help button home button Endocrine Society Endocrinology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Endocrinology, doi:10.1210/en.2004-0356
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
145/9/4251    most recent
Author Manuscript (PDF)
Right arrow Purchase Article
Right arrow View Shopping Cart
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Copyright Permission
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Wassen, F. W. J. S.
Right arrow Articles by Kuiper, G. G. J. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Wassen, F. W. J. S.
Right arrow Articles by Kuiper, G. G. J. M.
Endocrinology Vol. 145, No. 9 4251-4263
Copyright © 2004 by The Endocrine Society

Characteristics and Thyroid State-Dependent Regulation of Iodothyronine Deiodinases in Pigs

Frank W. J. S. Wassen, Willem Klootwijk, Ellen Kaptein, Dirk J. Duncker, Theo J. Visser and George G. J. M. Kuiper

Department of Internal Medicine (F.W.J.S.W, G.G.J.M.K, E.K., W.K., T.J.V.) and Department of Experimental Cardiology (D.J.D.), Erasmus Medical Center, 3015 GE Rotterdam, The Netherlands

Address all correspondence and requests for reprints to: George Kuiper, Department of Internal Medicine, Room Ee 502, Erasmus Medical Center, Dr Molewaterplein 50, 3015 GE Rotterdam, The Netherlands. E-mail: g.kuiper{at}erasmusmc.nl.


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
Three iodothyronine deiodinases (D1, D2, and D3) regulate local and systemic availability of thyroid hormone. D1 and D2 activate the prohormone T4 to the thyromimetic T3, and D3 inactivates T4 and T3 to rT3 and 3,3'-diiodothyronine, respectively. The expression of the three deiodinases is tightly regulated with regard to developmental stage and cell type to provide fine tuning of T3 supply to target cells. Most studies regarding distribution and regulation of deiodinases have been carried out in rodents. However, in different respects, rodents do not seem to be the optimal experimental model for human thyroid hormone physiology. For instance, D2 expression has been observed in human thyroid and skeletal muscle but not in these tissues in rodents. In this study, we have explored the pig as an alternative model. Porcine D1, D2, and D3 were cloned by RT-PCR, and their catalytic properties were shown to be virtually identical to those reported for human and rodent deiodinases. The tissue distribution of deiodinases was studied in normal pigs and in pigs made hypothyroid by methimazole treatment or in pigs made hyperthyroid by T4 treatment. D1 activity in liver and kidney was increased in T4-treated pigs. D2 activities in cerebrum and pituitary were decreased after T4 treatment and strongly increased after methimazole treatment. Remarkably, D2 activity in thyroid and skeletal muscle was induced in hypothyroid pigs. Significant expression of D3 was observed in cerebrum and was positively regulated by thyroid state. In conclusion, the pig appears to be a valuable model for human thyroid hormone physiology. The expression of D2 activity in thyroid and skeletal muscle is of particular interest for studies on the importance of this enzyme in (hypothyroid) humans.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
THYROID HORMONE IS essential for growth, development, and regulation of energy metabolism (1, 2, 3). Particularly well known is the critical role of thyroid hormone in the development and function of the central nervous system (4). Thyroid hormone is produced by the thyroid in the form of the biologically inactive precursor T4. The principal bioactive form of the hormone is T3. In humans, only about 20% of T3 is secreted by the thyroid; most circulating T3 is derived from outer ring deiodination (ORD) of T4 in peripheral tissues (1, 2, 3). Both T4 and T3 undergo inner ring deiodination (IRD) to metabolites that do not interact with T3 receptors, rT3 and 3,3'-diiodothyronine (3,3'-T2), respectively. Thus, ORD is regarded as an activating pathway, and IRD is regarded as an inactivating pathway. ORD is also the main pathway for the metabolism of rT3, representing another route for the generation of 3,3'-T2. Three iodothyronine deiodinases are involved in the deiodination of iodothyronines, i.e. D1, D2, and D3 (1, 2, 3).

In humans and rodents, D1 is located primarily in liver, kidney, and thyroid (5, 6, 7, 8, 9, 10, 11, 12). Lower D1 activities are expressed in other tissues, including rat anterior pituitary (13). Although D1 has both ORD and IRD activities, it appears to be particularly important for the generation of plasma T3 and clearance of plasma rT3 (1, 3). ORD of rT3 is the most efficient reaction catalyzed by D1, whereas IRD of both T4 and T3 are strongly accelerated by sulfation of these iodothyronines (1). Michaelis-Menten constant (Km) values for substrates of D1 are in the micromolar range. The enzyme is potently inhibited by the thyrostatic drug 6-propyl-2-thiouracil (PTU) (1, 2, 3, 14). D1 activity is positively regulated by T3, reflecting regulation of D1 expression by T3 at the pretranslational level (15). D1 activity in the thyroid and in FRTL-5 rat thy-roid cells is stimulated by TSH and by thyroid-stimulating antibodies (16).

In humans, D2 activity is found in brain, anterior pituitary, placenta, thyroid, and skeletal muscle, and D2 mRNA has also been detected in the human heart (3, 4, 17, 18, 19, 20, 21). In rodents, D2 is also expressed in brown adipose tissue (22, 23). D2 has only ORD activity, preferring T4 over rT3 as the substrate, with apparent Km values in the nanomolar range (17, 20). In general, D2 activity is increased in hypothyroidism and decreased in hyperthyroidism. Both pre- and posttranslational mechanisms are involved in the regulation of D2 expression by thyroid state, with distinct roles for T3, and for T4 and rT3, respectively (24, 25, 26, 27, 28). Although perhaps D2 in skeletal muscle may contribute to circulating T3, the enzyme is particularly important for local T3 production in brain and anterior pituitary (3, 29).

In humans and rodents, D3 is located in brain, placenta, pregnant uterus, and fetal tissues (7, 30, 31, 32, 33). D3 has only IRD activity and is thus important for the inactivation of thyroid hormone. It shows preference for T3 over T4 as the substrate, with apparent Km values in the nanomolar range (33). The high D3 activity in placenta, pregnant uterus, and different fetal tissues seems to serve the purpose of protecting the fetus against undue exposure to active thyroid hormone that may be detrimental for the development of different tissues, in particular the brain (4, 31, 32). In brain, D3 activity is increased in hyperthyroidism and decreased in hypothyroidism, but the mechanism of this regulation remains to be established (34, 35).

Most studies regarding distribution and regulation of deiodinases have been carried out in rats. However, in different respects, the rat does not seem to be the optimal experimental model for human thyroid hormone physiology. This is most obvious for the fetal and neonatal development of the tissues, which follows different patterns relative to the time of birth in humans vs. rats. Little D3 is expressed in rat liver at any stage of development, whereas high D3 activity is detected in fetal human liver or in liver of severely sick patients (7, 31, 36, 37). Furthermore, D2 expression has been observed in human thyroid and skeletal muscle but not in these tissues in the rat (17, 18, 19, 20, 21, 38, 39). Therefore, to investigate thyroid hormone metabolism in different tissues such as thyroid and skeletal muscle, we explored the pig as an alternative animal model. Iodothyronine deiodinase activities (D1, D2, and D3) have been described in liver, kidney, and placenta of pig fetuses as well as in neonatal pigs (40, 41). Our studies involved the investigation of the molecular characteristics, the tissue distribution, and the thyroid state-dependent regulation of the three iodothyronine deiodinases in pigs.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Animals and treatment
Two- to three-month old male and female Yorkshire x Landrace pigs were studied. All experiments were performed in accordance with the Guide for the Care and Use of Laboratory Animals (National Institutes of Health publication no. 86–23, revised 1996) and with prior approval of the Erasmus Medical Center Animal Care Committee. Two pigs (one male and one female) were made hypothyroid by oral treatment with methimazole at a dose of 5 mg/kg·d. Two pigs (one male and one female) were made hyperthyroid by oral treatment with T4 at a dose of 50 µg/kg·d. Both T4 and methimazole were administered in capsules (made by the Erasmus Medical Center Pharmacy with lactose as sweetener), which were mixed with the food. Four untreated pigs were included as controls. Once a week, body weight was measured, and animals were sedated by im injection of 10 mg/kg ketamine and 0.5 mg/kg midazolam; blood was collected for measurement of plasma T4 and T3 levels to ensure adequacy of the methimazole and T4 doses. At the end of the 4-wk period, pigs were sedated with ketamine (20 mg/kg im) and midazolam (0.5 mg/kg im), anesthetized with pentobarbital (20 mg/kg iv), intubated, and ventilated with a mixture of O2 and N2. Fluid-filled catheters (8 French) were inserted into the jugular vein for infusion of pentobarbital (10–15 mg/kg·h iv) and via a femoral artery into the aorta for measurement of arterial blood pressure. A high-fidelity microtipped pressure transducer was advanced via a carotid artery into the left ventricle for measurement of left ventricle pressure and its first derivative LVdP/dt, while a Swan-Ganz catheter was inserted via a femoral vein and advanced into the pulmonary artery for measurement of cardiac output according to the Fick method (42). All measurements were done in duplicate, after which animals underwent a midsternal thoracotomy, and the heart, liver, kidneys, skeletal muscle (Musculus iliopsoas), thyroid, pituitary, and cerebrum were isolated, collected in liquid N2, and stored at –80 C for further analysis.

