Fasting-Induced Increase in Type II Iodothyronine Deiodinase Activity ...

0013-7227/98/$03.00/0 Endocrinology Copyright © 1998 by The Endocrine Society

Vol. 139, No. 6 Printed in U.S.A.

Fasting-Induced Increase in Type II Iodothyronine Deiodinase Activity and Messenger Ribonucleic Acid Levels Is Not Reversed by Thyroxine in the Rat Hypothalamus* SABRINA DIANO, FREDERICK NAFTOLIN, FERNANDO GOGLIA, TAMAS L. HORVATH

AND

Department of Obstetrics and Gynecology, Yale University School of Medicine (S.D., F.N., T.L.H.), New Haven, Connecticut 06520; and the Department of General and Environmental Physiology (S.D., F.G.), University of Naples “Federico II,” Naples, Italy ABSTRACT The importance of local formation of T3 in the feedback effect of the thyroid gland on hypothalamic TRH-producing cells has been established. Primary failure of the thyroid gland results in a fall in circulating T4 and T3 levels, leading to an elevation in the production and release of TRH in the hypothalamic paraventricular nucleus. In contrast, during short term fasting, declining plasma levels of thyroid hormones coincide with suppressed TRH production and release. In the brain, the prevalent enzyme that converts T4 to T3 is type II iodothyronine deiodinase (DII). The present study was undertaken to determine whether a differential hypothalamic expression of type II deiodinase may exist in fasted rats and in animals that are hypothyroid due to the failure of the thyroid gland. Using in situ hybridization, we assessed type II deiodinase messenger RNA (mRNA) levels in the hypothalamus of rats that were control euthyroid, hyperthyroid (T4), hypothyroid induced by propylthiouracil (PTU), and fasted. A group of fasted rats also received exogenous T4. DII mRNA was detected around the third ventricle, including the ependymal layer and adjacent periventricular regions as well as in the arcuate nucleus and the external layer of the median eminence. Quantitative in situ hybridization analysis demonstrated that PTU treatment and short term fasting resulted in significant elevations in DII messenger levels compared with those in euthyroid controls. Three weeks of PTU administration induced a consistent decline in circulating T3 and undetectable T4 levels, whereas 3 days of fasting resulted in only a 50%

H

YPOTHYROIDISM leads to a change in the activity of iodothyronine deiodinases. The regulation of these enzymes occurs in a tissue-specific manner (1). For example, deiodinase type I (DI) is up-regulated in the thyroid gland of human and rat in hypothyroid conditions, whereas under the same circumstance, a decreased activity of this enzyme is present in other peripheral tissues (1– 4). On the other hand, deiodinase type II (DII), present in the central nervous system (5, 6), pituitary (7), brown adipose tissue (8), and placenta (9), shows increased activity when plasma T4 declines. Received December 10, 1997. Address all correspondence and requests for reprints to: Dr. Sabrina Diano, Department of Obstetrics and Gynecology, Yale Medical School, 333 Cedar Street, FMB 339, New Haven, Connecticut 06520. E-mail: [email protected]. * This work was supported by NSF Grant IBN-9728581, NIH Grant NS-36111, a fellowship from the University of Naples “Federico II,” and Regione Campania L.R. 31.12.1994, noB041, Italy.

fall in the concentration of serum thyroid hormones. Interestingly, however, the expression of the DII mRNA was more than 2-fold higher in fasted animals compared with the values in PTU-treated rats. Furthermore, although T4 administration repressed DII mRNA expression in euthyroid animals, the same treatment had no effect on the fasting-induced elevations of DII message. To assess whether DII enzymatic activity is also affected during food deprivation, hypothalami were dissected out, and DII activity was measured in control euthyroid, fasted, and fasted plus T4-treated rats. To determine whether comparable changes in plasma thyroid hormone levels induced by fasting and PTU treatment could have affected DII enzymatic activity in a similar manner, animals were injected ip with PTU for 5 days to decrease plasma thyroid hormones to levels similar to those caused by fasting. DII enzymatic assay showed a significant increase in DII activity in fasted and fasted plus T4-treated animals compared with those in euthyroid controls and PTU-treated rats. No significant changes were found in PTU-treated rats compared with euthyroid animals. These data indicate that during short term fasting, a signal of nonthyroid origin underlies the robust elevation of DII production and activity in the hypothalamus. Thus, we propose that during the initial phase of food deprivation, an increased negative thyroid feedback exists on the hypothalamus due to locally formed T3. This local hyperthyroidism may, in turn, induce the suppression of TRH under these conditions. (Endocrinology 139: 2879 –2884, 1998)

