| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
BRIEF COMMUNICATION |
Thyroid Section, Division of Endocrinology, Diabetes and Hypertension, Brigham and Women’s Hospital, Boston, Massachusetts 02115
Address all correspondence and requests for reprints to: Ann Marie Zavacki, Thyroid Section, Division of Endocrinology, Diabetes and Hypertension, Brigham and Women’s Hospital, HIM 641, 77 Ave Louis Pasteur, Boston, Massachusetts 02115. E-mail: azavacki{at}rics.bwh.harvard.edu.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
0.1-fold for both), they were 8-fold and 2-fold higher, respectively, relative to C3H mice. Thyroidal D1 and cerebral cortical type 3 deiodinase activity were unchanged between C3H-D2KO and C3H mice. In conclusion, C3H-D2KO mice have notably elevated serum T4 levels, and this, in conjunction with residual D1 activity, is likely an important role in the maintenance of euthyroid serum T3 concentrations. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Three rodent models exist in which the effects of deiodinase deficiency on extrathyroidal T4 to T3 conversion may be studied. These include the C3H/HeJ (C3H), Dio1–/– (D1KO), and Dio2–/– (D2KO) mice. In C3H mice, a decreased expression of the Dio1 gene results in both liver and kidney D1 activities being reduced to approximately 10% to 20% that of most wild-type mice, e.g. C57/BL6 (C57) (3, 4, 5). These mice have no obvious phenotypic abnormalities despite their relative deficiency of D1, with serum T3 and TRH-stimulated TSH levels being normal; however, T4 is increased by approximately 60% over that of C57 mice (3). Pituitary D2 activity is also decreased by approximately 50% in these mice, presumably as a result of high T4 levels increasing T4-mediated proteosomal degradation of this enzyme (3, 6, 7).
D1KO mice with a targeted disruption of the Dio1 gene are very similar to C3H with normal serum T3 and TSH, whereas T4 is elevated (8). These mice have no obvious abnormalities; however, a detailed analysis of these animals has revealed they are slightly heavier than wild-type siblings and have greater fecal iodothyronine excretion (8).
Mice with a targeted disruption of the Dio2 gene are also apparently euthyroid with normal serum T3 and hepatic D1 activity, although an elevation in serum T4 and TSH indicates a central insensitivity to T4 (9). D2KO mice exhibit isolated hypothyroidism in tissues that depend on D2-catalyzed T4 to T3 conversion to regulate thyroid status, i.e. brain, and brown adipose tissue. Thus, these animals have elevated TSH, are deaf, and are unable to sustain their body temperature during cold exposure as a result of impaired energy expenditure (9, 10, 11).
Still, it is remarkable that there are no further abnormalities in any of these mice. This leads one to wonder, is the T3-generating capacity of the remaining D1 and/or D2 in these animals able to compensate such that more severe defects are not manifested? Furthermore, what would happen if both D1 and D2 activity were impaired? To address this question, we backcrossed the targeted disruption of the Dio2 gene into a C3H background to generate the C3H-D2KO mice. Our results indicate that remarkably, despite very low D1 activity and no D2 expression, C3H-D2KO mice maintain euthyroid plasma T3 levels, and no apparent additional phenotypic abnormalities exist in these animals.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Generation of Dio1C3H/C3H-Dio2–/–, C3H-D2KO, and C57-D2KO mice
Male mice homozygous for a targeted disruption of the Dio2 gene (Dio2–/–) (9) were mated with C3H females and the resulting Dio1WT/C3H-Dio2+/– heterozygotes were interbred to generate Dio1WT/WT-Dio2+/+, Dio1C3H/C3H-Dio2+/+, Dio1WT/WT-Dio2–/–, and Dio1C3H/C3H-Dio2–/– mice. To generate C3H-backcrossed animals, male Dio2–/– mice were mated with C3H females to create F1 with the resulting male Dio2–/+ heterozygotes being crossed with a C3H female three more times to generate the F4. F4 Dio2–/+ heterozygous females were then mated with C3H males to ensure the Y chromosome was also C3H-derived in the F5 offspring. Heterozygous F5 males and females were then interbred to generate animals predicted to be 97% C3H (12) and these animals are referred to as C3H-D2KO mice. A similar strategy was used to generate animals in which the targeted disruption of the Dio2 allele was backcrossed into a C57 background, and these animals are referred to as D2KO mice in this article.
