| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Third Department of Internal Medicine, Interdisciplinary Graduate School of Medicine and Engineering, University of Yamanashi, Chuo City, Yamanashi 409-3898, Japan
Address all correspondence and requests for reprints to: Dr. Toyoshi Endo, M.D., Third Department of Internal Medicine, Interdisciplinary Graduate School of Medicine and Engineering, University of Yamanashi, Chuo City, Yamanashi 409-3898, Japan. E-mail: endot{at}yamanashi.ac.jp.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Objective: We therefore investigated whether osteocalcin and/or Cbfa-1 genes are expressed in malignant thyroid epithelial cells. We also studied the effects of Cbfa-1 on the expression of osteoblast-specific and thyrotropin receptor genes in thyrocytes.
Results: The human thyroid papillary carcinoma cell line BHP18–21 expresses bone-type osteocalcin mRNA at higher levels than in MG63 osteosarcoma cells. Northern blot analysis and EMSA using nuclear extracts from BHP18–21 cells and FRTL-5 cells demonstrated that cells of thyroid epithelial origin expressed Cbfa-1/Runx2, the main transcription factor for the expression of osteocalcin. When we transfected pcDNA3.1-human Cbfa-1 into FRTL-5 cells, Cbfa-1 increased the gene expression of alkaline phosphatase, type I collagen, and osteocalcin but suppressed the expression of thyrotropin receptor. We then stained the calcified regions of human papillary thyroid carcinoma tissues with antiosteocalcin antibody and found that malignant cells, as well as follicular epithelial cells, were immunopositive for osteocalcin. Northern blot analysis revealed that the Cbfa-1/Runx2 gene was strongly expressed in tissues from four cases of surgically resected papillary carcinoma.
Conclusions: Thyrocytes share characteristics with osteoblasts. Cbfa-1 may play a role in calcification processes in human thyroid papillary carcinoma tissues.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Several lines of evidence have recently demonstrated that core binding factor-
1 (Cbfa-1)/runt-related transcription factor (Runx)-2, a bone-specific transcription factor, plays an important role in osteoblast differentiation and bone formation (3). This transcription factor stimulates the expression of osteoblast-specific genes, such as osteocalcin, type I collagen, and alkaline phosphatase, that induce osteoprogenitor cells to differentiate to osteoblasts (4).
With regard to nonosteoblastic cells, human smooth vascular cells undergo a spontaneous osteo/chondrocytic conversion and begin expressing Cbfa-1/Runx-2 in vitro (5), and it is hypothesized that progressive changes in the expression of genes encoding bone-associated proteins may be involved in the regulation of vascular mineralization. Jung et al. (6) also reported that osteocalcin mRNA is expressed in nonosseous tissues such as the prostate, skeletal muscle, and ovary, although its physiological roles remain unknown.
The high frequency of calcification in thyroid papillary carcinoma tissues led us to study whether osteocalcin and/or Cbfa-1 genes are expressed in malignant thyroid epithelial cells. We report here that thyrocytes themselves express Cbfa-1 and that papillary carcinoma cells strongly express it.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The BHP 18–21 cell line, established from human thyroid papillary carcinoma, was cultured as described previously (7). FRTL-5 cells (CRL8395; American Type Culture Collection, Rockville, MD) were cultured (8) with (6H) or without 10 mU/ml TSH (Sigma-Aldrich, Inc., St. Louis, MO) for the indicated periods. Human osteosarcoma cells, MG63, and human liver cells, HepG2, were obtained from American Type Culture Collection. Human papillary thyroid cancer tissues and the surrounding normal tissues were obtained by surgery with informed consent. Calcified regions in the tissues were examined by echography and were postoperatively examined by von Kossa staining. Certificated pathologists at our hospital diagnosed the four papillary carcinoma tissues used.
