0013-7227/01/$03.00/0 Endocrinology Copyright © 2001 by The Endocrine Society
Vol. 142, No. 7 Printed in U.S.A.
Expression and Regulation of Type II Iodothyronine Deiodinase in Human Thyroid Gland* MASAMI MURAKAMI, OSAMU ARAKI, YASUHIRO HOSOI, YUJI KAMIYA, TADASHI MORIMURA, TAKAYUKI OGIWARA, HARUO MIZUMA, AND MASATOMO MORI First Department of Internal Medicine, Gunma University School of Medicine, Maebashi 371-8511, Japan ABSTRACT We have studied the expression of type II iodothyronine deiodinase (DII) in human thyroid tumors and cultured human thyroid cells to elucidate the mechanisms involved in the regulation of DII expression in human thyroid gland. Three cases with hyperfunctioning thyroid adenoma, including a case that showed an activating mutation of Gs␣ with a constitutive activation of cAMP production in cultured cells, and six cases with papillary thyroid carcinoma were analyzed in the present study. Free T3 was increased, whereas free T4 was within the normal range in all patients with hyperfunctioning thyroid adenoma. Thyroid tumor tissue and surrounding nontumor tissue were obtained at the time of surgery, and DII expression was compared between tumor tissue and nontumor tissue in each case. Northern analysis demonstrated the presence of DII messenger RNA (mRNA) approximately 7.5 kb in size in all of the tumor and nontumor tissues.
T
NEEDS TO be converted to T3 by iodothyronine deiodinase to exert its full biological activity (1). Type I iodothyronine deiodinase (DI) and type II iodothyronine deiodinase (DII) catalyze T4 activation. DI is present in thyroid gland, liver, kidney, and many other tissues, whereas DII is present in a limited number of tissues, including brain, pituitary, brown adipose tissue, and pineal gland in the rat. DI activity is known to decrease in the hypothyroid state and is believed to have a primary role in maintaining circulating T3 levels. DII activity, in contrast, increases in the hypothyroid state and plays a critical role in providing local intracellular T3. The Km of DII is approximately 2 nm for T4, which is 100 times lower than that of DI. Although DI activity is significantly inhibited by 6-propyl-2-thiouracil (PTU), DII activity is insensitive to PTU inhibition (1). Recently, a complementary DNA (cDNA) encoding DII was cloned from Rana catesbeiana tissues (2), and its mammalian counterpart was isolated (3). Subsequently, DII messenger RNA (mRNA) was unexpectedly detected in thyroid 4
Received February 13, 2001. Address all correspondence and requests for reprints to: Masami Murakami, M.D., First Department of Internal Medicine, Gunma University School of Medicine, Maebashi 371-8511, Japan. E-mail:
[email protected]. * This work was supported in part by a Grant-in-Aid for Scientific Research 09671024 (to M.M.) from the Ministry of Education, Science, Sports, and Culture, Japan, and a grant from the Ministry of Health and Welfare, Japan. Presented in part at the 71st Annual Meeting of the American Thyroid Association, Portland, Oregon, 1998. The sequence of the genomic clone has been submitted to DDBJ/EMBL/GenBank (AB039888).
DII mRNA and DII activity in hyperfunctioning thyroid adenoma were significantly increased compared with those in nontumor tissue in each case. In contrast, DII mRNA and DII activity in papillary thyroid carcinoma were decreased compared with those in nontumor tissue in each case. DII mRNA and DII activity in cultured human thyroid cells were significantly stimulated by TSH in a dose-dependent manner. The promoter activity of the human DII gene including the complete cAMP response element, transfected to cultured human thyroid cells, was stimulated by (Bu)2cAMP. In summary, these results suggest that DII expression in human thyroid gland is regulated at the transcriptional level through the TSH receptor-Gs␣-cAMP regulatory cascade, which may be related to the increase in circulating T3 level in patients with Graves’ disease and hyperfunctioning thyroid adenoma. (Endocrinology 142: 2961– 2967, 2001)
gland, skeletal muscle, and heart in humans (4, 5). DII mRNA levels were reported to be especially high in thyroids from patients with Graves’ disease and in follicular thyroid adenomas (4). These observations suggest previously unrecognized roles of DII in those tissues, including a possible contribution to circulating T3 levels (4, 5). It appears, therefore, important to study the mechanisms involved in the regulation of DII expression in those tissues. In the present report we analyzed DII expression in thyroid tumor tissues and normal surrounding tissues from patients with hyperfunctioning thyroid adenoma and papillary thyroid carcinoma and studied the regulation of DII expression in cultured human thyroid cells to elucidate the mechanisms involved in the regulation of DII expression in human thyroid gland. Materials and Methods Patients Table 1 shows the profiles of patients with thyroid tumor studied in the present report. Three patients (cases 1–3) with hyperfunctioning thyroid adenoma and six patients (cases 4 –9) with papillary thyroid carcinoma were studied. All of the patients with hyperfunctioning thyroid adenoma were demonstrated to have hot nodules by 123I scintigraphy. In case 1, primary culture of the cells from hyperfunctioning thyroid adenoma which showed an activating mutation of Gs␣ and that of the cells from surrounding nontumor tissue were performed. As previously described, constitutive activation of cAMP production was demonstrated in cultured cells obtained from hyperfunctioning thyroid adenoma tissue compared with the cells from surrounding nontumor tissue in this case (6). Informed consent was obtained from the patients for the use of thyroid tumor tissues. Use of the tissues did not adversely affect the clinical diagnosis or treatment of the patients.