Genotyping
Animals were tailed at 3 wk of age, and genomic DNA was extracted using DNAzol (Invitrogen, Carlsbad, CA) following the manufacturer’s instructions. Mice were genotyped by PCR using the following strategies:
Dio1.
Fifty nanograms of genomic DNA were amplified using Platinum GenoTYPE Tsp DNA polymerase (Invitrogen) following the manufacturer’s specifications with the following modifications: reactions contained 32 µM dATP, dCTP, dTTP, and 40 µM 7-deaza-dGTP (Roche, Indianapolis, IN) and 0.2 µl 
-32P dCTP (6000 Ci/mmol; NEN Life Science Products, Boston, MA), 1x PCR Enhancer solution (Invitrogen), 2.5 mM MgSO4, 2.5% DMSO, and 0.4 µM sense and antisense primers (sense: 5'-GCAGCGTCCATTCTCATTTAC-3', antisense 5'-TCTTAACGGACTGCCCAGG-3'). PCR products were resolved on a 10% TBE buffered acrylamide gel (Bio-Rad, Hercules, CA), which was then dried and exposed to film.
Dio2.
Fifty nanograms of genomic DNA were amplified using Red Taq DNA polymerase (Sigma, St. Louis, MO) following the manufacturer’s protocol using a mix containing the following primers: WT sense primer (5'-GTTTAGTCATGGAAGCAGCACTATG-3'), D2KO sense primer (5'-CGTGGGATCATTGTTTTTCTCTTG-3'), and common antisense primer (5'-CATGGCGTTAGCCAAAACTCATC-3') (13). Fragments were resolved by 1.5% agarose gel electrophoresis.
T3, T4, reverse T3 (rT3), and TSH measurements, and T3-charcoal uptake
Serum T3 concentrations were either measured as described previously (14, 15) using a standard curve prepared by diluting a known amount of T3 into charcoal-stripped mouse serum (Sigma) and primary anti-T3 antibody at a 1/100,000 final concentration or by using the same standard curve with COAT-A-COUNT total T3 kit (DPC, Los Angeles, CA). Serum T4 values and 125I T3 charcoal uptake to assess serum T3-binding proteins were measured as described previously (15). rT3 was measured by the Adaltis Reverse T3 RIA kit (Adaltis Italia S.p.A., Casalecchio di Reno, Italy, distributed by Polymedco, Cortland Manor, NY) using 50 µl of mouse serum and a standard curve prepared by diluting a known amount of rT3 into charcoal-stripped mouse serum. TSH was determined using the rat TSH 125I Biotrak Assay System from Amersham Biosciences (Piscataway, NJ) with slight modifications (16) with samples all falling within the linear range of a curve generated by the serial dilution of hypothyroid mouse serum.
D1, D2, and D3 assays
Deiodinase assays were performed as described previously (17). Briefly, tissues were sonicated in buffer containing 0.1 M KPO4, 1 mM EDTA, 0.25 M sucrose, and 10 mM dithiothreitol (DTT) and protein concentrations were determined using Bio-Rad Protein Assay reagent (Bio-Rad). For D1 assays of liver, kidney, thyroid, and pituitary, 5–15 µg, 15–50 µg, 3–6 µg, and 20–150 µg, respectively, were assayed with a final concentration of 10 mM DTT and 500 nM 125I rT3. For all other tissues analyzed for D1 expression 150 µg of tissue was assayed for 6 h with a final concentration of 10 mM DTT and 250 nM 125I rT3 with background levels of deiodination assessed under the same conditions with the addition of 1 mM propylthiouracil. For D2 assays, 20–125 µg of cerebral cortex or 50 µg of pituitary was assayed with a final concentration of 20 mM DTT, 5.0 nM 125I T4, and 1 mM propylthiouracil. D3 assays were performed using 20–125 µg of cerebral cortex protein with 10 mM DTT, 10 nM 125I T3, and 1 mM propylthiouracil.
