min at room temperature. The cDNA synthesis reaction was performed in a total volume of 55 µl, containing 1x first-strand cDNA buffer (Invitrogen), 10 nM DTT, 625 µM dNTPs, 5 µM oligo-deoxythymidine, 800 IU Superscript II. The reaction was performed for 10 min at room temperature, followed by 50 min at 42 C and 15 min at 70 C.
Using the Primer Express software program (PerkinElmer, Foster City, CA), we designed PCR primers and probes for the amplification of cDNA derived from selected transcripts. The Taqman probes carried a 5' carboxyfluorescein reporter label and a (4-(4-dimethylaminophenylazo)-benzoic acid dark quencher group. The following mRNAs’ expression was evaluated: IGF-I, IGFBP5, cathepsin-L, TPO, MEST, TGFBI, TM4SF6, and porphobilinogen deaminase as a normalizing control (44). The enzyme was activated by heating for 10 min at 95 C. We used a two-step PCR procedure, 60 sec at 60 C and 60 sec at 95 C for 40 cycles in a PCR mix containing 2 µl cDNA template, 1x quantitative PCR Mastermix (RT-QP2X-03, Eurogentec), 300 nM of each of the primers and 100 nM probe in a total volume of 30 µl. PCR, Taqman analysis and subsequent calculation were performed using the ABI/Prism 7700 Sequence Detector System (Applied Biosystem/PerkinElmer). In the intact Taqman probe, the 5' fluorescent reporter dye was quenched by the 3' dark quencher. The fluorogenic DNA probes were hydrolyzed during PCR upon hybridization to the template DNA by the 5' secondary structure dependent nuclease activity of the Taq DNA polymerase. Therefore, the increase in fluorescence intensity was therefore proportional to the accumulation of PCR product. The fluorescence intensity of the reporter label was normalized using the rhodamin derivative ROX as a passive reference label present in the buffer solution. The system generates a real-time amplification plot based upon the normalized fluorescence. The threshold cycle (Ct) was determined as the fractional cycle number at which the amount of amplified target reached a fixed threshold. This threshold was defined at a fluorescence value where all the amplification curves were above background fluorescence and still in exponential phase of amplification determined as a doubling of the fluorescence from one cycle to the other.
Gene expression in the samples from Tg-A2R mice was compared with samples coming from wild-type mice, using the 
Ct method described by Applied Biosystem.
Western Blotting Analysis
Protein extracts of total thyroid were obtained using the Trizol extraction method according to manufacturer’s instructions.
For these experiments, protein content is not a reliable marker of cellular content because of the markedly lower quantity of extracellular thyroglobulin in Tg-A2R mice (10). Therefore the cellular content was estimated according to the quantity of RNA extracted during Trizol reaction. Proteins were separated by SDS-PAGE and transferred to nitrocellulose membranes for 16 h at 30 V at 4 C. Membranes were incubated in a blocking solution containing 0.5% BSA and 5% dry milk in PBS. Caveolin-1 proteins were detected using rabbit polyclonal antibodies. The immune complexes were detected using a horseradish-peroxidase coupled antirabbit antibody and the enhanced chemiluminescence method (NEN Life Science Products, Boston, MA).
Immunohistochemistry
Tissue samples were fixed in paraformaldehyde (4%) for 18 h and embedded in paraffin by standard procedure. Sections (4 µm) were deparaffinized and rehydrated in a graded ethanol series. Endogenous peroxidase was blocked with 6% hydrogen peroxide for 30 min. Rabbit anti-caveolin-1 antibodies (1:200) were incubated for 90 min at room temperature. Dako Envision system (Dako, Carpinteria, CA) was used for detection of primary antibodies. Slides were counterstained with hematoxylin.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
Abbreviations: CIDE, Cell-inducing dff45-like effector; Ct, threshold cycle; DNase, deoxyribonuclease; dNTP, deoxynucleotide triphosphate; DTT, dithiothreitol; IGFBP, IGF binding protein; MEST, mesoderm-specific transcript; QRT-PCR, quantitative RT-PCR; SDS, sodium dodecyl sulfate; SSC, saline sodium citrate; Tg-A2aR, transgenic mice expressing the adenosine receptor; TPO, thyroperoxidase.
Received for publication June 25, 2003. Accepted for publication September 26, 2003.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
genes as a cause of toxic thyroid adenomas. J Clin Endocrinol Metab 82:2695–2701 subunit in thyroid glands of transgenic mice. Proc Natl Acad Sci USA 91:10488–10492This article has been cited by other articles:
 |
|
 |
|
  M. J. Costa, M. Senou, F. Van Rode, J. Ruf, M. Capello, D. Dequanter, P. Lothaire, C. Dessy, J. E. Dumont, M.-C. Many, et al. Reciprocal Negative Regulation between Thyrotropin/3',5'-Cyclic Adenosine Monophosphate-Mediated Proliferation and Caveolin-1 Expression in Human and Murine Thyrocytes Mol. Endocrinol., April 1, 2007; 21(4): 921 - 932. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Bruno, E. Ferretti, E. Tosi, F. Arturi, P. Giannasio, T. Mattei, A. Scipioni, I. Presta, R. Morisi, A. Gulino, et al. Modulation of Thyroid-Specific Gene Expression in Normal and Nodular Human Thyroid Tissues from Adults: An in Vivo Effect of Thyrotropin J. Clin. Endocrinol. Metab., October 1, 2005; 90(10): 5692 - 5697. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Kohrle Guard Your Master: Thyroid Hormone Receptors Protect Their Gland of Origin from Thyroid Cancer Endocrinology, October 1, 2004; 145(10): 4427 - 4429. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