Materials
Nonradioactive iodothyronines were obtained from Henning Berlin (Berlin, Germany); [3'-125I]T3 and [3',5'-125I]T4 were obtained from Amersham (Little Chalfont, UK); and [3',5'-125I]rT3 was prepared in our laboratory by radioiodination of 3,3'-T2 as described previously (43). [125I]T3 could be used without further purification, but [125I]T4 and [125I]rT3 were purified on Sephadex LH-20 (Pharmacia, Uppsala, Sweden) before each experiment. 125I-labeled and unlabeled T3 sulfate (T3S) and 3,3'-T2 sulfate (3,3'-T2S) were prepared as previously reported (44). Dithiothreitol (DTT), PTU, methimazole, T4 (for treatment of pigs), goldthioglucose (GTG), iodoacetate (IAc), and ß-mercaptoethanol were obtained from Sigma; electrophoresis-grade SDS-PAGE reagents, protein markers, and protein assay reagent were obtained from Bio-Rad (Richmond, IL); Sephadex LH-20 and diethylaminoethyl-dextran were obtained from Pharmacia; TRIzol reagent, synthetic oligonucleotides, rTaq polymerase, cell culture medium, and fetal bovine serum were obtained from GIBCO BRL (Breda, The Netherlands); and oligo(dT), and random hexamer primers, dNTP, RNase inhibitor, M-MLV reverse transcriptase, HindIII, and pGEM-T vector were obtained from Promega (Madison, WI).

Cloning of porcine deiodinases and construction of expression vectors
Total RNA was isolated from different porcine tissues (liver for D1 cloning, pituitary for D2 cloning, and cerebrum for D3 cloning) using TRIzol reagent. cDNA was obtained by reverse transcription of 5 µg total RNA using random hexamer primers as well as oligo(dT) primers and M-MLV reverse transcriptase. Initially, oligonucleotide primers homologous to sequences surrounding the start or stop codons of human, mouse, and rat deiodinases were designed (Table 1Go) and used for PCR reactions with the respective cDNA samples. For D1, the sense and antisense primers contain the start and stop codon, respectively. For D2, the sense primer is located just upstream of the start codon, whereas the antisense primer contains the stop codon. Unfortunately, for D3, it was necessary to locate the sense primer just downstream of the start codon, and the antisense primer was located in the selenocysteine insertion sequence (SECIS) element. The PCR products obtained (~750 bp for D1, ~850 bp for D2, and ~1400 bp for D3) were cloned in the pGEM-T vector and sequenced in both directions.


View this table:
[in this window]
[in a new window]
  		TABLE 1. Oligonucleotide primers used for cloning of porcine deiodinases by RT-PCR

 
Because a SECIS element is required for incorporation of selenocysteine (SeC) in selenoproteins, we prepared chimeric constructs in which the porcine deiodinase coding sequence cDNAs were inserted 5' to the SECIS element of the rat D1 gene. Expression plasmids were constructed by PCR of the pig (p) D1, pD2, and pD3 coding sequence in the pGEM-T plasmids mentioned earlier using primers located over or close to the translation start and stop codons and adapted with HindIII restriction sites (Table 2Go). For the D3 expression vector, a start codon and three amino acids were introduced (MLHS). These PCR products were cloned into the pGEM-T vector, excised with HindIII, and subcloned into the HindIII sites of the rD1SECIS-pcDNA3 plasmid. The latter was obtained by HindIII digestion of the G21-pcDNA3 plasmid (kindly provided by Dr. P. Reed Larsen, Harvard Medical School, Boston, MA) and contains part of the 3' untranslated region of rat D1, including the SECIS element. The inserts of these plasmids (pD1/2/3-rD1SECIS-pcDNA3) were sequenced in both directions.


View this table:
[in this window]
[in a new window]
  		TABLE 2. Oligonucleotide primers used for the construction of deiodinase expression vectors

 
Native deiodinases and expression of recombinant deiodinases
Tissues were homogenized and microsomal fractions were prepared by differential centrifugation, as previously described (8), in 100 mM phosphate buffer (pH 7.2) and 2 mM EDTA containing 1 mM DTT (PED1). Protein concentrations were determined using the Bio-Rad protein assay reagent and BSA as the standard. Aliquots of homogenates and microsomes were snap-frozen on dry ice/ethanol and stored at –80 C until further analysis.

Recombinant deiodinases were expressed in COS-1 cells after diethylaminoethyl-dextran-mediated transfection of expression plasmids as described (46, 47). Two days after transfection, the cells were rinsed with PBS, collected in 0.25 ml of 0.1 M phosphate buffer (pH 7.2), 2 mM EDTA, and 10 mM DTT (PED10), sonicated, aliquoted, snap-frozen on dry ice/ethanol, and stored at –80 C.

Deiodinase enzyme activity measurements
Deiodinase activities of native and recombinant enzyme preparations were analyzed either by quantitation of radioiodide released by ORD of outer ring 125I-labeled rT3 (D1) or T4 (D2) or by analysis of radioactive 3,3'-T2 generated by IRD of outer ring 125I-labeled T3 (D3) by HPLC.

D1 assay.
Appropriate amounts of liver homogenates or microsomal fractions were incubated in triplicate for 60 min at 37 C with 100,000 cpm [3',5'-125I]rT3 and varying amounts of unlabeled iodothyronines in 0.1 ml PE buffer containing 10 mM DTT (PED10). Blank incubations were carried out in the absence of microsomal protein (buffer blank). Reactions were stopped by addition at 4 C of 0.1 ml 5% (wt/vol) BSA in water followed by addition of 0.5 ml 10% (wt/vol) trichloroacetic acid in water. After pelleting of the precipitated [125I]iodothyronines, [125I]iodide was further isolated from the supernatant on LH-20 minicolumns, equilibrated, and eluted with 0.1 M HCl (8). In the case of COS cell homogenates, 50–100 µg protein was incubated in the same manner. Total deiodination was corrected for nonenzymatic deiodination in blank incubations with homogenate from nontransfected COS cells (<5% of total deiodination).

IRD was studied with outer ring-labeled T3S ([3'-125I]T3S). In this assay, IRD activity was the sum of [3'-125I]T2S, [3'-125I]T1S, and 125I– formed. The latter was formed by ORD of [125I]T2S, whereas [125I]T1S was formed by IRD of T2S. Liver microsomal fractions (10–250 µg protein/ml) were incubated for 60 min at 37 C with 10 nM [125I]T3S (200,000 cpm) in 0.1 ml PED10. The reaction was stopped by the addition of methanol (1:1) on ice, and the reaction mixtures were analyzed by reverse-phase HPLC as previously described (48).