Thyroidal radioiodine uptake
Mice were injected with 15,000 cpm/g body weight Na125I (NEN Life Science Products) ip. Two hours after injection (at a time when uptake was determined to be maximal and in a stationary phase for C57 mice, data not shown), mice were killed and serum and thyroid were collected and counted. Data are expressed as thyroidal cpm/cpm 1 ml serum.
Real-time PCR
Real-time PCR was performed as described previously using the QuantiTect SYBR Green PCR kit on an I-Cycler (Bio-Rad) (15).
Assessment of TSH bioactivity
TSH bioactivity was determined following previously described methods (18, 19). Briefly, 100,000 CHO JP 26–26 cells stably transfected with human TSH receptor (a generous gift of Drs. G. Vassart and S. Refetoff) were seeded in a Costar 48 multiwell dish in Ham’s F-12 media supplemented with 10% fetal calf serum, 15 µg/ml gentamicin, 50 µg/ml ampicillin, 2 mM glutamine, 1 mM sodium pyruvate, and 250 µg/ml geneticin (Life Technologies, Inc., Grand Island, NY). The next day, media was removed and cells were preincubated for 1 h in Krebs-Ringers HEPES buffer, pH 7.4, with 8 mM glucose with 0.05% BSA at 37 C. This was replaced with the same buffer plus 25 µM Rolipram and 10% of the serum to be tested for 1 h at 37 C. At the conclusion of the experiment, media was removed and the experiment terminated by the addition of 0.1 M HCl. Samples were neutralized using 0.2 M NaOH, and cAMP generated was determined using the cAMP [125I] RIA kit (Perkin-Elmer, Shelton, CT). Background cAMP generation was determined using serum from TSH-depleted mice that had been implanted with pellets releasing 5 µg T3/d (Innovative Research of America, Sarasota, FL) for 10 d in which T4 levels were completely suppressed (data not shown). Samples measured were also confirmed to be within the linear range of measurement by using serial dilutions of hypothyroid mouse serum and did not generate additional cAMP in CH0 JP 02 cells without stably transfected TSH receptor.
Statistical analysis
When two groups were compared, statistical significance was determined using a two-tailed Student’s t test, whereas when multiple groups were compared, one-way ANOVA with a Newman-Keuls posttest was used (Prism 4; GraphPad Software, San Diego, CA).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
). For D2 genotyping, we used 5' sense primers either directed against sequences in the wild-type Dio2 gene or the inserted neomycin gene and a common Dio2 antisense primer. This resulted in a 396 bp PCR product for the wild-type Dio2 allele, whereas the amplicon for the neomycin-containing disrupted Dio2 gene is 450 bp (Fig. 1B).
|
). Serum T4 levels were increased 73% in Dio1wt/wt-Dio2–/– mice lacking the Dio2 gene (P < 0.05) and 62% in Dio1C3H/C3H-Dio2+/+ mice containing the C3H allele of the Dio1 gene (P < 0.05) relative to wild-type (Dio1wt/wt-Dio2+/+) siblings in accordance with previous results (3, 4, 9). However, in Dio1C3H/C3H-Dio2–/– mice, T4 levels were even further elevated being double that of wild-type littermates (P < 0.01). TSH was not different between wild-type littermates and mice containing the C3H allele of the Dio1 gene (Table 1). Serum TSH of mice lacking a functional Dio2 gene was elevated by 34%, although this was not significantly different from wild-type littermates. However, serum TSH levels of the Dio1C3H/C3H-Dio2–/– mice were approximately 50% higher that of wild-type siblings (P < 0.01).
|
As a result of this concern, and the limitations imposed by the small number of animals homozygous for both markers generated in each cross (one of 16), we decided to use a second strategy for creating mice with low D1 expression lacking the Dio2 gene. Thus, mice with a targeted disruption of the Dio2 gene were backcrossed with C3H mice for five generations, generating animals with a genetic background predicted to be 97% C3H-derived, and these animals are referred to as C3H-D2KO mice (12). A similar strategy was used to generate animals in which the targeted disruption of the Dio2 allele was backcrossed into a C57 background with wild-type expression of the Dio1 allele, and these animals are referred to as D2KO mice (4).
