Differential Effects of Central Leptin, Insulin, or Glucose Administration during Fasting on the Hypothalamic-Pituitary-Thyroid Ax
http://www.100md.com
《内分泌学杂志》
Tupper Research Institute and Department of Medicine (C.F., P.S.S., E.S., S.S., R.M.L.), Division of Endocrinology, Diabetes, Metabolism
Molecular Medicine, New England Medical Center, Boston, Massachusetts 02111
Department of Endocrine Neurobiology (C.F.), Institute of Experimental Medicine, Hungarian Academy of Sciences, 1083 Budapest, Hungary
Division of Endocrinology, Diabetes
Hypertension (M.A.C., A.C.B.), Brigham and Women’s Hospital, Boston, Massachusetts 02115
Division of Endocrinology (R.S.R.), Universidade Federal de Sao Paulo, 04039-032 Sao Paulo, Brazil
Departments of Community Health (W.M.R.) and Neuroscience (R.M.L.), Tufts University School of Medicine, Boston, Massachusetts 02111
Division of Endocrinology (C.H.E.), University of Massachusetts Medical School, Worcester, Massachusetts 01655
Abstract
The reductions in circulating levels of leptin, insulin, and glucose with fasting serve as important homeostasis signals to neurons of the hypothalamic arcuate nucleus that synthesize neuropeptide Y (NPY)/agouti-related protein (AGRP) and -MSH/cocaine and amphetamine-regulated transcript. Because the central administration of leptin is capable of preventing the inhibitory effects of fasting on TRH mRNA in hypophysiotropic neurons primarily through effects on the arcuate nucleus, we determined whether the continuous administration of 30 mU/d insulin or 648 μg/d glucose into the cerebrospinal fluid by osmotic minipump might also have similar effects on the hypothalamic-pituitary-thyroid axis. As anticipated, the intracerebroventricular infusion of leptin reduced fasting-induced elevations in NPY and AGRP mRNA and increased proopiomelanocortin and cocaine and amphetamine-regulated transcript mRNA in the arcuate nucleus. In addition, leptin prevented fasting-induced reduction in pro-TRH mRNA levels in the paraventricular nucleus and in circulating thyroid hormone levels. In contrast, whereas insulin increased proopiomelanocortin mRNA and both insulin and glucose reduced NPY mRNA in arcuate nucleus neurons, neither prevented the fasting-induced suppression in hypophysiotropic TRH mRNA or circulating thyroid hormone levels. We conclude that insulin and glucose only partially replicate the central effects of leptin and may not be essential components of the hypothalamic-pituitary-thyroid regulatory system during fasting.
Introduction
FASTING RESULTS IN a number of adaptive responses to decrease energy expenditure. Included is the development of central hypothyroidism, characterized by decreased synthesis of TRH in the hypothalamic paraventricular nucleus (PVN) and low TSH and thyroid hormone levels in the bloodstream (1, 2, 3). Because thyroid hormone stimulates mitochondrial oxygen consumption and increases thermogenesis (4, 5), the reduction in circulating levels of thyroid hormone during fasting is presumed to be an important mechanism to conserve energy.
Data by Ahima et al. (6) and from our laboratories (1) have demonstrated that the amount of leptin circulating in the bloodstream serves as an important regulatory signal to the brain that modulates the responsiveness of the hypothalamic-pituitary-thyroid (HPT) axis to circulating levels of thyroid hormone. Thus, fasting-induced inhibition of the HPT axis can be completely reversed if circulating levels of leptin are increased by exogenous administration, despite continuation of the fast (1, 6).