2961
2962
Endo • 2001 Vol. 142 • No. 7
THYROID DII
TABLE 1. Clinical findings of patients with hyperfunctioning thyroid adenoma and papillary thyroid carcinoma Case no.
Age (yr)
Sex
Free T3 (2.4 – 4.3 pg/ml)a
Free T4 (0.81–2.13 ng/dl)a
TSH (0.34 –3.5 U/ml)a
Diagnosis
1 2 3 4 5 6 7 8 9
50 45 32 42 81 77 57 56 55
F F F F F F F F F
6.6 7.0 6.4 3.3 3.4 3.8 3.8 3.3 4.0
1.3 2.0 1.4 0.9 1.4 1.1 1.2 1.0 1.1
⬍0.05 ⬍0.05 ⬍0.05 1 0.8 0.6 1.4 2.7 2.1
Hyperfunctioning adenoma Hyperfunctioning adenoma Hyperfunctioning adenoma Papillary carcinoma Papillary carcinoma Papillary carcinoma Papillary carcinoma Papillary carcinoma Papillary carcinoma
a
Normal range is in parentheses.
Materials and methods [␣- P]UTP, [␣- P]deoxy-CTP, [ I]rT3, and [ I]T4 were purchased from NEN Life Science Products (Boston, MA). LH-20 was obtained from Pharmacia Biotech (Uppsala, Sweden). AG 50W-X2 resin and protein assay kit were purchased from Bio-Rad Laboratories, Inc. (Hercules, CA). T7 RNA polymerase and SP6 RNA polymerase were obtained from Nippon Gene (Tokyo, Japan). All other chemicals at the highest quality were obtained from Sigma (St. Louis, MO) or Wako Pure Chemical Industries, Ltd. (Osaka, Japan) unless otherwise indicated. 32
32
125
125
Hormone measurements Serum free T3 and free T4 were measured by enzyme immunoassay using Glaozyme [N] FT3-(S) and [N] FT4-(M) kits (SANYO Chemical Industries, Kyoto, Japan), respectively. Serum TSH was measured by two-site enzyme immunoassay using Glaozyme [New] TSH kit (SANYO Chemical Industries). Sensitivities for free T3, free T4, and TSH were 0.2 pg/ml, 0.04 ng/dl, and 0.014 U/ml, respectively. Intra- and interassay coefficients of variation were less than 4% and 8%, respectively, for each assay.
Tissue preparation Thyroid tissues were obtained from tumor tissues and surrounding nontumor tissues of the patients with hyperfunctioning thyroid adenoma and papillary thyroid carcinoma at the time of surgery for the measurement of deiodinase activity and RNA preparation. Thyroid tissues were also obtained from the patients with Graves’ disease at the time of subtotal thyroidectomy for the primary culture of thyroid cells.