RT-PCR
Total RNAs (10 µg) prepared using RNeasy minikits (QIAGEN Inc., Valencia CA) were reverse transcribed with reverse transcriptase (Takara Shuzo Co., Ltd., Shiga, Japan) in the presence of random primer. PCR primers used were: human osteocalcin (313 bp product), sense, 5'-CGAGACACCA TGAGAGCCCTCACA-3', antisense, 5'-CT AGACCGGGCCGTAGAAGCGCCG-3' (9); human Cbfa-1 (596 bp product), sense, 5'-ACGACAACCGCACCATGGT-3', antisense, 5'-CGGCCTCAGTGAGGGATG-3' (4); rat TSH receptor (392 bp product), sense, 5'-ATGAAGTAGACTGGAGGT-3', antisense, 5'-GGCATCAGGGTCTTGTAAG-3'; rat alkaline phosphatase (178 bp product), sense, 5'-GGAGGGAAGACCAG GTCTG-3', antisense, 5'-CATTTCCAAACAGGGGAC TCGCT-3'; and rat type I collagen (333 bp product), sense, 5'-GGACTTGGGGCAAG ACAGTCATC-3', antisense, 5'-GTCACGT TCAGTTGGTCAAAGAT-3'. RT-PCR efficiency was confirmed in samples by amplifying human glyceraldehyde-3-phosphate dehydrogenase (GAPDH) using sense, 5'-AACGTGTCAGTGGTGGTGGA CCT-3', antisense, 5'-ATGGCCCACATGG GCTCCAA-3'(286 bp product), or rat actin (329 bp product) using sense, 5'-CGTAAA GACCCTATGCCAA-3', antisense, 5'-AGC CATGCCAAAATGTCTCAT-3'. Detection of the BRAF mutation V600E was carried out by the methods of Namba et al. (10) using sense, 5'-TCATAATGCTTGCTCT GATAGG A-3', antisense, 5'-GGCCAAAAA TTTAATCAGTGGA-3'. PCR products were sequenced using a DNA sequencer (ABI PRISM 5'-310; Applied Biosystems, Tokyo, Japan).
EMSA
Preparation of nuclear extracts from cultured cells, and protocols for EMSA were as described elsewhere (11). Synthesized double-stranded oligonucleotide (sense, 5'-CTAGCGAGTATTGTGGTTAA TACG; antisense, 5'-CTAGCGTATTAAC CACAATACTCG) corresponding to the Cbfa-1 binding site on the osteocalcin promoter (12, 13) was labeled with [-32P] dCTP and Klenow fragment. In experiments using the antibody against Cbfa-1 (M-70; Santa Cruz Biotechnology, Inc., Santa Cruz, CA), nuclear extracts were incubated with antiserum or control serum for 30 min at room temperature before adding the labeled probe.
Northern blot analysis
Messenger RNAs (1 µg) from cultured cells or from human thyroid tissues were purified by Oligotex-dT30 (Takara Shuzo). Human Cbfa-1/Runx-2 cDNA in pBluescript was kindly donated by Professor T. Komori (Nagasaki University, Nagasaki, Japan). XhoI/SacII fragment containing the full coding region was ligated into pcDNA 3.1. Human osteocalcin and GAPDH cDNAs, amplified by PCR as above, were subcloned into pcDNA 3.1 and used as probes. Blotted filters were hybridized with 32P-labeled cDNA probes.
Immunohistochemistry
Human thyroid tissues were fixed with 4% paraformaldehyde. After treating sections (5 µm) with 3% H2O2 in methanol, immunostaining was performed with an ABC staining system (Santa Cruz Biotechnology). Antibovine osteocalcin monoclonal antibody, OCG-2, was purchased from Takara Shuzo. To confirm the reaction specificity for osteocalcin, we synthesized an antigen peptide, NH2-EAYR RFYGPV-COOH, which corresponds to amino acids 40–49 of bovine osteocalcin. OCG-2 antibody (10 µg/ml) was preincubated overnight at 4 C with 0.5 mg/ml synthetic peptide and was then used as preabsorbed antibody.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
BHP 18–21 cells, expressing paired box transcription factor-8 (Pax-8) and thyroid transcription factor-1 (TTF-1) (7), retain the characteristics of thyroid epithelial cells. Therefore, we carried out RT-PCR to determine whether osteocalcin mRNA was expressed in these papillary cancer cells. BHP18–21 cells mainly express a 313-bp transcript (Fig. 1A
). Sequence analysis of the PCR product revealed that it lacked intronic sequences (mature type transcript) and thus was completely identical with human osteocalcin cDNA (3). MG63 cells mainly expressed the 313-bp transcript and weakly expressed the 900-bp transcript (Fig. 1A). We were unable to amplify osteocalcin cDNA from HepG2 cells.
|
). Because it was reported that gene expression of osteocalcin depends on a key transcription factor, Cbfa-1 (4), we further studied whether Cbfa-1 mRNA is expressed in these three cell lines. RT-PCR successfully amplified Cbfa-1 cDNA in BHP18–21 and MG63 cells (Fig. 1A), and Northern analyses also showed that the amounts of Cbfa-1 mRNA (4.8 kb) in BHP thyroid cells were higher than that in MG63 cells (Fig. 1B).