It has been suggested that the major role of the DII is to maintain T3 homeostasis, producing adequate intracellular levels of T3 to ensure all T3-dependent cellular functions in the tissue (10, 11). In fact, a recent report demonstrated that in hypothyroid adult rats, the increased activity of DII is able to normalize T3 levels in the brain and other tissues after infusion of T4 at doses insufficient to adjust plasma T4 and T3 levels (12). In parallel with altered DII activity, failure of the thyroid gland [thyroidectomy or chemical inhibition of thyroid hormone production through 6-n-propyl-2-thiouracil (PTU), methimazole, or methimazole plus iopanoic acid] induces a modest elevation in hypothalamic TRH levels (13). In contrast, hypothyroid conditions induced by fasting coincide with suppressed production and release of TRH in the hypothalamic paraventricular nucleus and median eminence (14). The central mechanism that underlies the emergence of this apparent paradox in thyroid feedback is ill defined. We

2879

2880

FASTING UP-REGULATES HYPOTHALAMIC TYPE II DEIODINASE mRNA

propose that a differential expression of type II deiodinase during fasting-induced hypothyroidism and hypothyroidism due to the failure of the thyroid gland may contribute to the development of this apparent paradox in thyroid feedback. Thus, the aim of the present study was to assess the effects of fasting, T4, and the thyroid depressant, PTU, on DII messenger RNA (mRNA) levels in the hypothalamus. As mRNA is not necessarily translated into functional protein, the enzymatic activity of DII was also measured in those hypothalamic samples. Materials and Methods Animals and tissue preparation Forty-five male Sprague-Dawley rats from Charles River Laboratories (Wilmington, MA; 200 –250 g BW) were housed under a 12-h light, 12-h dark cycle in a temperature maintained at 21–23 C. Rats were divided into six experimental groups: intact animals (n 5 10), animals treated with PTU for 3 weeks (n 5 5) or for 5 days (n 5 5), euthyroid animals treated with T4 (n 5 5), 3-day food-deprived animals (n 5 10), and 3-day food-deprived animals treated with T4 (n 5 10). To induce hypothyroid conditions in the group of animals used for the in situ hybridization study, PTU was implanted sc in a pellet (100 mg/3-week release; Innovative Research of America, Sarasota, FL) (15–17). The pellet was kept in place for 3 weeks. The second group of PTU animals was ip injected for 5 days with 2 mg PTU/day in saline. Fasted and euthyroid animals received [l-T4 (Sigma Chemical Co., St. Louis, MO) by daily (for 3 days) ip injection at a dose of 200 mg/100 g BW. The last injection of T4 or PTU was given 24 h before death. All animals had free access to water. In the fasting experiments, rats were housed individually in cages to avoid coprophagia. For the in situ hybridization histochemistry, the brains (n 5 25) and the pituitary glands of euthyroid (n 5 5), hyperthyroid (n 5 5), 3-week PTU-treated (n 5 5), 3-day fasted (n 5 5), and 3-day fasted plus T4treated (n 5 5) animals were removed immediately after decapitation, frozen on dry ice, and stored at 280 C. For the measurement of DII activity, the brains of euthyroid (n 5 5), 5-day PTU-treated (n 5 5), 3-day fasted (n 5 5), and 3-day fasted plus T4-treated (n 5 5) animals were removed, frozen, and sliced coronally. The hypothalamic region between the optic chiasm and the mamillary bodies, containing the median eminence (ME), arcuate nucleus (ARC), and periventricular area (PE), was cut and used for the assay. Trunk blood was obtained from each animal, and the serum was separated and frozen at 220 C before measurement of thyroid hormone concentrations. For in situ hybridization, the frozen brains were allowed to equilibrate in a cryostat at 220 C. Coronal sections were cut at 16 mm and then mounted onto poly-l-lysine-coated slides. Brain sections were collected, beginning rostrally at the optic chiasm and continuing caudally to the premamillary area. Coronal sections from each animal were anatomically matched across experimental groups. The tissue sections were stored at 280 C until processing for in situ hybridization histochemistry. For DII activity, tissues were homogenized on ice in 5 vol 0.25 m sucrose and 10 mm HEPES (pH 7.0) containing 10 mm dithiothreitol (DTT), immediately frozen in dry ice, and stored at 280 C until assay.