D2 assay.
Appropriate amounts of homogenates or microsomal fractions (only homogenate for pituitary) were incubated for 60 min at 37 C with 1 nM (100,000 cpm) [3',5'-125I]T4 in the presence of 100 nM unlabeled T3 to inhibit D3 activity and in the absence or presence of 100 nM unlabeled T4 to saturate D2 in 0.1 ml PE buffer containing 25 mM DTT (PED25). Release of 125I– was determined and corrected for nonenzymatic deiodination as described earlier. The difference in fractional deiodination between incubations with 1 and 100 nM T4 represented low-Km D2 activity. In additional experiments, 1 nM labeled T4 was incubated with microsomal fractions or COS cell homogenates in the presence of varying amounts of unlabeled iodothyronines.

D3 assay.
Appropriate amounts of homogenates or microsomal fractions were incubated in triplicate for 1–4 h at 37 C with 1 nM (200,000 cpm) [125I]T3 in the absence or presence of 100 nM unlabeled T3 to saturate D3 in 0.1 ml PE buffer containing 50 mM DTT (PED50). The reactions were stopped by the addition of 0.1 ml ice-cold methanol. After centrifugation, 0.1 ml of the supernatant was mixed with 0.1 ml 0.02 M ammonium acetate (pH 4), and 0.1 ml of the mixture was applied to a 250 x 4.6 mm Symmetry C18 column connected to an Alliance HPLC system (Waters, Etten-Leur, The Netherlands) and eluted isocratically with a mixture of acetonitrile and 0.02 M ammonium acetate (pH 4; 33:67, vol/vol) at a flow of 1.2 ml/min. Radioactivity in the eluate was monitored online using a Radiomatic A-500 flow scintillation detector (Packard, Meriden, CT). Conversion of labeled T3 to radioactive 3,3'-T2 was corrected for nonenzymatic deiodination as observed in blanks incubated in the absence of microsomal protein (buffer blank). The difference in fractional deiodination between incubations with 1 and 100 nM T3 represented low-Km D3 activity. In additional experiments, 1 nM labeled T3 was incubated with microsomal fractions or COS cell homogenates in the presence of varying amounts of unlabeled iodothyronines.

Western blotting with D1 antiserum
Polyclonal antisera were raised in rabbits against the keyhole limpet hemocyanin conjugate of the synthetic peptide (C)NPEEVRAVLEKLHS (human D1 amino acid residues 236–249). This antiserum cross-reacts with pD1 (48).

Microsomal preparations from porcine liver and kidney (2.5–25 µg microsomal protein and BSA to 25 µg total protein) were separated on 12% SDS-PAGE gels in the Mini-Protean III system (Bio-Rad) according to manufacturer’s instructions. After electrophoresis, the proteins were blotted to nitrocellulose membrane (Hybond ECL; Amersham-Pharmacia Biotech), incubated with primary antiserum (1:500), and subsequently incubated with peroxidase-conjugated secondary antibody as described previously (47).

Affinity labeling of D1 with N-bromoacetyl (BrAc)-[125I]T3
BrAc[125I]T3 (1500 mCi/µmol) was synthesized as described (12, 48), and HPLC analysis demonstrated that the purity was at least 85%, with unreacted [125I]T3 as the main contaminant. Solutions of BrAc[125I]T3 (100,000 cpm, 0.03 pmol) in ethanol were pipetted into microcentrifuge tubes, and the solvent was evaporated by a stream of nitrogen. After the addition of 50 µl PED10 and vortexing, the liver or kidney microsomal fractions (100 µg protein) were added in a total volume of 25 µl PED10. COS cell homogenates with pD1 or rat D1 protein were used as controls. The mixtures were incubated for 15 min at 37 C. Reactions were terminated by the addition of SDS-PAGE gel-loading buffer, and samples were analyzed by SDS-PAGE (12% gel), followed by autoradiography to Kodak BioMax MS film (Eastman Kodak, Rochester, NY).

Hormone measurements
Plasma T4 and T3 levels were determined by RIA (49). Radiolabeled iodothyronines were obtained from Amersham-Pharmacia Biotech, and T4 and T3 antisera were produced previously (49). The sample volume was 50 µl for T4 and 25 µl for T3. The incubation volume was 0.5 ml RIA buffer (0.06 M barbital, 0.15 M NaCl, 0.1% BSA, and 0.6 g/liter 8-anilino-1-naphthalenesulfonic acid, pH 8.6). Mixtures were incubated in duplicate overnight at 4 C, and antibody-bound radioactivity was precipitated using Sac-Cel cellulose-coupled second antibody (IDS, Boldon, UK). The lower limit of detection was 2 nmol/liter for T4 and 0.08 nmol/liter for T3, and all plasma samples were measured in the same assay. The free fractions of plasma T4 and T3 were determined by equilibrium dialysis (50) and multiplied with the total T4 and T3 levels for calculation of the free T4 and free T3 concentrations.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
Cloning of porcine iodothyronine deiodinases
The coding sequences of the porcine deiodinases were cloned by RT-PCR on total RNA isolated from tissues with particularly high expression of these enzymes using primers derived from the nucleotide sequences of the human deiodinases. pD1 was cloned from liver, pD2 was cloned from pituitary, and pD3 was cloned from cerebrum. Initial attempts were carried out using forward sense primers representing the 5' flanking region and reverse antisense primers representing the 3' flanking region of the human deiodinase coding sequences (Table 1Go). This approach produced cDNA clones of the coding sequences of pD1 and pD2. For pD1, the primers used overlapped the start and stop codon, whereas for pD2, the forward sense primer was located 20 bp upstream of the start codon. Unfortunately, for the cloning of pD3, this approach was not successful. The forward sense primer had to be chosen downstream of the human D3 start codon, whereas the reverse antisense primer was located in the SECIS element. The deduced amino acid sequences are presented in Fig. 1Go, aligned with corresponding sequences of human and rat deiodinases. The amino acid sequence of pD1 shows 85% identity with human D1, 78% identity with rat D1, and 77% identity with mouse D1. The amino acid sequence of pD2 has 92% identity with human D2, 90% identity with rat D2, and 90% identity with mouse D2. Finally, the amino acid sequence of pD3 has 94% identity with human D3, 91% identity with rat D3, and 89% identity with mouse D3. Deiodinases contain a SeC residue in the catalytic center, which is essential for catalytic activity (2, 3, 46, 47). The core catalytic center of about 15 amino acid residues around the SeC residue is completely conserved in the porcine deiodinases, including the typical Cys to Ala substitution two residues N-terminal of the SeC residue in D2 enzymes (46).



View larger version (124K):
[in this window]
[in a new window]
  		FIG. 1. Comparison of human, porcine, and rodent deiodinases. Alignment of the deduced amino acid sequences of human (h), porcine (p), and rat (r) D1, D2, and D3 iodothyronine deiodinases. The SeC residue in the catalytic center is indicated by U.

 
Expression vectors were made as described in Materials and Methods using the oligonucleotide primers in Table 2Go. The expression vectors contained the rat D1 SECIS element downstream of the coding sequences. In the pD3 expression vector, the start codon and three amino acid residues (MLHS) were introduced (Table 2Go and Fig. 1Go).