C3H-D2KO mice have normal serum T3 values, whereas T4 is elevated
C3H and C3H-D2KO mice have litters that are comparable in size (5.6 ± 0.9 vs. 4.6 ± 0.7 pups/litter, mean ± SEM, P > 0.05), suggesting that no defects in embryonic development or fertility are associated with this genotype. C3H-D2KO male mice were slightly heavier and longer than C3H mice, weighing 14% more at 56 d of age (P < 0.01) and being 9.1 vs. 8.6 cm in length (P < 0.01).
Remarkably, like with the first strategy, measurement of serum T3 showed no difference among male C57, D2KO, C3H, and C3H-D2KO, indicating that despite a deficiency of both D1 and D2, C3H-D2KO mice maintain normal serum T3 levels (P > 0.05) (Fig. 2A). Additionally, charcoal T3 uptake assays were not significantly different between any of the groups, suggesting that the free fraction of serum T3 is not different between any of these animals (data not shown). T4 was increased 50% in the D2KO mice relative to that of C57 (P < 0.05), whereas that of the C3H-D2KO mice was slightly increased by 16% (P < 0.01) when compared with the already elevated C3H T4 values (Fig. 2B). Strikingly, when C3H-D2KO serum T4 is compared with that of C57, T4 is doubled in the C3H-D2KO mice. These results are mirrored when one compares the T3/T4 ratio in individual mice with the ratios being 0.14 ± 0.1, 0.10 ± 0.1, 0.09 ± 0.1, and 0.08 ± 0.01 for C57, D2KO, C3H, and C3H-D2KO mice, respectively, indicating a progressive increase in T4 levels, whereas T3 remains unchanged. When serum rT3 levels were measured, mice with impaired D1 expression (i.e. C3H and C3H-D2KO) had rT3 values approximately 2.5 times those of mice with wild-type D1 expression (i.e. C57 and D2KO) consistent with previous reports of serum rT3 levels being increased in C3H mice (Fig. 2C) (3).
|
). In agreement with previous results, serum TSH in D2KO mice was elevated, being over twice that of C57 controls (P < 0.05) (Fig. 2D) (9). TSH levels of C3H-D2KO mice were increased to a lesser extent than that of the D2KO mice, being increased by 40% relative to C3H (P < 0.01).
To determine whether the observed increases in TSH correlated with increased thyroidal activity, thyroidal 125I uptake was performed (Fig. 2E). No significant difference was observed between C57 and D2KO or C3H and C3H-D2KO animals after 2 h, although there appeared to be more 125I uptake in mice with a C3H vs. a C57 background despite their equivalent amounts of TSH. Because the bioactivity of TSH can vary (21), this suggested that C3H and C57 TSH could have differential biological potencies with respect to activating the TSH receptor. However, when the amount of cAMP generated by both C3H and C57 serum was measured using CHO JP 26–26 cells stably transfected with the TSH receptor (19) and then normalized to the amount of immunoreactive TSH, there was no difference between either group of animals (126.7 ± 23.5 vs. 110.4 ± 23.2 pmol of cAMP/ng TSH for C3H and C57, respectively, n = 5–6 mice/group).
Measurement of deiodinase activity in C57, D2KO, C3H, and C3H-D2KO mice
C3H and C3H-D2KO hepatic D1 activities were both reduced when compared with C57, being 1% and 11% that of C57 respectively (Table 2). Unexpectedly, hepatic D1 activity was significantly increased (8.4-fold) in C3H-D2KO mice relative to C3H (P < 0.01). Because D1 is a very sensitive marker of peripheral thyroid status (15), mRNA levels of D1 and two other hepatic thyroid hormone-responsive genes, Spot 14 and -glycerol phosphate dehydrogenase (-GPD) were also assessed (Table 3). D1 mRNA levels closely paralleled those of D1 activity with D1 mRNA levels of C3H and C3H-D2KO being 1% and 6.7% that of C57, respectively, and C3H-D2KO D1 mRNA levels being increased 6-fold relative to C3H. However, measurement of mRNA expression of both spot 14 and -GPD by real-time PCR (15) showed no change between C3H and C3H-D2KO mice.