In situ hybridization histochemistry A 481-bp fragment of complementary DNA (cDNA) of DII was amplified based on the RT-PCR reaction, using specific oligonucleotide primers derived from the coding region of the rat DII sequence (59-ACT CGG GAA TTC TGC TCA AGC ACG T-39 and 59-ACA TGG ATC CTC TTG GTT CCG GTG CT-39) (18). Total RNA was extracted from the rat pituitary glands by the guanidium thiocyanate-phenol-chloroform method using TRIzol reagent (Life Technologies, Grand Island, NY) and transcribed using the First-Strand cDNA Synthesis Kit (Pharmacia Biotech, Piscataway, NJ). PCR reaction was carried out using the following protocol: 3 mg cDNA template, 0.5 mm primers, 1.25 mm MgCl2, 80 mm deoxy-NTP, and 2 U Taq DNA polymerase. The thermal profiles were 94 C for 1 min, 55 C for 1 min, and 72 C for 1 min, with a final 10-min extension period. The resulting fragment, purified from agarose gel

Endo • 1998 Vol 139 • No 6

using the QIA Quick Gel Extraction Kit (Qiagen, Chatsworth, CA), was digested with EcoRI and BamHI inserted in pBluescript vector (Stratagene, La Jolla, CA) (19). Linearized DNA was transcribed using T7 polymerase [antisense complementary RNA (cRNA) probe] and T3 polymerase (sense cRNA probe; Riboprobe Combination System T3/T7, Promega Corp., Madison, WI) and labeled with [35S]UTP (Amersham; 10 mCi/ml). The radiolabeled cRNA probe was then purified by passing the transcription reaction solution over a G-50 column (Pharmacia Biotech), and fractions were collected and counted using a scintillation counter. The purified cRNA probes were heated at 80 C for 2 min with 500 mg/ml yeast transfer RNA and 50 mm DTT in water before being diluted to an activity of 5.0 3 107 dpm/ml with hybridization buffer containing 50% formamide, 0.25 m sodium chloride, 1 3 Denhardt’s solution, and 10% dextran sulfate. Sections with this hybridization solution (150 ml/slide) were incubated overnight at 50 C. After hybridization, the slides were washed four times (10 min each time) in 4 3 SSC (standard saline citrate) before ribonuclease (RNase) digestion (20 mg/ml for 30 min at 37 C) and rinsed at room temperature in decreasing concentrations of SSC that contained 1 mm DTT (2, 1, and 0.5 3; 10 min each) to a final stringency of 0.1 3 SSC at 65 C for 30 min (20). After dehydration in increasing alcohols, the sections were exposed to b-Max Hyperfilm (Amersham, Arlington Heights, IL) for 5 days before being dipped in Kodak NTB-2 liquid emulsion diluted 1:1 with distilled water. The dipped autoradiograms were developed 21 days later with Kodak D-19 developer (Eastman Kodak, Rochester, NY) and fixed, and the sections were counterstained through the emulsion with hematoxylin. Sections were examined under bright- and darkfield illumination. Several control experiments were carried out to test the specificity of the hybridization method and the DII probe. First, sections were incubated as described above with hybridization solution containing the sense strand probe synthesized by using T3 polymerase to transcribe the coding strand of the DNA insert. Second, the hybridization was also attempted on sections that had been pretreated with RNase (20 mg/ml for 30 min at 37 C) to degrade tissue RNA. Third, tissue sections were incubated in radiolabeled probe and then in an excess of unlabeled probe, which competed with the radiolabeled probe, eliminating the increased signal. Moreover, to assess the thermal stability of the hybrid, different series of sections were rinsed in 0.1 3 SSC at 75, 80, 85, 90, 94, and 98 C.

Assay of type II deiodinase DII activity was measured based on the release of radioiodide from the 125I2-labeled substrate. Using 100,000 cpm [59-125I]rT3 (SA, .750 mCi/mg; New England Nuclear, Boston, MA) for each sample, an incubation mixture containing 100 mg tissue homogenate in 0.1 m potassium phosphate buffer (pH 7.0), 1 mm EDTA, and 50 ml of substrate for a final concentration of 2 nm rT3 (Sigma Chemical Co., St. Louis, MO), 20 mm cofactor DTT, and 1 mm PTU, pH 7.0, was incubated in duplicate at 37 C for 1 h. Although T4 is a better substrate for 59DII, the physiological responses of the enzyme are equally reflected with either substrate (21, 22). The reactions were stopped by the addition of 50 ml ice-cold 5% BSA followed by 400 ml 10% ice-cold trichloroacetic acid and centrifuged at 4000 3 g for 20 min. The supernatant was further purified by cation exchange chromatography on 1.6 ml Dowex 50 W-X2 (100 –200 mesh; Sigma Chemical Co.). The iodide was then eluted twice with 1 ml 10% glacial acetic acid and counted in a g-counter. As a control, homogenate was substituted by buffer, and the amount of 125I2 produced was subtracted from the sample results. Enzymatic activity is expressed in femtomoles of I22 released per h/mg protein. For the determination of proteins, the bicinchonic acid assay (BCA Protein Assay, Pierce, Rockford, IL) was employed.