Cultured thyroid cells from each well were washed twice with the washing buffer (100 mm potassium phosphate, pH 7.0), scraped off, and transferred into 500 l ice-cold buffer (100 mm potassium phosphate, pH 7.0, containing 20 mm dithiothreitol). After centrifugation at 3,000 rpm for 15 min, the supernatants were discarded. Pellets were sonicated in 100 l sonication buffer (100 mm potassium phosphate, pH 7.0, containing 20 mm dithiothreitol and 1 mm EDTA) per tube. Iodothyronine deiodinase activity was measured as previously described (9). In brief, the samples were incubated with 2 nm [125I]T4, which was purified using LH-20 column chromatography on the day of experiment, in the incubation buffer (100 mm potassium phosphate, pH 7.0, containing 20 mm dithiothreitol, 1 mm EDTA, and 1 mm PTU) for 1 h at 37 C in duplicate. The reaction was terminated by adding 100 l 2% BSA and 800 l 10% trichloroacetic acid. After centrifugation at 3,000 rpm for 10 min, the supernatant was applied to a small column packed with AG 50W-X2 resin (bed volume, 1 ml) and eluted with 2 ml 10% glacial acetic acid. Separated 125I was counted with a ␥-counter. Nonenzymatic deiodination was corrected by subtracting I⫺ released in tissue-free tubes. The protein concentration was determined by Bradford’s method using BSA as a standard (10). The deiodinating activity was calculated either as the percentage of I⫺ released or as picomoles of I⫺ released per mg protein/h after multiplication by a factor of 2 to correct random labeling at the equivalent 3⬘ and 5⬘ positions. In some experiments DII activity was also measured by the difference between tracer iodide released at 2 and 100 nm T4; the latter concentration was shown to saturate DII activity, but not DI activity (4). Equimolar concentrations of I⫺ and T3 were produced by deiodination of T4 as assessed by descending paper chromatography (hexane-tertiary amyl alcohol-2 n ammonia), as previously described (9).
Human thyroid cell culture
RNA preparation and Northern analysis
Thyroid tissues obtained from the patients with Graves’ disease were dispersed by digestion with collagenase (CLS 2,Worthington Biochemical Corp., Freehold, NJ) and dispase (Life Technologies, Inc., Gaithersburg, MD) as previously described (6, 7). Subsequently, thyroid cells were cultured in an atmosphere of 5% CO2 and 95% air at 37 C in Coon’s modified Ham’s F-12 medium supplemented with 5% calf serum, penicillin (50 U/ml), streptomycin (50 U/ml), and a six-hormone mixture containing bovine TSH (1 mU/ml), insulin (10 g/ml), hydrocortisone (5 nm), transferrin (5 g/ml), glycyl-l-histidyl-l-lysine-acetate (10 ng/ ml), and somatostatin (10 ng/ml). TSH is required to reorganize thyroid cells into three-dimensional and functionally differentiated thyroid follicles (8). After 24 h of culture, the medium was replaced with a medium without TSH, and then the cells were cultured for 3 days before the experiment.
Total RNA was isolated from thyroid tumor tissues, surrounding nontumor tissues, or cultured thyroid cells from each well by the modified acid guanidinium thiocyanate phenol-chloroform method according to Chomczynski and Sacchi (11). Northern analyses were performed as previously described (9). Human DII cDNA fragment containing residues 110-1051 (3) and human glyceraldehyde-3-phosphate dehydrogenase (G3PDH) cDNA fragment containing residues 71–1053, which were cloned into pCRII (Invitrogen, San Diego, CA), were used to synthesize complementary RNA (cRNA) probes with [␣-32P]UTP and SP6 RNA polymerase or T7 RNA polymerase, respectively. Ten micrograms of total RNA per lane were electrophoresed on a 1.4% agarose gel containing 2 m formaldehyde and transferred overnight in 20 ⫻ SSC (1 ⫻ SSC ⫽ 150 mm sodium chloride and 15 mm trisodium citrate) to a nylon membrane (Biodyne, Pall BioSupport Corp., East Hills, NY). RNA was cross-linked to the nylon membrane with a UV Stratalinker (Stratagene, La Jolla, CA). The membrane was prehybridized with the hybridization buffer (50% formamide, 0.2% SDS, 5% dextran sulfate, 50 mm HEPES, 5 ⫻ SSC, 5 ⫻ Denhart’s solution, and 250 g/ml denatured salmon sperm DNA) at 68 C for 2 h. Subsequently, the membrane was hybridized at 68 C overnight with the hybridization buffer containing a human DII cRNA probe. The membrane was washed twice in 2 ⫻ SSC/0.1% SDS at 25 C for 15 min and twice in 0.1 ⫻ SSC/0.1% SDS at 68 C for 1 h. Autoradiography was established by exposing the filters for 6 –24 h to x-ray film (Kodak XAR-2, Eastman Kodak Co., Rochester, NY) at ⫺70 C. After the detection of DII mRNA, the probe was stripped off, and blots
Measurement of deiodinase activity Thyroid microsomal fractions were obtained from tumor tissues or surrounding nontumor tissues as previously described (4). Briefly, thyroid tissues were homogenized in the homogenization buffer (50 mm Tris-HCl and 0.25 m sucrose, pH 7.5, containing 20 mm dithiothreitol and 1 mm EDTA). Homogenates were centrifuged at 20,000 ⫻ g for 20 min at 4 C, and the resultant supernatants were centrifuged at 200,000 ⫻ g for 90 min at 4 C to prepare microsomal fractions. Microsomal pellets were resuspended in the homogenization buffer and stored at ⫺70 C.