|
|
). Although renal D1 activity was also decreased in D2KO mice when compared with C57, this difference was not found to be significant (P > 0.05). Thyroidal D1 activity was not significantly different between C57 and D2KO or C3H and C3H-D2KO mice. Notably, mice with a C3H background had approximately one half the thyroidal D1 activity of mice with a C57 background. Because of the increase in hepatic and renal D1 activity in the C3H-D2KO mice when compared with C3H animals, we assayed cerebral cortex, testis, brown adipose tissue, white adipose tissue, lung, heart, intestine, gastrocnemius muscle, and skin for D1 expression. Most tissues assayed had either no or very low (<50 fmol/min/mg) D1 activity (data not shown). However, large intestine did have D1 activity comparable to that found in C3H liver (0.19 pmol/min/mg C57, 0.30 pmol/min/mg D2KO, 0.13 pmol/min/mg C3H, and 0.34 pmol/min/mg C3H-D2KO), although no increase of the magnitude observed between C3H and C3H-D2KO hepatic D1 activity was apparent.
In accordance with previous results, D2 activity in cerebral cortex and pituitary was decreased in C3H when compared with C57 mice (Table 2) (4). Cerebral cortical D3 was not different between C57 and D2KO mice, unlike previous reports in which it is elevated (Table 2) (9). However, D3 activity was the same between C3H and C57 mice, in agreement with previous results (3).
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-GPD relative to C3H, demonstrating a remarkable adaptive capacity.
How do the C3H-D2KO mice maintain normal serum T3 levels? Although the sources of the circulating T3 in the C3H mouse have not been defined, the substantial increase in T4 relative to that of C57 mice suggests that increased thyroidal activity is playing at least a partial role. Our results showing a higher thyroidal 125I uptake in C3H mice relative to C57 mice support this conclusion (Fig. 2E).
In the rodent, it is thought that D1 and D2 play an approximately equal role in peripheral T3 production (reviewed in [2 ]). Thus, the lower D2 activity found in C3H mice vs. C57 (as a consequence of the posttranslational effect of the marked increase in serum T4) observed in this (Table 2) and earlier studies (4) argues that hepatic and renal D1, albeit much lower than that of C57 animals, is providing a substantial fraction of the T3 generated from the increased serum T4 in the C3H mouse. Thus, the relative deficiency of D2 already present in the C3H mouse can explain the greater percentage increase in serum T4 in the D2KO relative to the C57 mouse than in the C3H-D2KO vs. C3H mouse (1.5-fold vs. 1.2-fold, Fig. 2B). The greater stress caused by the requirement to normalize T3 in the D2KO mouse in the C57 background resulting from its lower initial T4 may also explain why the serum TSH is significantly higher in the D2KO than in the C57 parent strain. It seems likely that in the C3H-D2KO, the small increase in the already elevated serum T4, when combined with the 8- and 2-fold increases in hepatic and renal D1, result in circulating T3 homeostasis. Additionally, our data rule out both an increase in thyroidal D1 and ectopic induction of D1 in other tissues as making contributions to peripheral T4 to T3 conversion in the C3H-D2KO mice (Table 2).
The fact that total D1 activities in the C3H-D2KO liver and kidney are much lower than they are in the D2KO mice but the serum T3/T4 ratios are virtually the same (0.08 vs. 0.10) suggests that there may be a considerable excess in total body D1 activity. A comparison of the C3H-D2KO mouse with mice with no D1 and D2 will provide further insight into the role of this small residual D1 in the C3H-D2KO in the maintenance of thyroid hormone homeostasis (22).
T3 can be inactivated by inner-ring deiodination through D3 or D1 and thus a decrease in T3 clearance rates could also play a role in maintaining euthyroid serum T3 levels in the C3H-D2KO mice (2). However, cerebral cortical D3 levels of C3H-D2KO animals are not different from that of C3H or C57 mice (Table 2), suggesting that there is no change in D3-mediated T3 clearance in these animals. Furthermore, D1KO mice have no change in their clearance rates of physiological amounts of T3; thus, it is not likely that mice with low levels of D1 such as C3H would have a change in their T3 clearance rates (8).