Hormone measurements Plasma T4, T3, and T4-binding capacity were measured by immunoassay in the Clinical Chemistry Laboratory of the Yale-New Haven Hospital.

FASTING UP-REGULATES HYPOTHALAMIC TYPE II DEIODINASE mRNA Data recording on Image 1 image analyzer and statistical analysis The density of the hybridization product was assessed in the different experimental groups. To digitally analyze, quantitate, and compare the amount of deiodinase type II mRNA, an Image-1/AT image processor (Universal Imaging Corp., West Chester, PA) using an Olympus IMT-2 inverted microscope with dark field optics (Olympus Corp., Lake Success, NY) and a Hamamatsu CCD camera (Hamamatsu Photonics, Hamamatsu, Japan) was employed. Six sections per animal were selected from the same area to assess the intensity of the hybridization product. The total surface covered by the hybridization product was assessed within a test region measuring 2 3 105 mm2 that contained the ME-ARC region and the PE. The threshold for measurement was assessed for each slide by determining the background labeling in the nearby ventromedial nucleus. After collecting the data, the means and ses were calculated. For each experiment, means were compared between experimental groups using one-way ANOVA with mean comparisons by the StudentNewman-Keuls method. A level of confidence of P , 0.05 was used to determine significant differences.

Results Hormone measurements

The circulating levels of T4 and T3, and the T4-binding capacity of animals in the six different groups are shown in Table 1. In situ hybridization histochemistry

The riboprobe generated to label deiodinase type II mRNA appeared to be specific, as silver grains were not found over cells after hybridization with the sense strand probe. In addition, no hybridization product was detected when sections were pretreated with RNase and then hybridized with the antisense probe, and reduction in the hybridization signal was found after incubation with an excess of unlabeled cRNA probe. The thermal stability of the hybrids was also assessed to test the specificity of the labeling. Increasing the temperature of the posthybridization washes, a reduction in the background labeling was observed. Only when the temperature of the final posthybridization wash exceeded 90 C was a decline in the density of the silver grains over labeled cells also observed. In situ hybridization histochemistry revealed a hybridization signal representing DII mRNA in the hypothalamus of the animals from all experimental groups. Within the hypothalamus, the most abundant labeling was present in the ARC-ME region and the PE (Fig. 1). No difference in the pattern of hybridization signal was observed depending upon the section. The periventricular labeling was localized to the ependymal layer of the third ventricle, where the hybridization was detected in the proximal one third of the ventricular wall, whereas no hybridization product was de-

2881

tected in the paraventricular nucleus and other hypothalamic nuclei adjacent to the third ventricle. Labeling was observed in the ependymal layer of the third ventricle only between the suprachiasmatic area and the mamillary bodies. No labeling was detected in the PE posterior to the retrochiasmatic area. In the ME, silver grains were concentrated either in the external layer, adjacent to the surface of the brain, or in the internal layer, adjacent to the wall of the third ventricle. Silver grains were also present in the ARC, particularly in its medial aspects. A quantitative analysis of the in situ hybridization experiments demonstrated differences between the amount of hybridization signal present in the PE and ME-ARC of animals from the different experimental groups (Fig. 2). A lower amount of the hybridization signal was detected in the euthyroid rats. In the fasted and fasted plus T4-treated rats, the amounts of hybridization signal were, respectively, 2.1 and 2.2 times higher (P , 0.05) than that in euthyroid animals. Although an increase in hybridization product was also detected in PTU-treated rats (1.47 times higher; P , 0.05), the amount of this increase was significantly lower than that in fasted and fasted plus T4-treated rats (P , 0.05). In fasted plus T4-treated animals, T4 injections did not affect DII mRNA levels, which were 2.2 times higher (P , 0.05) than the euthyroid values. No significant difference in the hybridization signal was found between fasted (2.1 times higher) and fasted plus T4-treated rats (2.2 times higher). On the other hand, in hyperthyroid rats, a decrease in the hybridization signal was detected compared with that in euthyroid animals, indicating an effect of T4 on DII mRNA under this condition. In addition, measurements of the hybridization signal in the PE and the ARC-ME independently showed different intensities of the hybridization product depending on the region examined. For example, the intensity of the PE of fasted rats was 4.29 times higher (P , 0.05) than that of the same area of the euthyroid animals, whereas the ME-ARC was only 1.61 times more intense compared with that of euthyroid rats (P , 0.05). In the fasted plus T4-treated animals, the PE and ARC-ME labeling were, respectively, 3.89 and 1.79 times greater than those in intact animals (P , 0.05). In the PTU-treated rats, the ARC-ME signal was 1.58 times significantly higher (P , 0.05) than the control value, whereas the PE signal was 1.36 times higher, but not statistically different (P . 0.05), compared with that in euthyroid rats. T4-treated rats compared with the euthyroid rats, showed no significant difference in the level of the DII mRNA in the PE (1.1 times higher), but, on the other hand, they showed a decrease of labeling in the ARC-ME regions