Catalytic characterization of native and recombinant porcine deiodinases
Deiodinase activities were studied under initial reaction rate conditions, with conversion rates being linear with protein concentration and incubation time. Pig liver microsomes were used as a source of native pD1, and its activity was compared with that of recombinant pD1 expressed in COS-1 cells (Fig. 2Go). Native and recombinant pD1 showed ORD activity with T4 and rT3 and IRD activity with T3S (Fig. 2AGo); much higher rates for ORD of rT3 than of T4 (Fig. 2AGo); much higher rates for IRD of T3S than of nonsulfated T3 (Fig. 2AGo); equal dose-dependent inhibition of the ORD of [125I]rT3 by unlabeled iodothyronines with approximate IC50 values of 0.2 µM rT3, 2 µM T4, and 10 µM T3 (Fig. 2BGo); similar apparent Km values for rT3 ({approx}0.2 µM) as determined by Lineweaver-Burk analysis (Fig. 2CGo; Table 3Go); and identical sensitivity to well-known D1 inhibitors with approximate IC50 values of 0.02 µM GTG, 1 µM IAc, and 10 µM PTU (Fig. 2DGo). The potency of PTU inhibition increased 10-fold if the rT3 substrate concentration was increased from 10 to 100 nM (data not shown), which is in agreement with the uncompetitive nature of PTU inhibition (2, 14).



View larger version (35K):
[in this window]
[in a new window]
  		FIG. 2. Characterization of pD1 enzyme. A, ORD of T4, T3, and rT3 as well as IRD of T3S by recombinant (Rec) D1 expressed in COS cells (open bars) and native (Nat) D1 in porcine liver microsomes (closed bars). Assay mixtures contained 10 nM 125I substrate (100,000 cpm), 10 mM DTT, and 0.3 (lysate) or 0.05 (liver microsomes) mg protein/ml and were incubated for 60 min at 37 C. B, Inhibition of the ORD of [125I]rT3 by recombinant D1 (open symbols) or native D1 enzyme (closed symbols) by 0.1–100 µM unlabeled rT3, T4, or T3. Assay mixtures contained 10 nM [125I]rT3 (100,000 cpm), 10 mM DTT, and 0.1 (lysate) or 0.01 (liver microsomes) mg protein/ml and were incubated for 60 min at 37 C. C, Double reciprocal plot of the rate of rT3 deiodination catalyzed by recombinant (Rec) D1 enzyme (open symbols) and native (Nat) D1 enzyme (closed symbols). D, Inhibition of the ORD of [125I]rT3 by recombinant D1 (open symbols) and native D1 (closed symbols) by increasing concentrations of GTG (0.001–10 µM), IAc (0.01–100 µM), or PTU (0.1–1000 µM). Assay mixtures contained 10 nM [125I]rT3 (100,000 cpm), 10 mM DTT, and 0.1 (lysate) or 0.01 (liver microsomes) mg protein/ml and were incubated for 60 min at 37 C. ctrl, Control.

 

View this table:
[in this window]
[in a new window]
  		TABLE 3. Apparent Km values for preferred substrates of porcine iodothyronine deiodinases

 
These findings demonstrate that native and recombinant pD1 have identical catalytic properties, substrate specificities, and inhibitor sensitivities and that these characteristics are in good agreement with previous studies of human and rat D1 (2, 8, 11). However, when the deiodination of [3'-125I]T3S by porcine liver microsomes was studied in comparison with rat and human liver microsomes, a remarkable difference in product formation was observed (Fig. 3Go). T3S was very rapidly deiodinated by all three D1 enzymes through IRD to 3,3'-T2S as the initial product (IRD of outer ring-labeled [3'-125I]T3S gives [3'-125I]T2S). Although this intermediate was completely converted by rat D1 via ORD (only 125I– produced), some IRD to [3'-125I]T1S by human D1 was observed, whereas 3,3'-T2S was converted by pD1 to almost equal extents via ORD and IRD (note that 125I– is produced by ORD of 3,3'-T2S and 3'-T1S). These differences in deiodination of 3,3'-T2S by pD1, human D1, and rat D1 via ORD or IRD were also documented using [3'-125I]T2S as the substrate (data not shown).



View larger version (30K):
[in this window]
[in a new window]
  		FIG. 3. Deiodination (IRD and ORD) of sulfated iodothyronines by human D1, pD1, and rat D1 enzyme. Liver microsomal fractions (10–250 µg protein/ml) were incubated for 60 min at 37 C with 10 nM [3'-125I]T3S (200,000 cpm) in 0.1 ml PED10 buffer. The reaction was stopped by the addition of methanol, and the mixture was analyzed by reverse-phase HPLC as described (48 ). In this assay, IRD activity was the sum of [125I]T2S, [125I]T1S, and 125I– formed. The latter was formed by ORD of [125I]T2S, whereas [125I]T1S was formed by IRD of T2S. Whereas pD1 catalyzes both ORD and IRD of T2S, human and rat D1 cause only ORD of T2S.

 
For comparison with the catalytic activity of recombinant pD2, pig thyroid homogenate (from methimazole-treated pigs) was used as a source of native pD2 (Fig. 4Go). Native and recombinant pD2 demonstrated ORD activity with T4 and rT3 but no IRD activity with these iodothyronines or T3 as substrates (Fig. 4AGo); higher rates for ORD of T4 than of rT3 (Fig. 4AGo); identical dose-dependent inhibition of the ORD of [125I]T4 by unlabeled iodothyronines, with approximate IC50 values of 10 nM T4, 100 nM rT3, and much greater than 1000 nM T3 (Fig. 4BGo); identical apparent Km values for T4 ({approx}8 nM; Fig. 4CGo and Table 3Go); and similar effects of inhibitors, with approximate IC50 values of 1 µM GTG, greater than 1000 µM IAc, and greater than 1000 µM PTU (Fig. 4DGo). These results demonstrate that native and recombinant pD2 are identical with regard to catalytic activity, substrate specificity, and inhibition by GTG and that these properties are very similar to those previously reported for human and rat D2 (17, 20).



View larger version (33K):
[in this window]
[in a new window]
  		FIG. 4. Characterization of pD2 enzyme. A, ORD of T4, T3, or rT3 by recombinant (Rec) pD2 (open bars) from transfected COS cell homogenates and native (Nat) D2 (closed bars) in thyroid homogenate of methimazole-treated pig. Assay mixtures contained 1 nM 125I substrate (100,000 cpm), 25 mM DTT, and 0.2 (lysate) or 0.4 (thyroid homogenate) mg protein/ml and were incubated for 60 min at 37 C. B, Inhibition of the ORD of [125I]T4 by recombinant (open symbols) or native (closed symbols) pD2 enzyme by unlabeled T4, rT3, and T3. Assay mixtures contained [125I]T4 (100,000 cpm), 25 mM DTT, and 0.12 (lysate) or 0.21 (thyroid homogenate) mg protein/ml and were incubated for 60 min at 37 C. C, Double reciprocal plot of T4 deiodination by recombinant (Rec) D2 (open symbols) and native (Nat) D2 enzyme (closed symbols). D, Inhibition of the ORD of [125I]T4 by recombinant (open symbols) or native D2 (closed symbols) by GTG (0.1–100 µM), PTU (10–1000 µM), or IAc (10–1000 µM). Assay mixtures contained 1 nM [125I]T4 (100,000 cpm), 25 mM DTT, and 0.27 (lysate) or 0.2 (thyroid homogenate) mg protein/ml. ctrl, Control.

 
To compare the catalytic activities of native and recombinant pD3, pig cerebrum microsomal preparations were used as a source of native enzyme (Fig. 5Go). Native and recombinant pD3 showed IRD activity with T4 and T3 but no ORD activity with these iodothyronines or rT3 as substrates (Fig. 5AGo); higher rates for IRD of T3 than of T4 (Fig. 5AGo); similar dose-dependent inhibition of the IRD of [125I]T3 by unlabeled iodothyronines, with approximate IC50 values of 20 nM T3, 50 nM T4, and much greater than 1000 nM rT3 (Fig. 5BGo); identical apparent Km values for T3 ({approx}20 nM; Fig. 5CGo and Table 3Go); and similar sensitivity to inhibitors, with approximate IC50 values of 1 µM GTG, greater than 1000 µM IAc, and greater than 1000 µM PTU (Fig. 5DGo). These results show that native and recombinant pD3 have identical properties regarding catalytic activity, substrate specificity, and inhibitor sensitivity, which are in good agreement with those previously reported for human and rat D3 (33).