What is the basis of this increase in T4 in C3H-D2KO animals? Their elevated TSH relative to C3H suggests that thyroidal activity should be increased with more production of T4 (Fig. 2D). Yet, thyroidal 125I uptake is not increased in C3H-D2KO mice relative to C3H (Fig. 2E). However, thyroidal 125I uptake does indicate that C3H already have a greater capacity to concentrate iodide than C57 and that their thyroids may be more active, producing more T4 (Fig. 2E). This difference, when taken in context of the similar serum TSH levels found in the C3H and C57 animals, implies that these strains may have different TSH bioactivity. However, our data indicate there is no difference between C3H and C57 TSH bioactivity and thus the increase in C3H thyroidal 125I uptake may be the result of other yet to be defined compensatory mechanisms in these mice. These findings are in agreement with studies by Pohlenz et al. (16) comparing T4, TSH, and thyroidal radioiodine uptake between several strains of mice, including C57, which also indicate that C57 TSH is not less bioactive than that of other mouse strains. An additional factor that could further enhance the accumulation of T4 in C3H-D2KO mice is the decreased serum T4 clearance rate previously reported in D2KO mice (9).
The increase in D1 enzyme activity in C3H-D2KO liver is associated with a parallel increase in mRNA levels in agreement with several other studies showing a close correlation between D1 activity and mRNA levels (Tables 2 and 3) (3, 4, 5). Although the Dio1 gene is a known to be a very T3-responsive gene, expression of the T3-responsive hepatic genes, spot 14 and -glycerol phosphate dehydrogenase, was not increased in C3H-D2KO mice, suggesting that their livers are not thyrotoxic (15). Alternatively, these genes may simply not be sensitive enough markers to detect a slight alteration in thyroid status or some other component of Dio1 transcription or mRNA stability may be affected.
rT3 levels are elevated 2.5-fold in C3H mice as expected when compared with C57 animals (3) with no further increase in C3H-D2KO mice. Notably, there is also no change in rT3 concentrations between C57 and D2KO mice. This strongly suggests that D2 plays a minimal role in the clearance of rT3 (Fig. 2C).
Although the tissue-specific thyroid status of the C3H-D2KO mice remains to be investigated in detail, overall our results indicate that few phenotypic consequences not already associated with the D2KO mouse result from a combined deficiency in Dio1 and loss of Dio2 gene expression. Our results with the C3H-D2KO mouse further underscore the potent network of mechanisms that exists to maintain active thyroid hormone at appropriate levels.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Disclosure Statement: The authors have nothing to declare.
First Published Online November 30, 2006
Abbreviations: D1 or D2, Type 1 or 2 iodothyronine deiodinases; DTT, dithiothreitol; rT3, reverse T3.
Received July 31, 2006.
Accepted for publication November 25, 2006.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  J. A. Hall and A. C. Bianco Triumphs of the Thyroid Despite Lesser Conversion Endocrinology, June 1, 2009; 150(6): 2502 - 2504. [Full Text] [PDF]
|
|
|
|
|
|
V. A. Galton, M. J. Schneider, A. S. Clark, and D. L. St. Germain Life without Thyroxine to 3,5,3'-Triiodothyronine Conversion: Studies in Mice Devoid of the 5'-Deiodinases Endocrinology, June 1, 2009; 150(6): 2957 - 2963. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. N. Walpita, A. D. Crawford, E. D. R. Janssens, S. Van der Geyten, and V. M. Darras Type 2 Iodothyronine Deiodinase Is Essential for Thyroid Hormone-Dependent Embryonic Development and Pigmentation in Zebrafish Endocrinology, January 1, 2009; 150(1): 530 - 539. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. Sanchez, P. S. Singru, R. Acharya, M. Bodria, C. Fekete, A. M. Zavacki, A. C. Bianco, and R. M. Lechan Differential Effects of Refeeding on Melanocortin-Responsive Neurons in the Hypothalamic Paraventricular Nucleus Endocrinology, September 1, 2008; 149(9): 4329 - 4335. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Ohguchi, T. Tanaka, A. Uchida, K. Magoori, H. Kudo, I. Kim, K. Daigo, I. Sakakibara, M. Okamura, H. Harigae, et al. Hepatocyte Nuclear Factor 4{alpha} Contributes to Thyroid Hormone Homeostasis by Cooperatively Regulating the Type 1 Iodothyronine Deiodinase Gene with GATA4 and Kruppel-Like Transcription Factor 9 Mol. Cell. Biol., June 15, 2008; 28(12): 3917 - 3931. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