TABLE 1. Effects of PTU treatments (n 5 5), T4 injection (n 5 5), food deprivation (FD; n 5 5), and food deprivation plus T4 injection (n 5 5) on plasma levels of T4 (micrograms per dl), T3 (nanograms per dl), and TBC (micrograms per dl)

T4 T3 TBC

Euthyroid

3 weeks of PTUa

5dPTUa,b

T4a,b

FDa,b

FD 1 T4a,b

4.9 6 0.7 56 6 5.4 3.9 6 0.8

,1 13 6 2.0 9.1 6 1.1

2.2 6 0.3 27 6 2.9 4.5 6 0.5

22.7 6 1.3 105 6 8.7 ,1

2.8 6 0.6 30 6 3.5 5.1 6 0.4

14.2 6 0.9 107 6 9.1 1.1 6 0.3

T4, T3, and TBC were also measured in euthyroid rats (n 5 5). a P , 0.05 for all values compared with the euthyroid animals. b P , 0.05 compared with 3 weeks of PTU treatment.

2882


Endo • 1998 Vol 139 • No 6

FIG. 1. Dark- and brightfield micrographs demonstrating in situ hybridization products for DII mRNA in the PE, ARC, and ME. Dark and bright images are low and high power magnifications, respectively, of the medial hypothalamus of euthyroid (A–A3), PTU-treated (B–B3), T4-treated (C–C3), fasted (D–D3), and fasted plus T4-treated (E–E3) animals. Bar scales for dark- and brightfield images represent 200 and 100 mm, respectively.

(0.79 times lower; P , 0.05), indicating a T4 effect only in this latter area. DII enzymatic activity

DII enzymatic activity is shown in Fig. 3. DII activity was significantly elevated (P , 0.05) only in fasted (11.9 6 0.5 fmol/hzmg protein) and fasted plus T4 (12.1 6 1.1 fmol/hzmg protein) animals compared with that in euthyroid rats (7.2 6 1.02 fmol/hzmg protein). No significant changes were observed in the PTU-treated animals (6.9 6 1.3 fmol/hzmg protein) that had similar plasma thyroid hormone levels as fasted rats.

Discussion

The present results demonstrated that increases in the enzymatic activity and mRNA levels of DII occur during short term fasting. T4 replacement in fasted animals did not reverse the increase in DII activity and mRNA levels in the hypothalamus. Three weeks of treatment with PTU induced a modest increase in DII mRNA levels, whereas T4 administration to euthyroid animals suppressed DII mRNA levels only in the ME-ARC region. DII enzymatic activity in hypothalamic tissue fragments of 5-day treated PTU animals with plasma thyroid hormone levels comparable to those in fasted rats showed no significant difference compared with levels in euthyroid rats, and their


FIG. 2. A graph showing the results of a quantitative analysis of DII mRNA levels (n 5 5 for each experimental group) using one-way ANOVA. Results are expressed as the mean 6 SEM. a, Significantly different (P , 0.05) from euthyroid values; b, significantly different (P , 0.05) from hypothyroid values.