THYROID DII were rehybridized with human G3PDH cRNA probe as a control. Hybridization and washing were performed as described above, and the membrane was exposed for 1 h. mRNA levels were quantitated by densitometry using NIH Image (version 1.61), and the optical density of the DII band 7.5 kb in length was corrected for G3PDH. RNA samples for comparison were analyzed on the same blot.
Isolation of genomic DNA of human DII and subcloning of the human DII 5⬘-flanking region into the luciferase expression vector Partial human DII cDNA (residues 110 –1051) was labeled with [␣-32P]deoxy-CTP by the random primer method. The probe was used to screen a human placenta FIXII genomic DNA library (Stratagene) and positive clones were obtained. Sequencing analysis revealed that one of the clones was made up of human DII cDNA residues 1–356, the preceding 12.8-kb 5⬘-flanking region, and the subsequent 6.8-kb intron, as described previously (12). To create the human DII promoter-luciferase construct, the 832-, 784-, and 693-bp fragments immediately upstream of the translation start site were generated by PCR using a human DII genomic clone in Fix II (Stratagene) as a template. Forward primers including the KpnI site (underlined) were 5⬘-ATAGGTACCATCCTGGCCAAAGTAAAG-3⬘ (832 bp), 5⬘-ATAGGTACCCCAAGATTAGGCTTTCACT-3⬘ (784 bp), and 5⬘-ATAGGTACCACTTTGCACCACAGACAG-3⬘ (693 bp), and the reverse primer including a XhoI site (underlined) was 5⬘-ATACTCGAGCTTCTCTGCCTCCTGAGT-3⬘. The resulting single PCR fragment was subcloned into the pGL3 basic luciferase expression vector (Promega Corp., Madison, WI), between the XhoI and KpnI sites. The identity and orientation of the PCR fragment were then characterized by restriction enzyme analysis and DNA sequencing. Large scale plasmid DNA purification was performed using Maxiprep kit (QIAGEN, Valencia, CA).
Transient transfection and dual luciferase activity assay The human DII promoter-luciferase constructs were transfected to cultured human thyrocytes with Lipofectamine reagent (Life Technologies, Inc.), following the protocol provided by the manufacturer. Each transfection was performed using 0.45 g firefly luciferase reporter construct DNA and 0.05 g of an internal control plasmid pRL-TK (which contains a herpes simplex virus thymidine kinase promoter upstream of the Renilla luciferase gene; Promega Corp.). Five hours after transfection, the transfection medium was replaced by culture medium. Twenty-four hours after transfection, where indicated, 1 mm (Bu)2cAMP was added to the culture medium. After 24 h, the medium was removed, wells were rinsed with PBS, and then the cells were lysed in passive lysis buffer (Promega Corp.) at room temperature for 30 min. Culture plates were subjected to one cycle of freezing (⫺80 C) and thawing (room temperature) to ensure complete cell lysis. The samples were transferred to microcentrifuge tubes and then centrifuged at 12,000 ⫻ g for 30 sec at 4 C. Supernatants were used for assay of luciferase activities. Firefly and Renilla luciferase activities were sequentially measured using the Dual Luciferase Reporter Assay System (Promega Corp.) according to the manufacturer’s instructions. Firefly luciferase activities were normalized based on the Renilla luciferase activity in each well.