The effect of leptin on the HPT axis is mediated primarily through monosynaptic connections to TRH neurons in the PVN from two leptin-sensitive neuronal populations in the arcuate nucleus (7) that have opposing effects on TRH biosynthesis. These include neuropeptide Y (NPY)/agouti-related protein (AGRP)-synthesizing neurons that are inhibitory to hypophysiotropic TRH neurons (8, 9, 10), and -MSH/cocaine and amphetamine-regulated transcript (CART)-synthesizing neurons that are stimulatory (11, 12). In addition to changes in circulating leptin levels, however, fasting also reduces circulating levels of insulin and glucose (13, 14) that may also serve as regulatory signals to the central nervous system. Like leptin, the central administration of insulin or glucose inhibits food intake (15, 16, 17) and has similar effects on the expression of proopiomelanocortin (POMC) and NPY mRNA in arcuate nucleus neurons (18, 19, 20, 21, 22, 23). We hypothesized, therefore, that insulin and/or glucose may also contribute to the regulation of the HPT axis during fasting through actions on the arcuate nucleus. Accordingly, we compared the effects of the central administration of leptin on the HPT axis during fasting to that of insulin and glucose.
Materials and Methods
The experiments were carried out on adult male Sprague-Dawley rats, weighing 200–250 g. Animals were housed under standard environmental conditions (light from 0600–1800 h; temperature, 22 ± 1 C; rat chow and water ad libitum). All experimental protocols were reviewed and approved by the Institutional Animal Care and Use Committee of the Tufts-New England Medical Center.
Adult rats (n = 36) were implanted with a 22-gauge stainless steel guide cannula (Plastics One, Inc., Roanoke, VA) into the lateral cerebral ventricle under stereotaxic control (coordinates from bregma: anterior-posterior, –0.8; lateral, 1.2; dorsal-ventral, 3.5) through a burr hole in the skull. The cannula was secured to the skull with three stainless steel screws and dental cement and was temporarily occluded with a dummy cannula. Bacitracin ointment was applied to the interface of the cement and the skin after surgery. One week after intracerebroventricular cannulation, an osmotic minipump (Alzet model 1003D; Alza Corp., Palo Alto, CA) was implanted under sodium pentobarbital anesthesia (35 mg/kg body weight ip) intradermally between the scapulas and connected with durameter vinyl tubing (Scientific Comodities Inc., Lake Havasu City, AZ) to a 28-gauge needle that was permanently inserted into and extended 1 mm below the external guide cannula. The animals were divided into five groups. The first group (group 1) had free access to food, whereas in the remaining groups (groups 2–5), food was withdrawn over the 3 d of experimentation. Osmotic minipumps delivered either artificial cerebrospinal fluid (aCSF) (140 mM NaCl, 3.35 mM KCl, 1.15 mM MgCl2, 1.26 mM CaCl2, 1.2 mM Na2HPO4, 0.3 mM NaH2PO4, 0.1% BSA, pH 7.4) (groups 1 and 2), 10 μg/24 h mouse leptin (Lilly Pharmaceutical Co, Indianapolis, IN) in aCSF (group 3), 30 mU porcine insulin (Lilly) in aCSF (group 4), or 648 μg glucose in aCSF (group 5) for 3 d at a rate of 1 μl/h.
At the completion of the experiment, the animals were anesthetized with sodium pentobarbital between 0900 and 1200 h, CSF was taken from the cisterna magna for mouse leptin, porcine insulin, and glucose measurements; brown fat was dissected from the interscapular region; blood was taken from the inferior vena cava for measurement of serum TSH, thyroid hormone, and glucose levels; and the animals were immediately perfused with fixative as described below. Blood was collected into polypropylene tubes and centrifuged for 15 min at 4000 rpm, and the plasma was stored at –80 C until assayed. The brown fat was snap-frozen in dry ice and stored at –80 C until processed for type 2 iodothyronine deiodinase (D2) enzymatic activity and uncoupling protein-1 (UCP1) mRNA.
Tissue preparation for in situ hybridization histochemistry
Under sodium pentobarbital anesthesia, the animals were perfused transcardially with 20 ml 0.01 M PBS, pH 7.4, containing 15,000 U/liter heparin sulfate, followed by 150 ml 4% paraformaldehyde in PBS. The brains were removed and postfixed by immersion in the same fixative for 2 h at room temperature. Tissue blocks containing the hypothalamus were cryoprotected in 20% sucrose in PBS at 4 C overnight and then frozen on dry ice. Serial 18-μm-thick coronal sections through the rostrocaudal extent of the PVN and the arcuate nucleus were cut on a cryostat (Leica CM3050 S; Leica Microsystems, Nussloch GmbH, Germany) and adhered to SuperFrost Plus glass slides (Fisher Scientific, Pittsburgh, PA) to obtain four sets of slides, each set containing every fourth section through the PVN and the arcuate nucleus. Cannula placement was confirmed by light microscopic examination. The tissue sections were desiccated overnight at 42 C and were stored at –80 C until prepared for in situ hybridization histochemistry.