View larger version (33K):
[in this window]
[in a new window]
  		FIG. 5. Characterization of pD3 enzyme. A, IRD of T4 and T3 or ORD of rT3 by recombinant (Rec) pD3 enzyme (open bars; transfected COS cell homogenates) and native (Nat) D3 enzyme (closed bars; pig cerebrum microsomal fraction). Assay mixtures contained 10 nM 125I substrate (200,000 cpm), 50 mM DTT, and 0.03 (lysate) or 2 (cerebrum microsomes) mg protein/ml and were incubated for 60 min at 37 C. B, Inhibition of the IRD of [125I]T3 by recombinant D3 (open symbols) or native D3 (closed symbols) by increasing concentrations of unlabeled T3, T4, or rT3 (1–1000 nM). Conditions were 10 nM [125I]T3 (100,000 cpm), 50 mM DTT, and 0.02 (lysate) or 1 (cerebrum) mg protein/ml. C, Double reciprocal plot of T3 deiodination by native (Nat) D3 enzyme (closed symbols) and recombinant (Rec) D3 enzyme (open symbols). D, Inhibition of the IRD of [125I]T3 by recombinant (open symbols) or native D3 (closed symbols) by GTG (0.1–100 µM), PTU (10–1000 µM), or IAc (10–1000 µM). Assay mixtures contained 10 nM [125I]T3, 50 mM DTT, and 1 (cerebrum microsomes) or 0.02 (lysate) mg protein/ml. ctrl, Control.

 
Tissue distribution and thyroid state-dependent regulation of pig deiodinases
To study the regulation of pD1, pD2, and pD3 in several tissues by thyroid state, two pigs were made hypothyroid by methimazole treatment, two pigs were made hyperthyroid by high-dose T4 treatment, and four untreated pigs were studied as euthyroid controls. Treatment with methimazole resulted in a decrease in serum T4 and T3 levels from about 35 and 0.8 nmol/liter, respectively, at baseline to below the detection limit within 2–3 wk in both pigs (Fig. 6Go). Four weeks after onset of treatment, animals showed a 25% decrease in heart rate, a 15% decrease in cardiac output, and a 40% decrease in left ventricle pressure first derivative compared with euthyroid pigs, whereas body weight gain was normal (Table 4Go). The thyroid weight, corrected for body weight, doubled in 4 wk (Table 4Go). The porcine TSHß subunit is highly homologous (80%) with the human TSHß subunit (51). Nevertheless, due to poor cross-reactivity of porcine TSH in our human TSH assay (Immulite 2000; DPC, Los Angeles, CA), we could not detect TSH in serum from euthyroid animals; we could only detect TSH in serum from hypothyroid pigs (data not shown). Conversely, treatment with L-thyroxine caused a 4-fold increase of serum T4 levels after 2–3 wk of treatment (Fig. 6Go). Four weeks after the onset of treatment, animals showed slight increases in heart rate (15%) and cardiac output (20%) compared with euthyroid pigs, despite a 10% lower body weight (Table 4Go).



View larger version (26K):
[in this window]
[in a new window]
  		FIG. 6. Plasma T4 and T3 levels (total and free). Pigs were treated with methimazole (closed symbols, n = 2) or T4 (open symbols, n = 2) during 4 wk as described in Materials and Methods. Once every week, blood samples were collected and total T4 levels were measured to ensure adequacy of the treatment. At the end of the 4-wk treatment period, all plasma samples were measured in the same assay for total and free T4 and T3 (see Materials and Methods). The total T4 and T3 plasma levels of the untreated pigs were similar to the levels at wk 0 of the treated pigs and did not change over the 4-wk period (data not shown).

 

View this table:
[in this window]
[in a new window]
  		TABLE 4. Hemodynamic parameters and body/tissue weight of experimental animals

 
Tissue deiodinase activities were assayed in homogenates and microsomal fractions, except for pituitary in which analyses were done only in homogenates. Only deiodinase activities determined in homogenates are presented here (Fig. 7Go). Activities expressed per milligram of protein were 2- to 5-fold higher in microsomes than in homogenates, but similar thyroid state-dependent changes were observed in both fractions. In euthyroid pigs, the highest D1 activities were observed in liver. Hepatic D1 activities were 20% lower in methimazole-treated pigs than in control animals, but they were markedly (>3-fold) increased in livers from hyperthyroid pigs (Fig. 7AGo). Renal D1 activity was not only markedly (>3-fold) increased in the hyperthyroid animals but also (2-fold) in the methimazole-treated pigs (Fig. 7AGo). Thyroid and skeletal muscle showed low D1 activity (<0.1 pmol/min·mg protein), which was not regulated by thyroid state (data not shown). No significant D1 activity was detected in brain and pituitary homogenates.



View larger version (18K):
[in this window]
[in a new window]
  		FIG. 7. Thyroid state-dependent regulation of deiodinase activity in various tissues. A, D1 activities (mean and range shown) in liver and kidney homogenates of methimazole-treated pigs (Hypo/closed bars, n = 2), thyroxine-treated pigs (Hyper/hatched bars, n = 2), or untreated pigs (Eu/open bars, n = 4). Assay mixtures contained 0.1 µM [125I]rT3 (100,000 cpm), 10 mM DTT, and 0.1 mg protein/ml and were incubated for 30 min at 37 C. B, D2 activities (mean and range shown) in cerebrum (brain), pituitary (Pit), thyroid, and skeletal muscle (Musculus iliopsoas) homogenates of methimazole-treated pigs (closed bars, n = 2), thyroxine-treated pigs (hatched bars, n = 2), and untreated pigs (open bars, n = 4). Note that the D2 activities in cerebrum and muscle homogenates are multiplied by a factor 10. Assay mixtures contained 1 nM [125I]T4 (100,000 cpm), 25 mM DTT, and 1 mg protein/ml and were incubated for 60 min at 37 C. C, D3 activity (mean and range shown) in cerebrum (brain) homogenates of methimazole-treated pigs (closed bars, n = 2), thyroxine-treated pigs (hatched bars, n = 2), or untreated pigs (open bars, n = 4). Assay mixtures contained 1 nM [125I]T3 (200,000 cpm), 50 mM DTT, and 1 mg protein/ml and were incubated for 60 min at 37 C.

 
We have recently developed an anti-D1 antiserum that cross-reacts with cat and pD1 protein (48). Microsomal fractions from liver, kidney, and thyroid were analyzed by immunoblotting with this antiserum. In microsomal fractions prepared from livers of hyperthyroid pigs, the amount of D1 protein was similar or slightly increased compared with microsomal fractions of euthyroid livers (Fig. 8Go). In microsomal fractions of liver from hypothyroid animals, the amount of D1 protein was strongly reduced. In microsomal fractions prepared from kidney (Fig. 8Go), essentially the same results were obtained (i.e. strong decrease of D1 protein content in preparations from hypothyroid pigs compared with euthyroid and hyperthyroid pigs). In microsomal fractions of thyroid tissue, no D1 protein could be detected on Western blots, which was in line with the very low D1 activities.



View larger version (37K):
[in this window]
[in a new window]
  		FIG. 8. Western blot analysis of D1 protein (apparent molecular mass, 27 kDa) in porcine liver and kidney microsomal fractions. Increasing amounts of microsomal fractions (2.5–25 µg protein, supplemented with BSA to 25 µg) were probed with an anti-D1 antiserum generated against a 14-amino acid peptide corresponding to the C-terminal sequence of human D1 (see Materials and Methods). Pigs 1 and 2 were methimazole treated; pigs 3 and 4 were thyroxine treated; and pigs 5 and 6 were untreated. D1 activity values (pmol rT3/min·mg microsomal protein) are indicated. Assay conditions were 0.1 µM [125I]rT3, 10 mM DTT, and 0.01 mg microsomal protein/ml. Migration distances of molecular mass markers (kilodaltons) are indicated on the left. eu, Euthyroid; hyper, hyperthyroid; hypo, hypothyroid.