2963
serum TSH was undetectable in all patients with hyperfunctioning thyroid adenoma. DII activity and DII mRNA in thyroid tumor tissues and surrounding nontumor tissues
Because not only DII, but also DI, is expressed in human thyroid gland, it is important to distinguish those activities (4). We have attempted to distinguish DII activities from DI activities by two methods. First, T4 deiodinating activity was measured as the release of I⫺ from various concentrations of T4 in the presence of 1 mm PTU in the microsomal fraction of hyperfunctioning thyroid adenoma of case 2 (Table 1). As shown in Fig. 1, a double reciprocal plot of deiodinating activity demonstrated that the Km for T4 was 1.24 nm, and the maximum velocity (Vmax) was 4.88 pmol/mg protein䡠h. A low Km of T4 deiodinating activity is compatible with DII activity. Km and Vmax are comparable to those reported in the previous study using thyroid tissue from a patient with Graves’ disease (4). The Vmax of 4.88 pmol/mg protein䡠h is more than 5-fold that of human skeletal muscle, human placenta, or human epidermal keratinocytes (5, 13). DII activity was then measured as the release of I⫺ from 2 nm T4 in the presence of 1 mm PTU in the microsomal fractions of tumor tissues and surrounding nontumor tissues (Table 2). Although DII activity was greater in the tissue of hyperfunctioning thyroid adenoma than in surrounding nontumor tissue in each case, DII activity was lower in the tissue of papillary thyroid carcinoma than in surrounding nontumor tissue in each case. Thus, the DII activity ratio (tumor tissue vs. nontumor tissue) was greater than 1 in patients with hyperfunctioning thyroid adenoma and less than 1 in patients with papillary thyroid carcinoma, as shown in Table 2. Because it has been suggested that PTU is a poor inhibitor of DI at low substrate levels, we also measured DII activity in some tissues as the difference in fractional T4 deiodination between low (2 nm) and high (100 nm) substrate levels; the latter concentration has been shown to saturate DII activity, but not DI activity (4). As the data shown in parentheses in Table 2 indicate, the results are in good agreement with those obtained by the release of I⫺ from 2 nm T4 in the presence of 1 mm PTU, and DII activities measured by two different
Statistics All values are expressed as the mean ⫾ se. Statistical differences were evaluated by ANOVA, followed by Dunnett’s test using StatView 5.0 (Abacus Concepts, Inc., Berkeley, CA).
Results Serum free thyroid hormones and serum TSH in the patients with thyroid tumor
As shown in Table 1, serum free thyroid hormones and serum TSH were within the normal range in all patients with papillary thyroid carcinoma. In contrast, serum free T3 was increased, serum free T4 was within the normal range, and
FIG. 1. Double reciprocal plot of iodothyronine deiodinase activity in hyperfunctioning thyroid adenoma. Iodothyronine deiodinase activity was measured in a microsomal fraction of hyperfunctioning thyroid adenoma in case 2 using various concentrations of [125I]T4 in the presence of 1 mM PTU as described in Materials and Methods.
2964
Endo • 2001 Vol. 142 • No. 7
THYROID DII
TABLE 2. Type II iodothyronine deiodinase (DII) activity in tumor tissue and surrounding nontumor tissue Case no.
DII activity in tumor (pmol/mg protein 䡠 h)
DII activity in nontumor (pmol/mg protein 䡠 h)
DII activity ratio (tumor/nontumor)
Diagnosis
1 2 3 4 5 6 7 8 9
6.67 4.23 (5.12) 1.28 0.04 0.04 0.12 (0.05) 0.17 (0.52) 0.12 (0.02) 0.23
2.78 0.51 (0.39) 0.30 1.09 0.10 0.33 (0.33) 0.65 (1.26) 0.63 (0.21) 1.63
2.40 8.29 (13.12) 4.27 0.04 0.40 0.36 (0.15) 0.26 (0.41) 0.19 (0.10) 0.14
Hyperfunctioning adenoma Hyperfunctioning adenoma Hyperfunctioning adenoma Papillary carcinoma Papillary carcinoma Papillary carcinoma Papillary carcinoma Papillary carcinoma Papillary carcinoma
DII activity was measured as the release of I⫺ from 2 nM T4 in the presence of 1 mM PTU. DII activity was also measured as the difference in fractional T4 deiodination between low (2 nM) and high (100 nM) substrate levels in the absence of PTU, and the data are shown in parentheses.
FIG. 2. Northern analysis of DII mRNA in tumor tissue (T) and surrounding nontumor tissue (N) in hyperfunctioning thyroid adenoma (A) and papillary thyroid carcinoma (B). Total RNA was isolated separately from tumor tissues and nontumor tissues, electrophoresed, transferred to the nylon membrane, and hybridized with cRNA probes of human DII and human G3PDH. The exposure time for DII mRNA was 24 h. Each lane represents 10 g total RNA obtained from each tissue.