In situ hybridization histochemistry
Every fourth section of the PVN was hybridized with a 1241-bp single-stranded [35S]UTP-labeled cRNA probe for pro-TRH, and every fourth section through the arcuate nucleus was hybridized with single-stranded [35S]UTP-labeled cRNA probe for NPY (24), AGRP (25), POMC (26), and CART (27) as previously described (10). Hybridizations were performed under plastic coverslips in a buffer containing 50% formamide, a 2-fold concentration of standard sodium citrate, 10% dextran sulfate, 0.5% sodium dodecyl sulfate, 250 μg/ml denatured salmon sperm DNA, and 6 x 105 cpm radiolabeled probe for 16 h at 56 C. Slides were dipped into Kodak NTB2 autoradiography emulsion (Eastman Kodak Co., Rochester, NY), and the autoradiograms were developed after 2–4 d of exposure at 4 C.
Image analysis
Autoradiograms were visualized under dark-field illumination using a COHU 4910 video camera (COHU, Inc., San Diego, CA). The images were analyzed with a Macintosh G4 computer (Apple Computers, Cupertino, CA) using Scion Image (Scion Corp., Frederick, MD). Background density points were removed by thresholding the image, and integrated density values (density x area) of hybridized neurons in the same region of each side of the PVN or arcuate nucleus were measured in five to six consecutive sections for each animal depending upon the radiolabeled probe (TRH = 6, NPY = 5, AGRP = 5, CART = 5, and POMC = 5). Nonlinearity of radioactivity in the emulsion was evaluated by comparing density values with a calibration curve created from autoradiograms of known dilutions of the radiolabeled probes immobilized on glass slides in 1.5% gelatin fixed with 4% formaldehyde and exposed and developed simultaneously with the in situ hybridization autoradiograms.
For mRNAs shown to be affected by insulin or glucose above, a more detailed analysis was performed using the approach described previously. However, integrated density values were measured for each hybridized neuron in the arcuate nucleus sections at the same level of the rostral caudal extent of the median eminence and expressed as a histogram, reflecting the number of arcuate nucleus neurons with integrated density values above background ranging from 0.1–110 density units at intervals of 5 density units.
Hormone and glucose measurements
Serum T3 and T4 were measured by the Diagnostic Products Corp. (Los Angeles, CA) TKT41 and TKT31 assay systems, respectively. The intraassay coefficients of variation for these assays were 3.1 and 2.8%, respectively. The interassay coefficients of variation for these assays were 5.7 and 5.9%, respectively. Serum TSH was measured by the rat TSH assay kit RPA 554 obtained from GE Healthcare (formerly Amersham Bioscience, Piscataway, NJ). The intraassay coefficient of variation for this assay was 4.8%, and its interassay coefficient of variation was 8.5%. Leptin was measured by the Linco Research (St. Charles, MO) ML-82K mouse leptin assay. Its intraassay coefficient of variation was 3.3%, and its interassay coefficient of variation was 4.9%. Insulin was measured by the Linco PI-12K porcine insulin assay. The blood glucose levels were measured by glucometer (Medisense Precision Xtra; Abbott, Bedford, MA). CSF glucose levels were measured by the Freestyle Blood Glucose Monitoring System (Therasense, Alameda, CA).