FIG. 3. A graph showing rT3 DII activity expressed in femtomoles per h/mg protein in euthyroid control (n 5 5), 5-day PTU-treated (n 5 5), T4-treated (n 5 5), fasted (n 5 5), and fasted plus T4-treated (n 5 5) rats. a, Significantly different (P , 0.05) compared with euthyroid and PTU animals.

values were significantly lower than those in fasted and fasted plus T4-treated rats. The iodothyronines are considered to be the most important regulators of DII activity. However, the action of thyroid hormones on DII seems to be site specific in the central nervous system. For example, thyroid hormones are known to induce a rapid suppression of DII activity in the pituitary and cerebral cortex of hypothyroid rats through a mechanism that involves both pre- (23) and posttranslational mechanisms (23–25). In the hypothalamus, hypothyroidism conditions are known to influence DII mRNA levels (6) and activity (5, 26). Interestingly, PTU and T4 had different effects on DII

2883

mRNA content in various parts of the hypothalamus; T4 and PTU did not significantly alter DII mRNA levels in the PE region of the hypothalamus, whereas, in accordance with a recent report (6), in the ME-ARC region, T4 and PTU suppressed and elevated DII mRNA content, respectively. To determine whether the ME-ARC and PE regions of the mediobasal hypothalamus may also exhibit differential changes in DII activity in response to PTU or T4 treatments, further biochemical studies are warranted using more sensitive sampling methods, such as the micropunch technique of Palkovits (27). Surprisingly, DII mRNA levels in the PE region that were unaffected by PTU and T4 treatments were robustly elevated by short term fasting with or without T4 replacement. A less impressive increase could also be detected in the ME-ARC region. Corresponding to these changes, activity measurements demonstrated elevated hypothalamic DII activity in fasted and fasted plus T4-treated animals compared with that in euthyroid control rats. To date, no previous reports are available regarding DII activity in short term fasted adult rats. However, a recent report by Pascual-Leone et al. (28) supports the idea that dietary restriction, at least in its initial phase, parallels elevations in hypothalamic DII activity. They reported DII activity changes in perinatal rats that are exposed to food restrictions. Although in that study a different regimen of food restriction was employed, an increase in fetal DII activity was detected after 5 days of the restricted diet administered to the dams (28). In comparing the differential effects of food restriction vs. thyroid hormone levels on hypothalamic DII mRNA, the following inferences can be made. Signals arising from food restriction predominantly affect DII-producing cells at the transcriptional or mRNA-processing level in the subependymal region of the PE, whereas thyroid hormones affect DII mRNA in cells of the ME-ARC region. Similar changes in thyroid hormone levels induced by fasting or PTU showed a significant difference in the enzymatic activity between the two groups of animals. Although fasting induced a statistically significant increase in DII activity in the hypothalamus compared with that in euthyroid controls, no significant changes were detected in the hypothalamus of 5-day PTU-treated animals. The differential regulation of DII activity and mRNA levels by fasting and thyroid hormones may help to explain why declining circulating thyroid hormone levels coincide with suppressed and elevated hypothalamic TRH production, the former being characteristic of fasting and the latter of primary hypothyroidism. Although circulating T4 levels are undetectable in hypothyroid animals, only a 50% decrease could be detected in fasted animals compared with euthyroid values. Therefore, we propose that during food deprivation, at least in its initial phase, the overproduction of DII induces high concentrations of T3 in the hypothalamus. In turn, this local hyperthyroid condition may lead to the suppression of TRH production and release. The present as well as previous reports (5, 6) indicate the lack or minimal activity of DII in the paraventricular nucleus where hypophysiotropic TRH neurons are located. Hence, it is likely that neurons of the ARC projecting to the paraventricular nucleus mediate the effect of T3 that is formed locally. The appearance of DII mRNA in the ependymal zone and the ME (6) together with our recent observation of glial fibrillary acidic protein in cells expressing

2884


DII mRNA (29) and a previous report (30) on the expression of DII in glial cells in neonatal rat brain strongly indicate that DII-producing cells are astrocytes and tanycytes. These glial cells provide an extensive network of cellular processes in the ARC (31–34) and suggest a paracrine action on PVN-projective ARC neurons via the production of thyroid hormones. In our previous study (29), we have shown that neurons in the ARC heavily project to neuroendocrine TRH cells in the PVN, and we recently found that a robust NPY innervation of TRH cells (35) originates from the ARC (36). On the other hand, the presence of DII mRNA in the external layer of the ME suggests that formation of T3 at this site may directly affect TSH production in the anterior pituitary and/or influence the release of TRH from the neuronal terminals around the portal vessels. In conclusion, our study indicates that the signal for elevated DII levels during fasting does not originate in the thyroid gland, and thus, suppressed circulating thyroid hormone levels under this condition are not the cause but, rather, are the result of suppressed TRH production and release. The mechanisms by which a differential regulation of DII mRNA occurs in fasting require further study. For example, consideration may be given to both the product of the ob gene, leptin, and glucocorticoids. It is known that during fasting, plasma leptin and corticosterone levels are low and elevated, respectively. Furthermore, systemic administration of leptin to fasted animals has been reported to reduce corticosterone levels (37) and increase TRH mRNA (38), which is also elevated during fasting in adrenalectomized rats treated with corticosterone to normal plasma levels (14). Thus, these observations raise the possibility that a mechanism involving interactions between leptin and glucocorticoids may be an important regulator of the hypothalamic-pituitary-thyroid axis.