methods show a significant positive correlation (r ⫽ 0.983; P ⬍ 0.001). Figure 2 shows Northern analysis of DII mRNA in tumor tissues and nontumor tissues from patients with hyperfunctioning thyroid adenoma and papillary thyroid carcinoma. Although the results of 24-h exposure are shown, a longer exposure clearly demonstrated the DII mRNA with approximately 7.5 kb in all tumor tissues and nontumor tissues. Although DII mRNA was greater in the tissue of hyperfunctioning thyroid adenoma than in nontumor tissue in each case, DII mRNA was lower in the tissue of papillary thyroid carcinoma than in nontumor tissue in each case. Thus, the DII mRNA ratio (tumor tissue vs. nontumor tissue) was greater than 1 in patients with hyperfunctioning thyroid
adenoma and less than 1 in patients with papillary carcinoma, as shown in Fig. 3. Effect of TSH on DII expression in cultured human thyroid cells
In the next experiment the effects of TSH on DII mRNA and DII activity in cultured thyroid cells obtained from patients with Graves’ disease were studied. Cultured human thyroid cells were incubated with various concentrations of TSH for 6 h. The DII mRNA level corrected for G3PDH mRNA in cultured thyroid cells was increased by TSH in a dose-dependent manner as shown in Fig. 4, A and B. DII activity in cultured thyroid cells was also significantly in-
THYROID DII
2965
FIG. 3. DII mRNA in tumor tissue and nontumor tissue of hyperfunctioning thyroid adenoma and papillary thyroid carcinoma. The OD of the DII mRNA band was corrected for G3PDH mRNA, and the DII mRNA ratio (tumor tissue/nontumor tissue) was determined.
creased by TSH in a dose-dependent manner as shown in Fig. 4B. These results indicate that DII expression in the human thyroid gland is regulated by the TSH receptor-mediated pathway. Analysis of cAMP response element in DII promoter in cultured human thyroid cells
To elucidate the mechanisms involved in the regulation of DII expression by the TSH receptor-mediated pathway, we studied whether DII promoter activity is stimulated by cAMP in cultured human thyroid cells. We isolated a human DII genomic clone that contains 12.8 kb of the 5⬘-flanking region (12). As shown in Fig. 5A, a complete cAMP response element (TGACGTCA) was observed ⫺801 to ⫺794 upstream of the translation start site. Human DII promoterluciferase constructs including the 832-, 784-, and 693-bp fragments upstream of the translation start site were generated, as shown in Fig. 5B and transfected to cultured human thyroid cells. The 832-bp construct including the cAMP response element demonstrated stimulation by (Bu)2cAMP, whereas other constructs lacking the cAMP response element were not stimulated by (Bu)2cAMP, as shown in Fig. 5C. These results indicate that the functional cAMP response element present in the 5⬘-flanking region of the human DII gene and human DII is transcriptionally regulated by a cAMP-dependent mechanism in human thyroid cells. Discussion
In the present study DII activity and DII mRNA were detected in all tissues from hyperfunctioning thyroid adenoma and papillary thyroid carcinoma. It was clearly demonstrated that DII activity and DII mRNA were increased in the tissue of hyperfunctioning adenoma compared with surrounding nontumor tissue in each case. In contrast, DII activity and DII mRNA were decreased in the tissue of papillary thyroid carcinoma compared with surrounding nontumor tissue in each case. It has been reported that the
FIG. 4. A, Northern analysis of DII mRNA in cultured human thyroid cells stimulated by TSH. Total RNA was isolated from cultured human thyroid cells stimulated by various concentrations of TSH for 6 h, electrophoresed, transferred to the nylon membrane, and hybridized with cRNA probes of human DII and human G3PDH. Each lane represents 10 g total RNA obtained from cells in each well. B, DII mRNA and DII activity in cultured human thyroid cells stimulated by TSH. The OD of the DII mRNA band was corrected for G3PDH mRNA, and DII mRNA was expressed as percentage of DII mRNA in control cells. DII activity was measured in sonicate of cells that were stimulated by various concentrations of TSH for 6 h. DII activity in control cells was 0.26 pmol/mg protein䡠h, and DII activity was expressed as a percentage of that in control cells. Each bar represents the mean ⫾ SE of four wells. Statistical differences were evaluated by ANOVA, followed by Dunnett’s test of comparison to a control. *, P ⬍ 0.01.