 
BrAc[125I]T3 has been used extensively as affinity label for D1, allowing the specific identification and quantitation of D1 in liver microsomal fractions from many species (12). Affinity labeling was used here for the semiquantitative analysis of D1 protein in liver and kidney microsomal fractions (Fig. 9Go). Three prominent radioactively labeled protein bands were visible with apparent molecular masses of 27 kDa, 32 kDa, and 56 kDa. The 56-kDa protein was previously identified as protein disulfide isomerase (12). The 32-kDa protein band is not identical to D1 protein because its labeling could not be prevented by rT3 and PTU (12). The 27-kDa protein band represents D1 protein because affinity labeling could be prevented by rT3 and PTU incubation (12), and it comigrates with the 27-kDa D1 protein present in homogenates from COS cells transfected with the pD1 expression vector (Fig. 9Go). The labeling intensities of the 27-kDa D1 protein strongly correlate with the D1 activities measured in liver and kidney microsomal fractions from euthyroid, hypothyroid, and hyperthyroid pigs. This is especially apparent for kidney microsomal fractions in which the D1 labeling intensities in both the hypothyroid and hyperthyroid state are stronger than in the euthyroid state, which is in line with the D1 activity levels (Figs. 7AGo and 9Go).



View larger version (50K):
[in this window]
[in a new window]
  		FIG. 9. Labeling patterns obtained by SDS-PAGE and autoradiography after reaction of liver and kidney microsomal fractions (100 µg protein) with BrAc[125I]T3 in the presence of 10 mM DTT at 37 C. Pigs 1 and 2 (pig numbers indicated at bottom) were methimazole treated; pigs 3 and 4 were thyroxine treated; and pigs 5 and 6 were untreated. D1 activity values (pmol rT3/min·mg microsomal protein) are indicated inside the autoradiogram. Migration distances of molecular mass markers (kilodaltons) are indicated on the left. eu, Euthyroid; hyper, hyperthyroid; hypo, hypothyroid.

 
In euthyroid pigs, D2 activity was higher in pituitary than in brain (cerebrum), whereas D2 activity was low to undetectable in thyroid and skeletal muscle (Fig. 7BGo). D2 activities in brain and pituitary were increased more than 10-fold in hypothyroid pigs and decreased by 40% in hyperthyroid pigs compared with euthyroid controls. A dramatic induction of D2 activity was observed in thyroid of hypothyroid pigs to levels similar to those in hypothyroid pituitary. Furthermore, thyroid weight increased markedly after methimazole treatment compared with control and T4-treated pigs (Table 4Go). A striking induction of D2 expression was also seen in skeletal muscle (Musculus iliopsoas) to levels approaching those in hypothyroid brain (Fig. 7BGo). D2 activities were undetectable in liver and kidney even in hypothyroid animals.

Irrespective of thyroid state, significant D3 activities were only detected in brain (cerebrum). Cerebral D3 activity was increased by 40% in hyperthyroid pigs and decreased by more than 50% in hypothyroid pigs vs. euthyroid controls (Fig. 7CGo).

Porcine heart homogenates from euthyroid control pigs, hypothyroid pigs, and hyperthyroid pigs were analyzed for D2 as well as D3 activity. Hearts were subdivided into left and right ventricle (mainly ventricle wall), left atrium, right atrium, and atrium septum. Rather low D2 activities (<0.1 fmol/min·mg protein) and D3 activities (<0.1 fmol/min·mg protein) could be detected in these samples, irrespective of thyroid state (data not shown).


   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
Our studies demonstrate a high degree of homology between the amino acid sequences of the porcine iodothyronine deiodinases and the corresponding human enzymes. Also, the catalytic properties of the pig deiodinases are virtually identical to those previously reported for the human and rat enzymes. Thus, pD1 possesses both ORD and IRD activities, has apparent Km values in the micromolar range, accepts rT3 as the preferred (ORD) substrate, and shows markedly facilitated IRD of sulfated vs. nonsulfated T3. Sulfation has also been reported to facilitate the IRD of T4 by human and rat D1 (1, 2, 3, 8), but this was not studied here with pD1. The only difference noted between pD1, human D1, and rat D1 was the higher susceptibility of 3,3'-T2S to undergo IRD by pD1 than by human D1 and, in particular, rat D1. Much like the enzyme in humans and rats, pD1 is extremely sensitive to inhibition by GTG, IAc, and PTU.

The catalytic properties of pD2 and pD3 are also identical to those reported for the corresponding human and rat deiodinases. Thus, pD2 only shows ORD activity, prefers T4 over rT3 as the substrate, and shows apparent Km values in the nanomolar range. pD3 has only IRD activity, prefers T3 over T4 as the substrate, and shows apparent Km values intermediate between those of pD1 and pD2. D2 and D3 from humans and rats do not catalyze the deiodination of sulfated substrates, but this was not studied here with pD2 and pD3. Much like the human and rat enzymes, pD2 and pD3 are only inhibited by GTG at 100-fold higher concentrations than those required for pD1 inhibition, whereas pD2 and pD3 are hardly affected by IAc and PTU. GTG and IAc are thought to inhibit D1 by reacting with the selenolate form of the catalytic SeC residue, whereas PTU is thought to react with the selenenyl iodide intermediate generated during catalysis. The much lower potencies of the effects of these inhibitors on D2 and D3 suggest that these enzymes follow a catalytic mechanism different from that of D1, although they also contain a SeC residue in their active centers (46, 47).

To further investigate the suitability of the pig as an animal model for human thyroid hormone metabolism, we studied the tissue distribution of the different deiodinases as well as their regulation by thyroid state. Treatment with methimazole successfully induced hypothyroidism, as indicated by the reduction of both plasma T4 and T3 (both total and free) to undetectable levels, the large increase in thyroid weight, and the marked decrease in cardiac output. Conversely, treatment with T4 resulted in large increases in serum total and free T4, decreases in thyroid and body weight, and increases in heart rate and cardiac output, attesting to the hyperthyroid state of the animals. This was supported by the significant increases in D1 activities in liver and kidney of the T4-treated pigs. Previous studies have shown increased D1 activity in liver homogenates of hyperthyroid rats (52). In addition, a close correlation between rodent hepatic and renal D1 mRNA levels and enzyme activity was found (53, 54, 55), suggesting predominant regulation at the pretranslational level. In studies with thyroid hormone receptor (TR)-deficient mice, it was established that TRß is mainly responsible for D1 regulation in liver, whereas in kidney, regulation relied solely on TRß (56). Identification of T3 receptor-binding response elements (TREs) in the human D1 gene promoter (15, 57) further indicates that the thyroid hormone regulation of D1 is exerted at the level of gene transcription. Whether similar TREs are present in the pD1 promoter remains to be investigated.

Remarkably, D1 activity showed only a small decrease in liver and actually a marked increase in kidney of hypothyroid pigs. The increased D1 activity in kidney of hypothyroid pigs compared with euthyroid pigs was reflected in increased labeling intensity upon affinity labeling with BrAcT3. In general, the BrAcT3 affinity-labeling data correlates strongly with the D1 activity measurements. Meanwhile, the immunoblotting experiments revealed decreased D1 protein content in livers and kidneys of hypothyroid animals, whereas the D1 protein content of tissue from euthyroid and hyperthyroid pigs was very similar. There appears to exist a discrepancy between the D1 protein content as determined by immunoblotting and the D1 activity measurements. This is the case for the hypothyroid pigs, in which activities are similar (liver) or increased (kidney), while the D1 protein content is decreased (liver and kidney), compared with euthyroid animals. Also, for hyperthyroid pigs, the activity is clearly increased in liver and kidney compared with euthyroid pigs, whereas the protein content is similar for euthyroid and hyperthyroid pigs. We have recently observed alternative splicing of D1 mRNA in human liver (58). Nine D1 mRNA variants have been identified so far, all encoding truncated proteins from which two still contain the catalytic center (GenBank accession nos. AY560374–AY560383). The epitope of our D1 antiserum is at the C terminus, and it would not detect all of these D1 variants, provided that these also exist at the protein level. So, it could be the case that, in porcine liver and kidney, variant D1 proteins exist that contribute to D1 activity, which are not detected by immunoblotting. Efforts to raise polyclonal D1 antisera directed against different epitope(s) have not been successful so far.