Brown adipose tissue (BAT) measurements
D2 activity was measured as previously described (28). Approximately 250 μg total BAT lysate protein was incubated for 2 h in the presence of 1 nM [125I]5'T4, 20 mM dithiothreitol, and 1 mM propylthiouracil. Specific T4 to T3 conversion was calculated by subtracting nonspecific deiodination in tubes containing the same amount of lysate protein obtained from human embryonic kidney cells. The background activity of these samples was less than 2%. Deiodinase activity was expressed as femtomoles of T4 deiodinated per minute per milligram of protein. Total RNA was extracted using TRIzol Reagent (Invitrogen, Carlsbad, CA) and used to synthesize cDNA using SuperScript First-Strand Synthesis System for RT-PCR (Invitrogen). The generated cDNAs were used in a real-time PCR using the QuantiTect SYBR Green PCR kit in I-Cycler (Bio-Rad, Hercules, CA). Standard curves (five-point serial dilution of mixed experimental and control groups cDNA) were analyzed in each assay and used as calibrators to the relative quantification of product generated in the exponential phase of amplification curve. The r2 value was below 0.99 for all standard curves, and the amplification efficiency varied between 90 and 100%. Sample quantification was calculated by the standard curve and corrected by the internal control -actin mRNA in all experiments. Specific oligonucleotides for rat UCP1 and -actin were designed using Beacon Designer 2.06 (Premier Biosoft International, Palo Alto, CA).
Statistical analyses
Results are presented as mean ± SEM. One-way ANOVA was used to examine the effect of group on each variable of interest, followed by the Newman-Keuls post hoc test to determine whether the ANOVA showed a significant (P < 0.05) group effect. Because of skewness of the CART data, the natural log of the variables was used to test significance. To explore whether a combination of T3, T4, and TRH would be able to further separate the groups, discriminant function analysis was used, with group differences being evaluated with Hotelling’s t2 statistic. Density values of individual neurons in each experimental group for NPY and POMC were examined for biomodality by testing the kurtosis values for each animal after correcting for skewness by taking the natural log of the NPY and POMC density values. Analyses were conducted using the SPSS statistical program version 11.5 (SPSS Inc., Chicago, IL).
Results
Effect of central leptin, insulin, and glucose administration on body weight and serum and CSF hormone parameters in fasted animals
Body weight change and serum and CSF hormone levels are shown in Table 1. Fasting resulted in an approximately 15% weight loss during the 3 d of experimentation. Central leptin and insulin administration had no influence on weight loss of the animals, and although glucose appeared to diminish the effects of fasting on weight loss, the difference was not significant.
Murine leptin and porcine insulin levels were elevated in the CSF only in the leptin- and insulin-treated groups, respectively. Glucose levels tended to decrease in the CSF of fasted animals, and a further decrease was observed in the leptin-treated fasted animals. The glucose level in the CSF of glucose-treated fasted animals was significantly increased compared with fasted rats receiving aCSF but was not different when compared with the glucose levels of fed rats. Serum glucose levels of fasted animals were significantly decreased compared with fed animals, and the central administration of leptin caused an even further decline. However, the central administration of neither insulin nor glucose had any effect on peripheral glucose levels.
As expected, fasting resulted in a fall in serum T4 and T3 levels. Central leptin treatment completely reversed the effects of fasting on serum T3 levels and attenuated the effects of fasting on serum T4 levels. The central administration of either insulin or glucose, however, had no significant effects on thyroid hormone levels. Serum TSH levels were not significantly changed by any of the treatments. Leptin administration significantly increased the T3/T4 ratio in the peripheral blood, whereas fasting and the other treatments had no effect.
Effect of central leptin, insulin, and glucose administration on BAT UCP1 mRNA levels and D2 activity in fasted animals
Fasting significantly decreased UCP1 mRNA levels in BAT (Table 1). Leptin administration to fasting animals, however, increased UCP1 mRNA to levels even above fed controls, and insulin treatment prevented the fasting-induced inhibition of the UCP1 mRNA. Glucose had no significant effect. In contrast, only leptin treatment altered D2 activity. Fasting, insulin, and glucose treatment had no effect, but central leptin administration resulted in an approximately 2-fold increase in D2 activity.