cAMP regulatory cascade is constitutively activated in hyperfunctioning thyroid adenoma, mostly due to an activating mutation in the TSH receptor gene or Gs␣ gene (14), and constitutive activation of cAMP production has been demonstrated in cultured cells from hyperfunctioning thyroid adenoma that showed an activating mutation of Gs␣ compared with cells from surrounding nontumor tissue in one of the cases analyzed in the present study (6). The present results, therefore, strongly suggest that DII activity and DII mRNA are increased as a result of constitutive activation of the cAMP regulatory cascade in hyperfunctioning thyroid adenoma. To avoid the variations in the results of DII expression among thyroid tissue samples because of possible difference
2966
THYROID DII
Endo • 2001 Vol. 142 • No. 7
FIG. 5. A, Partial sequence of the 5⬘flanking region of the human DII gene. The translation start site is designated ⫹1. The potential cAMP response element and TATA box are underlined. B, Schematic illustrations of human DII promoter-luciferase reporter constructs. The potential cAMP response element and TATA box are indicated as boxes. C, (Bu)2cAMP induction of promoter activity of human DII promoterluciferase reporter constructs. Firefly luciferase activity was corrected for Renilla luciferase activity expressed by the herpes simplex virus-thymidine kinase promoter as an internal control. The ratio of the luciferase activity in cultured human thyroid cells transfected with constructs incubated with 1 mM (Bu)2cAMP and without (Bu)2cAMP is shown. The basal levels of firefly luciferase activity corrected for Renilla luciferase activity of DII 0 Luc, DII ⫺693 Luc, DII ⫺784 Luc, and DII ⫺832 Luc were 0.06, 0.37, 0.32, and 0.15 arbitrary units, respectively. Each bar represents the mean ⫾ SE of three wells.
in the procedure of surgery or the preparation of samples, we compared DII expression in tumor tissue and surrounding nontumor tissue in each case. Although it has been suggested that DII activity and DII mRNA are increased in hyperfunctioning thyroid adenoma (4), DII activity and DII mRNA in surrounding nontumor tissue of hyperfunctioning thyroid adenoma are not known. In the present study DII activity and DII mRNA were decreased in surrounding nontumor tissue compared with those in tissue from hyperfunctioning thyroid adenoma in each case. These results indicate that DII activity and DII mRNA are increased only in the tissue of hyperfunctioning thyroid adenoma and not in surrounding nontumor tissue. It is noteworthy that only serum free T3 was increased, and serum free T4 was within the normal range in all patients with hyperfunctioning thyroid adenoma, whereas serum free T3 and serum free T4 were within the normal range in all patients with papillary thyroid carcinoma in the present study. These results suggest that increased T3 production by DII in hyperfunctioning thyroid adenoma may contribute to elevated circulating T3 levels (4). The present results demonstrated, for the first time, that DII activity and DII mRNA were decreased in the tissue from papillary thyroid carcinoma compared with surrounding nontumor tissue in each case. Therefore, there appears to be a striking difference in DII expression between hyperfunc-
tioning thyroid adenoma and papillary thyroid carcinoma. The decrease in DII activity and DII mRNA in papillary thyroid carcinoma may be associated with dedifferentiation or malignant transformation of thyroid follicular cells. We performed primary culture of human thyroid cells obtained from patients with Graves’ disease to study the regulation of DII expression in human thyroid gland. DII activity and DII mRNA were detected in cultured human thyroid cells and were stimulated by TSH in a dose-dependent manner. These results suggest that DII expression in human thyroid cells is regulated by the TSH receptormediated mechanism, which involves the cAMP regulatory cascade. Because serum TSH is suppressed in patients with hyperfunctioning thyroid adenoma, constitutive activation of the TSH receptor-cAMP regulatory cascade, presumably due to an activating mutation in the TSH receptor gene or the Gs␣ gene, is responsible for the increased expression of DII in hyperfunctioning thyroid adenoma tissue in the present study. Analysis of the promoter region of human DII gene revealed the presence of a complete cAMP response element (TGACGTCA) ⫺801 to ⫺794 upstream of the translation start site of the gene. The 832-bp human DII promoter-luciferase construct including the cAMP response element transfected to cultured human thyroid cells was stimulated by (Bu)2cAMP, whereas other constructs lacking the cAMP re-
THYROID DII
sponse element transfected to cultured human thyroid cells were not stimulated by (Bu)2cAMP. These results indicate that the functional cAMP response element is present in the human DII gene, and human DII expression is transcriptionally regulated by a cAMP-dependent mechanism in human thyroid cells. Recently, cAMP-mediated mechanisms have been reported to be involved in the pretranslational regulation of DII expression in human skeletal muscle cells (9), rat astrocytes (15), and rat pineal gland (16). In human skeletal muscle cells, cAMP-stimulated DII expression was demonstrated to be suppressed by thyroid hormones at both pretranslational and posttranslational levels (9). It is, therefore, of considerable interest to investigate the possible negative regulation of DII expression by thyroid hormones in human thyroid gland in further studies. Recently, the complete sequence of the human DII gene has appeared in GenBank (AC007372), and Bartha et al. (17) demonstrated the presence of functional cAMP response element in human DII promoter using HEK-293 cells. In conclusion, these results suggest that human thyroid DII expression is regulated transcriptionally through the TSH receptor ␣-subunit of Gs protein-cAMP regulatory cascade, which may be related to the increase in circulating T3 levels in patients with Graves’ disease and hyperfunctioning thyroid adenoma.