To our knowledge, this is the first study that compares liver and kidney D1 protein content determined by immunoblotting and activity measurements in euthyroid, hypothyroid, and hyperthyroid animals. Only one previous study by DePalo et al. (59) showed increased abundance of the 27-kDa D1 protein band on Western blots of liver microsomes from hyperthyroid rats compared with samples from hypothyroid rats, which is in line with our results. In that study, no immunoblot data and no D1 activity measurements for euthyroid rats were presented. In a human hepatoma cell line (HepG2), it was found that T3 treatment influences the alternative splicing of TR{alpha} mRNA, thus changing the balance toward TR{alpha}2 encoding mRNA (60). Although regulation of D1 expression by T3 is generally considered to occur mainly at the transcriptional level (15, 53, 54, 55, 56, 57), effects of T3 on D1 splicing cannot be excluded and remain to be investigated.

Finally, we cannot exclude that the lack of a large reduction in hepatic D1 activity and the increase in renal D1 activity in methimazole-treated rats could represent direct effects of methimazole rather than the hypothyroid state it induces. This has been demonstrated for the marked induction of phenol uridine-5'-diphospho-glucuronyltransferase activity in livers of methimazole-treated rats, which was not prevented by administration of T4 replacement doses (61).

The regulation of D2 expression by thyroid state has been extensively studied in rat brain and pituitary (17, 24, 25, 26, 62, 63, 64). Our findings regarding the effects of hypo- and hyperthyroidism on D2 activity in pig brain and pituitary are in excellent agreement with these previous studies in rats. Thus, brain and pituitary D2 activities were strongly increased in hypothyroid pigs and markedly decreased in hyperthyroid animals compared with the euthyroid controls. As demonstrated in rats, the negative control of D2 expression in brain and pituitary by thyroid hormone involves two different mechanisms (62). First, down-regulation of D2 mRNA expression by thyroid hormone is probably mediated by the nuclear T3 receptor, although a putative negative TRE in the promoter region of the D2 gene remains to be identified (39, 65). Second and more importantly, D2 also undergoes substrate-induced enzyme inactivation by selective proteolysis, which is exerted by the substrates T4 and rT3 rather than by T3 (26, 28, 64, 67).

One of the remarkable findings in our study was the expression of high D2 activity in thyroid of hypothyroid pigs, whereas little thyroidal D2 activity was found in euthyroid animals. Normal human thyroid tissue expresses D2, thus contributing to the plasma T3 pool (18, 21). In patients with Graves’ disease or hyperfunctioning adenomas, D2 activities increase significantly, sometimes causing relatively high circulating free T3 levels (18, 21, 68). Thyroid function is normally undisturbed in patients with thyroid carcinoma, but in some patients with large or widely metastatic follicular carcinoma, increased T4 to T3 conversion due to overexpression of D2 was found (69). In vitro studies using human thyroid cells have shown that D2 expression is up-regulated by TSH through the cAMP-protein kinase A pathway (18, 21, 38). The effect of cAMP is exerted at the pretranslational level, probably by stimulation of D2 gene transcription as suggested by the identification of a cAMP response element in the promoter region of the human D2 gene (66, 70). The expression of D2 in human but not in rat thyroid has been associated with the presence of a thyroid transcription factor 1 response element in the D2 gene promoter in humans but not in rats (39). It is very likely that such cAMP and thyroid transcription factor 1 response elements are also present in the promoter of the pD2 gene and that the increased D2 activity in hypothyroid pigs is the consequence of elevated TSH levels acting through the cAMP-protein kinase A pathway.

Perhaps the most striking finding in our study was the expression of high D2 activity in skeletal muscle of hypothyroid pigs in contrast to the insignificant D2 activity in skeletal muscle of euthyroid animals. Expression of D2 mRNA and activity in human skeletal muscle has been previously reported (17, 20). Also, cultured human skeletal muscle cells express D2 mRNA and activity in particular in the absence of thyroid hormone and after ß-adrenergic stimulation of the cAMP-mediated pathway (19). D2 expression in skeletal muscle may play a role in local T3 production in particular in hypothyroid subjects. The increased D2 activity in skeletal muscle of hypothyroid pigs might in part be explained by a reduction of T4-induced D2 proteolysis (67).

Significant expression of D3 was only observed in brain (cerebrum) but not in other tissues. The positive regulation of brain D3 expression by thyroid state is in agreement with previous studies of the regulation of D3 in rat brain (34, 35). The mechanism of this thyroid hormone-dependent regulation of D3 expression remains to be established.

We detected rather low D2 and D3 enzyme activities in porcine heart samples, which were not regulated by thyroid state. Several studies have reported D2 and D3 activity in heart homogenates, albeit at low levels. In rat heart, D3 activity was induced during hypertrophy and cardiac failure (71). Prolonged treatment with methimazole increased D2 activity in mouse and rat heart (72). In several studies, D2 mRNA was detected in rodent or human heart, either by RT-PCR (72, 73) or by Northern blotting (20). At the moment, no studies are available that describe D2 or D3 activities in human heart samples. It is obvious that more investigations are needed, in particular with regard to possible changes in deiodinase activities in human or porcine heart during ventricular hypertrophy and cardiac failure and after myocardial infarction.

In conclusion, the pig appears to be a good animal model for human thyroid hormone metabolism, considering the high degree of homology between the structures, functional properties, tissue distribution, and thyroid state-dependent regulation of the porcine and human iodothyronine deiodinases. The expression of D2 in porcine thyroid and skeletal muscle is of particular interest for studies on the importance of the enzyme in these tissues of (hypothyroid) human subjects.


   Acknowledgments
 
We thank Hans van Toor for synthesis of radiolabeled iodothyronines and hormone measurements and Ronald van der Wal for plasmid DNA sequencing.


   Footnotes
 
This work was supported by The Netherlands Organization of Scientific Research (Grants 903-40-194 and 920-03-151) and the Quality of Life Research Program of the European Union (Grant QLG3-CT-2000-00930).

This work was presented in part at the 73rd Annual Meeting of the American Thyroid Association, Washington, DC, 2001 (Abstract 38).

The porcine iodothyronine deiodinase coding sequences have been submitted to the GenBank database under accession nos. AY533206 (type I iodothyronine deiodinase), AY533207 (type II iodothyronine deiodinase), and AY533208 (type III iodothyronine deiodinase).

Abbreviations: BrAc, Bromoacetyl; DTT, dithiothreitol; GTG, goldthioglucose; IAc, iodoacetate; IRD, inner ring deiodination; ORD, outer ring deiodination; p, pig; PTU, 6-propyl-2-thiouracil; SeC, selenocysteine; SECIS, selenocysteine insertion sequence; 3'-T1,3'-monoiodothyronine; 3,3'-T2, 3,3'-diiodothyronine; 3,3'-T2S, 3,3'-T2 sulfate; T3S, T3 sulfate; TR, thyroid hormone receptor; TRE, T3 receptor-binding response element.

Received March 18, 2004.