14.

15. 16. 17. 18.

19. 20. 21. 22. 23. 24. 25. 26. 27. 28.

References 1. St Germain DL 1994 Iodothyronine deiodinases. Trends Endocrinol Metab 5:36 – 42 2. Chopra IJ 1977 A study of extrathyroidal conversion of thyroxine to 3,3959triiodothyronine in vitro. Endocrinology 101:453– 463 3. Green WL 1978 Metabolism of thyroid hormones by the rat thyroid tissue in vitro. Endocrinology 103:826 – 837 4. Kaplan MM 1986 Regulatory influences on iodothyronine deiodination in animal tissues. In: Hennemann (ed) Thyroid Hormone Metabolism. Marcel Dekker, New York, pp 231–253 5. Riskind PN, Kolodny JM, Larsen PR 1987 The regional hypothalamic distribution of type II 59-monodeiodinase in euthyroid and hypothyroid rats. Brain Res 420:194 –198 6. Tu HM, Kim SW, Salvatore D, Bartha T, Legradi G, Larsen PR, Lechan RM 1997 Regional distribution of type 2 thyroxine deiodinase messenger ribonucleic acid in rat hypothalamus and pituitary and its regulation by thyroid hormone. Endocrinology 138:3359 –3368 7. Visser TJ, Kaplan MM, Leonard JL, Larsen PR 1983 Evidence for two pathways of iodothyronine 59-deiodination in rat pituitary that differ in kinetics, propylthiouracil sensitivity, and response to hypothyroidism. J Clin Invest 71:992–1002 8. Silva JE, Larsen PR 1983 Adrenergic activation of triiodothyronine production in brown adipose tissue. Nature 305:712–713 9. Kaplan MM, Shaw EA 1984 Type II iodothyronine 59-deiodination by human and rat placenta in vitro. J Clin Endocrinol Metab 59:253–257 10. Calvo O, Obregon MJ, Ruiz de Ona C, Escobar del Rey F, Morreale de Escobar G 1990 Congenital hypothyroidism, as studied in rats. Crucial role of maternal thyroxine but not of 3:5,39-triiodothyronine in the protection of the fetal brain. J Clin Invest 86:889 – 899 11. Leonard JL, Kaplan MM, Visser TJ, Silva JE, Larsen PR 1981 Cerebral cortex responds rapidly to thyroid hormones. Science 214:571–573 12. Escobar-Morreale HF, Obregon MJ, Escobar del Rey F, Morreale de Escobar G 1995 Replacement therapy of hypothyroidism with thyroxine alone does not ensure euthyroidism in all tissues, as studied in thyroidectomized rats. J Clin Invest 96:2828 –2838 13. Rondeel JMM, de Greef WJ, Klootwijk W, Visser TJ 1992 Effects of hypo-

29.

30. 31.

32. 33. 34.

35. 36.

37. 38.