To confirm the expression of Cbfa-1 protein in thyrocytes, we performed EMSA using a synthetic oligonucleotide corresponding to the Cbfa-1 binding site on the osteocalcin promoter (12) (Fig. 1C). A major protein-DNA complex was formed with the nuclear extracts from FRTL-5 and BHP18-cells but not with the extracts from HepG2 cells. Formation of the protein-DNA complex from FRTL-5 and BHP18–21 cells was inhibited by homologous unlabeled oligonucleotide. In addition, when anti-Cbfa-1 antibody was added to the reaction mixture, the protein-DNA complex was supershifted. However, particularly in BHP18–21 cells, there is a band near that for Cbfa-1, which is decreased by the competitor but is not supershifted by the antibody, thus suggesting that there is an unknown protein other than Cbfa-1 that binds to the oligonucleotide and that Cbfa-1 is expressed in both cell lines of thyroid epithelial origin. It is noteworthy that the complex formation from BHP18–21cells was more intense than that from FRTL-5 cells and that TSH increased the complex formation in FRTL-5 cells cultured without TSH for 5 d.
Transfection of Cbfa-1 cDNA into FRTL-5 cells
To study the role of Cbfa-1 in thyrocytes, we transfected human Cbfa-1 cDNA into FRTL-5 cells and established a stable cell line. Cbfa-1 increased osteoblast marker genes, alkaline phoshatase, type I collagen, and osteocalcin in stable transformants (Fig. 1D), as is the case in osteoblasts (4). In contrast, however, TSH receptor (TSH-R) gene expression was suppressed by Cbfa-1. On the other hand, when TSH was added to the FRTL-5 cells cultured in the absence of TSH for 7 d, Cbfa-1 mRNA levels increased (Fig. 1C). This increase was apparently more prominent than the results obtained by EMSA assay (Fig. 1C). Although, the amount of Cbfa-1 did not always parallel the DNA binding activity, the difference may be related to the culture time of FRTL-5 cells without TSH.
Expression of Cbfa-1 gene in human papillary thyroid carcinoma tissues
Figure 1F shows the results of Northern blot analysis in four cases of human papillary thyroid carcinoma tissues and their surrounding normal tissues using 32P-human Cbfa-1 cDNA as a probe. Normal tissues express little Cbfa-1 mRNA, but extremely high levels were expressed in all papillary cancer tissues investigated. When we examined the presence of V600E BRAF mutation, one sample (line 5) was positive but other three (lane 6–8) were negative.
Immunostaining of human papillary thyroid cancer tissues with antiosteocalcin monoclonal antibody
We stained severely calcified human papillary cancer tissue with antiosteocalcin antibody. Figure 2B shows malignant cells surrounded by an amorphous substance that was stained with the antibody. This positive staining was eliminated by preincubating the antibody with synthetic antigenic peptide (Fig. 2C). In noncalcified regions, papillary carcinoma cells were positive for osteocalcin, and among these, highly positive cells were observed (Fig. 2E). Osteocalcin-positive cells were also observed in the other three papillary carcinoma tissue samples (data not shown). Interestingly, the follicular epithelial cells of the surrounding normal tissues were also stained by the antibody in the same sections (Fig. 2H), suggesting that both carcinoma cells and follicular epithelial cells express osteocalcin.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
We found that Cbfa-1 is strongly expressed in cultured cells of papillary carcinoma origin, BHP18–21, as well as in surgically resected human thyroid carcinoma tissues. Osteocalcin, another bone-specific gene, was also transcribed in these cells and malignant tissues. A nonspliced form of osteocalcin mRNA has been reported in several nonosseous tissues (6). In contrast, the major osteocalcin gene transcripts in BHP18–21 cells had no intronic sequences (Fig. 2A), indicating that they were the mature bone type. At present, BHP18–21 thyrocytes are the only nonosseous cells known to express mature-type osteocalcin mRNA.
The roles of Cbfa-1 in the thyroid were investigated by transfecting Cbfa-1 cDNA into FRTL-5. Cbfa-1 stimulated the expression of alkaline phoshatase, type I collagen, and osteocalcin, as is the case in osteoblasts (4). These results suggest a sequence of molecular events related to calcification, beginning with the overexpression of Cbfa-1, in papillary carcinoma cells.
Transfection of Cbfa-1 cDNA into FRTL-5 cells suppressed TSH-R gene expression. Conversely, addition of TSH increases Cbfa-1 mRNA levels in the cells. It has reported that TSH-R gene expression depends on TTF-1 in thyrocytes (17). We previously reported that TSH-R mRNA is lower in papillary thyroid cancer than surrounding normal tissues (18). Fabbro et al. (19) showed that TTF-1 mRNA is always detectable in papillary carcinoma. It is therefore unlikely that TTF-1 is the main factor regulating the expression of TSH-R in carcinoma cells. We suspect that Cbfa-1 plays a role in TSH-R gene expression, but further elucidation is required.
|
Footnotes |
|---|
First Published Online April 1, 2008
Abbreviations: Cbfa-1, Core binding factor-1; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; Runx, runt-related transcription factor; TSH-R, TSH receptor; TTF, thyroid transcription factor.
Received December 20, 2007.
Accepted March 20, 2008.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
gene in T cells during thymic development. Mol Cell Biol 15:1662–1670[Abstract]
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