Accepted for publication May 26, 2004.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Visser TJ 1996 Pathways of thyroid hormone metabolism. Acta Med Austriaca 23:10–16[Medline]
  2. Kohrle J 2002 Iodothyronine deiodinases. Methods Enzymol 347:125–167[Medline]
  3. Bianco AC, Salvatore D, Gereben B, Berry MJ, Larsen PR 2002 Biochemistry, cellular and molecular biology, and physiological roles of the iodothyronine selenodeiodinases. Endocr Rev 23:38–89[Abstract/Free Full Text]
  4. Bernal J 2002 Action of thyroid hormone in brain. J Endocrinol Invest 25:268–288[Medline]
  5. Schreck R, Schnieders F, Schmutzler C, Kohrle J 1994 Retinoids stimulate type I iodothyronine 5'-deiodinase activity in human follicular thyroid carcinoma cell lines. J Clin Endocrinol Metab 79:791–798[Abstract]
  6. Chanoine JP, Braverman LE, Farwell AP, Safran M, Alex S, Dubord S, Leonard JL 1993 The thyroid gland is a major source of circulating T3 in the rat. J Clin Invest 91:2709–2713
  7. Richard K, Hume R, Kaptein E, Sanders JP, van Toor H, de Herder WW, den Hollander JC, Krenning E, Visser TJ 1998 Ontogeny of iodothyronine deiodinases in human liver. J Clin Endocrinol Metab 83:2868–2874[Abstract/Free Full Text]
  8. Visser TJ, Kaptein E, Terpstra OT, Krenning EP 1988 Deiodination of thyroid hormone by human liver. J Clin Endocrinol Metab 67:17–24[Abstract]
  9. Brtko J, Bobalova J, Podoba J, Schmutzler C, Kohrle J 2002 Thyroid hormone receptors and type I iodothyronine 5'-deiodinase activity of human thyroid toxic adenomas and benign cold nodules. Exp Clin Endocrinol Diabetes 110:166–170[CrossRef][Medline]
  10. Beech SG, Walker SW, Dorrance AM, Arthur JR, Nicol F, Lee D, Beckett GJ 1993 The role of thyroidal type I iodothyronine deiodinase in triiodothyronine production by human and sheep thyrocytes in primary culture. J Endocrinol 136:361–370[Abstract]
  11. Borges M, Ingbar SH, Silva JE 1990 Iodothyronine deiodinase activities in FRTL5 cells: predominance of type I 5'-deiodinase. Endocrinology 126:3059–3068[Abstract]
  12. Schoenmakers CH, Pigmans IG, Visser TJ 1992 Species differences in liver type I iodothyronine deiodinase. Biochim Biophys Acta 1121:160–166[CrossRef][Medline]
  13. Baur A, Bauer K, Jarry H, Kohrle J 1997 3,5-Diiodo-L-thyronine stimulates type I 5'-deiodinase activity in rat anterior pituitaries in vivoand in reaggregate cultures and in GH3 cells in vitro. Endocrinology 138:3242–3248[Abstract/Free Full Text]
  14. Sanders JP, van der Geyten S, Kaptein E, Darras VM, Kuhn ER, Leonard JL, Visser TJ 1997 Characterization of a propylthiouracil-insensitive type I iodothyronine deiodinase. Endocrinology 138:5153–5160[Abstract/Free Full Text]
  15. Toyoda N, Zavacki AM, Maia AL, Harney JW, Larsen PR 1995 A novel retinoic X receptor-independent thyroid hormone response element is present in the human type I deiodinase gene. Mol Cell Biol 15:5100–5112[Abstract]
  16. Toyoda N, Nishikawa M, Horimoto M, Yoshikawa N, Mori Y, Yoshimura M, Masaki H, Tanaka K, Inada M 1990 Graves immunoglobulin G stimulates iodothyronine 5'deiodinating activity in FRTL-5 rat thyroid cells. J Clin Endocrinol Metab 70:1506–1511[Abstract]
  17. Croteau W, Davey JC, Galton VA, St Germain DL 1996 Cloning of the mammalian type II iodothyronine deiodinase. A selenoprotein differentially expressed and regulated in human and rat brain and other tissue. J Clin Invest 98:405–417[Medline]
  18. 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[Medline]
  19. 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[Abstract/Free Full Text]
  20. Salvatore D, Bartha T, Harney JW, Larsen PR 1996 Molecular biological and biochemical characterization of human type 2 selenodeiodinase. Endocrinology 137:3308–3315[Abstract]
  21. Murakami M, Araki O, Hosoi Y, Kamiya Y, Morimura T, Ogiwara S, Mizuma H, Mori M 2001 Expression and regulation of type II iodothyronine deiodinase in human thyroid gland. Endocrinology 142:2961–2967[Abstract/Free Full Text]
  22. De Jesus LA, Carvalho SD, Ribeiro MO, Schneider M, Kim S, Harney J, Larsen PR, Bianco AC 2001 The type 2 iodothyronine deiodinase is essential for adaptive thermogenesis in brown adipose tissue. J Clin Invest 108:1379–1385[CrossRef][Medline]
  23. Galton VA, Martinez E, Hernandez A, St Germain EA, Bates JM, St Germain DL 2001 The type 2 iodothyronine deiodinase is expressed in the rat uterus and induced during pregnancy. Endocrinology 142:2123–2128[Abstract/Free Full Text]
  24. St Germain DL 1988 The effects and interactions of substrates, inhibitors, and the cellular thiol-disulfide balance on the regulation of type II iodothyronine 5'-deiodinase. Endocrinology 122:1860–1868[Abstract]
  25. Silva JE, Leonard JL 1985 Regulation of rat cerebrocortical and adenohypophyseal type II 5'-deiodinase by thyroxine, triiodothyronine, and reverse triiodothyronine. Endocrinology 116:1627–1635[Abstract]
  26. Leonard JL, Silva JE, Kaplan MM, Mellen SA, Visser TJ, Larsen PR 1984 Acute posttranscriptional regulation of cerebrocortical and pituitary iodothyronine 5'-deiodinases by thyroid hormone. Endocrinology 114:998–1004[Abstract]
  27. Gereben B, Kollar A, Harney JW, Larsen PR 2002 The mRNA structure has potent regulatory effect on type 2 iodothyronine deiodinase expression. Mol Endocrinol 16:1667–1679[Abstract/Free Full Text]
  28. Gereben B, Goncalves C, Harney JW, Larsen PR, Bianco AC 2000 Selective proteolysis of human type 2 deiodinase: a novel ubiquitin-proteasomal mediated mechanism for regulation of hormone inactivation. Mol Endocrinol 14:1697–1708[Abstract/Free Full Text]
  29. Schneider MJ, Fiering SN, Pallud SE, Parlow AF, St Germain DL, Galton VA 2001 Targeted disruption of the type 2 selenodeiodinase gene results in a phenotype of pituitary resistance to T4. Mol Endocrinol 15:2137–2148[Abstract/Free Full Text]
  30. Campos-Barros A, Hoell T, Musa A, Sampaolo S, Stoltenburg G, Pinna G, Eravci M, Meinhold H, Baumgartner A 1996 Phenolic and tyrosyl ring iodothyronine deiodination and thyroid hormone concentrations in the human central nervous system. J Clin Endocrinol Metab 81:2179–2185[Abstract]
  31. Darras VM, Hume R, Visser TJ 1999 Regulation of thyroid hormone metabolism during fetal development. Mol Cell Endocrinol 151:37–47[CrossRef][Medline]
  32. Galton VA, Martinez E, Hernandez A, St Germain EA, Bates JM, St Germain DL 1999 Pregnant rat uterus expresses high levels of the type 3 iodothyronine deiodinase. J Clin Invest 103:979–987[Medline]
  33. Salvatore D, Low SC, Berry M, Maia AL, Harney JW, Croteau W, St Germain DL, Larsen PR 1995 Type 3 iodothyronine deiodinase: cloning, in vitro expression, and functional analysis of the placental selenoenzyme. J Clin Invest 96:2421–2430
  34. Kaplan MM, Yaskoski KA 1980 Phenolic and tyrosyl ring deiodination of iodothyronines in rat brain homogenates. J Clin Invest 66:551–552
  35. Tu HM, Legradi G, Bartha T, Salvatore D, Lechan RM, Larsen PR 1999 Regional expression of the type 3 iodothyronine deiodinase messenger ribonucleic acid in the rat central nervous system and its regulation by thyroid hormone. Endocrinology 140:784–790[Abstract/Free Full Text]