Endo • 1998 Vol 139 • No 6

thyroidism on hypothalamic release of thyrotropin-releasing hormone in rats. Endocrinology 130:651– 656 Van Haasteren GAC, Linkels E, Lootwijk W, van Toor H, Rondeel JMM, Themmen APN, de Jong FH, Valentijn K, Vaudry H, Bauer K, Visser TJ, de Greef WJ 1995 Starvation-induced changes in the hypothalamic content of prothyrotrophin releasing hormone (pro-TRH) mRNA and the hypothalamic release of proTRH-derived peptides: role of the adrenal gland. J Endocrinol 145:143–153 Walker JS, Levy G 1988 Kinetics of drug action in disease states. XXXIV. Effect of experimental thyroid disorders on the pharmacodynamics of phenobarbital, ethanol and pentylenetetrazol. J Pharmacol Exp Ther 249:6 –10 Walker JS, Levy G 1989 Induction of experimental thyroid dysfunction in rats with implantable pellets of thyroxine or propylthiouracil. J Pharmacol Methods 21:223–229 Ling E, O’Brien P, Salerno T, Iannuzzo CD 1988 Effects of different thyroid treatments on the biochemical characteristics of rabbit myocardium. Can J Cardiol 4:301–306 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 tissues. J Clin Invest 98:405– 417 Sambrook J, Fritsch EF, Maniatis T 1989 Molecular Cloning. Cold Spring Harbor Laboratory, Cold Spring Harbor Simerly RB, Chang C, Muramatsu M, Swanson LW 1990 Distribution of androgen and estrogen receptor mRNA-containing cells in the rat brain: an in situ hybridization study. J Comp Neurol 294:76 –95 Leonard JL, Mellen SA, Larsen PR 1982 Thyroxine 59-deiodinase activity in brown adipose tissue. Endocrinology 112:1153–1155 Silva JE, Larsen PR 1985 Potential of brown adipose tissue type II thyroxine 59-deiodinase as local and systemic source of triiodothyronine in rats. J Clin Invest 76:2296 –2305 Burmeister LA, Pachucki J, St. Germain DL 1997 Thyroid hormones inhibit type 2 iodothyronine deiodinase in the rat cerebral cortex by both pre- and posttranslational mechanisms. Endocrinology 138:5231–5237 Leonard JL, Silva JE, Kaplan MM, Mellen S, Visser TJ, Larsen PR 1984 Acute posttranscriptional regulation of cerebrortical and pituitary iodothyronine 59-deiodinase by thyroid hormone. Endocrinology 114:998 –1004 Silva JE, Leonard JL 1985 Regulation of rat cerebrocortical and adenohypophyseal type II 59-deiodinase by thyroxine, triiodothyronine, and reverse triiodothyronine. Endocrinology 116:1627–1635 Serrano-Lozano A, Montiel M, Morell M, Morata P 1993 59 deiodinase activity in brain regions of adult rats: modifications in different situations of experimental hypothyroidism. Brain Res Bull 30:611– 616 Palkovits M 1980 Guide and Map for the Isolated Removal of Individual Cell Groups from the Rat Brain. Akademia Kiado, Budapest Pascual-Leone AM, Alaez C, Calvo R, Martin MA, Obregon MJ 1994 Effects of thyroid hormone deiodination on regulation of thyroid axis in undernourished rats. Am J Physiol 267:E983–E989 Diano S, Naftolin F, Goglia F, Csernus V, Horvath TL, Monosynaptic pathway between the arcuate nucleus expressing glial type II iodothyronine 59deiodinase mRNA and the median eminence-projective TRH cells of the rat paraventricular nucleus. J Neuroendocrinol, in press Guadano-Ferraz A, Obregon MJ, St Gremain DL, Bernal J 1997 The type 2 iodothyronine deiodinase is expressed primarily in glial cells in the neonatal rat brain. Proc Natl Acad Sci USA 94:10391–10396 Blier R 1972 Structural relationship of ependymal cells and their processes within the hypothalamus. In Knigge KM, Scott DE, Weindl A (eds) Brain Endocrine Interaction. Median Eminence: Structure and Function. Karger, Basel, pp 306 –318 Bruni JE 1974 Scanning and transmission electron microscopy of the ependymal lining of the third ventricle. J Can Sci Neurol 1:59 –73 Rodriguez EM, Gonzalez CB, Delannoy L 1979 Cellular organization of the lateral and postinfundibular regions of the median eminence in the rat. Cell Tissue Res 201:377– 408 Akmayer IF, Fidelina OV 1976 Morphological aspects of the hypothalamichypophyseal system. VI. The tanycytes: their relation to the sexual differentiation of the hypothalamus. An enzyme-histochemical study. Cell Tissue Res 173:407– 416 Toni R, Jackson IMD, Lechan RM 1990 Neuropeptide Y-immunoreactive innervation of thyrotropin-releasing hormone-synthesizing neurons in the rat paraventricular nucleus. Endocrinology 126:2444 –2453 Diano S, Goglia F, Naftolin F, Horvath TL 1997 GABAergic NPY input of the parvicellular TRH cells: its origin in the arcuate nucleus and regulation by thyroid hormones. 4th International Meeting on NPY, London, UK. Regul Pept 71:217 (Abstract OC34) Ahima RS, Prabakaran D, Mantzoros C, Qu D, Lowell B, Maratos-Flier E, Flier JS 1996 Role of leptin in the neuroendocrine response to fasting. Nature 382:250 –252 Legradi G, Emerson CH, Ahima RS, Flier JS, Lechan RM 1997 Leptin prevents fasting-induced suppression of prothyrotropin-releasing hormone messenger ribonucleic acid in neurons of the hypothalamic paraventricular nucleus. Endocrinology 138:2569 –2576