4. 5. 6.
7. 8. 9. 10. 11. 12.
13.
14.
Acknowledgments The authors are indebted to Drs. Takayuki Kasahara, Tetsuo Negishi, and Makoto Imamura for technical assistance and useful discussion.
15.
References 1. Larsen PR 1996 Thyroid hormone transport, cellular uptake, metabolism and molecular action. In: DeGroot LJ, Larsen PR, Hennemann G (eds) The Thyroid and Its Disease. Churchill Livingstone, New York, pp 61–111 2. Davey JC, Becker KB, Schneider MJ, St Germain DL, Galton VA 1995 Cloning of a cDNA for the type II iodothyronine deiodinase. J Biol Chem 270:26786 –26789 3. Croteau W, Davey JC, Galton VA, St Germain DL 1996 Cloning of the mammalian type II iodothyronine deiodinase: a selenoprotein differentially
16.
17.
2967
expressed and regulated in human and rat brain and other tissues. J Clin Invest 98:405– 417 Salvatore D, Tu H, Harney JW, Larsen PR 1996 Type 2 iodothyronine deiodinase is highly expressed in human thyroid. J Clin Invest 98:962–968 Salvatore D, Tibor B, Harney JW, Larsen PR 1996 Molecular biology and biochemical characterization of the human type 2 selenodeiodinase. Endocrinology 137:3308 –3315 Kamiya Y, Murakami M, Yanagita Y, Koitabashi H, Nagamachi Y, Hosoi Y, Ogiwara T, Mizuma H, Iriuchijima T, Mori M 1998 Primary culture of cells from hyperfunctioning thyroid adenoma with an activating mutation of G␣s. Mol Cell Endocrinol 138:137–142 Williams DW, Wynford-Thomas D 1989 Animal cell culture: human thyroid epithelial cells. In: Pollard JW, Walker JM (eds) Methods in Molecular Biology. Humana Press, Totowa, NJ, vol 5:139 –149 Inoue K, Horiuchi R, Kondo Y 1980 Effect of thyrotropin on cell orientation and follicle reconstruction in rotated suspension culture of hog thyroid cells. Endocrinology 107:1162–1168 Hosoi Y, Murakami M, Mizuma H, Ogiwara T, Imamura M, Mori M 1999 Expression and regulation of type II iodothyronine deiodinase in cultured human skeletal muscle cells. J Clin Endocrinol Metab 84:3293–3300 Bradford MM 1976 A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 72:248 –254 Chomczynski P, Sacchi N 1987 Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal Biochem 162:156 –159 Araki O, Murakami M, Morimura T, Kamiya Y, Hosoi Y, Kato Y, Mori M 1999 Assignment of type II iodothyronine deiodinase gene (DIO2) to human chromosome band 14q24.23q24.3 by in situ hybridization. Cytogenet Cell Genet 84:73–74 Kaplan MM, Pan C, Gordon PR, Lee J-K, Gilchrest BA 1988 Human epidermal keratinocytes in culture convert thyroxine to 3,5,3⬘-triiodothyronine by type II iodothyronine deiodination: novel endocrine function of the skin. J Clin Endocrinol Metab 66:815– 822 Parma J, Duprez L, Van Sande J, Hermans J, Rochmans P, Van Vilet G, Costagliola S, Rodien P, Dumont JE, Vassart G 1997 Diversity and prevalence of somatic mutations in the thyrotropin receptor and Gs␣ genes as a cause of toxic thyroid adenomas. J Clin Endocrinol Metab 82:2695–2701 Pallud S, Lennon AM, Ramauge M, Gavaret JM, Croteau W, Pierre M, Courtin F, St Germain DL 1997 Expression of the type II iodothyronine deiodinase in cultured rat astrocytes is selenium-dependent. J Biol Chem 272:18104 –18110 Kamiya Y, Murakami M, Araki O, Hosoi Y, Ogiwara T, Mizuma H, Mori M 1999 Pretranslational regulation of rhythmic type II iodothyronine deiodinase expression by -adrenergic mechanism in the rat pineal gland. Endocrinology 140:1272–1278 Bartha T, Kim S-W, Salvatore D, Gereben B, Tu HM, Harney JW, Rudas P, Larsen PR 2000 Characterization of the 5⬘-flanking and 5⬘-untranslated regions of the cyclic adenosine 3⬘,5⬘-monophosphate-responsive human type 2 iodothyronine deiodinase gene. Endocrinology 140:229 –237