Effect of central leptin, insulin, or glucose administration on gene expression of feeding-related peptides in the arcuate nucleus
In agreement with the literature (18), fasting increased the NPY and AGRP expression and decreased the POMC and CART mRNA levels in the arcuate nucleus (Figs. 1–4). All groups showed significant differences (P < 0.01) by ANOVA. Leptin, insulin, and glucose exerted similar effects on NPY-producing arcuate nucleus neurons in fasted rats (Fig. 1). All three substances significantly decreased NPY mRNA to the level observed in fed animals, despite continuation of the fast (Fig. 1). In contrast, leptin prevented the fasting-induced increase in AGRP mRNA levels, but insulin and glucose had no effect (Fig. 2). In addition, whereas both leptin and insulin significantly reversed the fasting-induced fall of the POMC mRNA levels, glucose had no effect (Fig. 3). Leptin resulted in a significant increase in CART gene expression in the arcuate nucleus compared with fasting animals, but the central administration of insulin or glucose had no effect (Fig. 4).
Regional differences were not observed in the response of the arcuate nucleus neurons to any of the treatments. Analysis of density values for individual arcuate nucleus neurons expressing the two mRNAs affected by leptin and insulin or glucose revealed that compared with fasting animals, leptin, insulin, and glucose all resulted in a significant and marked shift of the density values for NPY mRNA in fasting animals to the left (Fig. 5A), whereas leptin and insulin-treated fasting animals showed a shift of the density values for POMC mRNA to the right (Fig. 5B). Testing for bimodality of the density values for the individual cells examined was performed by computing kurtosis values. For none of the animals in either the NPY or POMC groups could the hypothesis of normality be rejected.
Effect of central leptin, insulin, or glucose administration on TRH mRNA in hypophysiotropic neurons in the PVN
In fed animals, neurons containing pro-TRH mRNA were readily visualized by in situ hybridization histochemistry, symmetrically distributed in the medial and periventricular parvocellular subdivisions of the PVN on either side of the third ventricle (Fig. 6A). As previously recognized, fasting caused a marked reduction in the hybridization signal (Fig. 6B), whereas leptin administration completely prevented the fasting-induced inhibition of TRH mRNA (Fig. 6C). In contrast, no significant effect on the density of silver grains over pro-TRH neurons in the PVN in fasting animals was observed after insulin administration (Fig. 6D). Central glucose administration to fasted animals had a tendency to further decrease TRH mRNA levels below that observed in the fasted animals receiving aCSF (Fig. 6E), and they were significantly different when examined by t test (P = 0.028). However, the values were not significantly different when subjected to a more rigorous statistical analysis using ANOVA and Newman-Keuls testing or when the effect of glucose on the HPT axis was assessed by discriminant function analysis using combinations of T3, T4, and TRH mRNA. The results of the image analyses are shown in Fig. 6F.
Discussion
Two main theories have been proposed as to how the brain senses the status of the peripheral energy stores (29). According the lipostatic theory, the brain monitors the storage and metabolism of fat, whereas the glucostatic theory is based on the hypothesis that the brain senses the storage and use of carbohydrates (29). It is now recognized that the main adiposity signal to the brain is leptin (18), with circulating leptin levels directly proportional to the total amount of stored fat (30). The glucostatic signal is glucose itself sensed by specific populations of glucose-sensing neurons in the central nervous system (31, 32). Insulin also signals to the brain, and its peripheral level is dependent on both the amount of adipose tissue and circulating glucose levels (29). Interestingly, neurons in a discrete location in the hypothalamus, the arcuate nucleus, are a major central target for all three signaling substances, and in some instances the same arcuate nucleus neurons express leptin and insulin receptors and are glucose sensitive (33). This is the case for NPY-producing arcuate nucleus neurons that are inhibited by all three substances and -MSH-producing neurons that are activated by all three substances (18, 19, 20, 21, 22, 23). Because NPY and -MSH have critical roles in regulation of the HPT axis (7), we determined whether in addition to leptin, insulin and glucose might also affect hypophysiotropic TRH gene expression and circulating levels of TSH and thyroid hormones.
Similar to previous observations (1, 34), the central administration of leptin to fasting animals prevented fasting-induced inhibition of the TRH mRNA in hypophysiotropic neurons. Although a direct effect of leptin on TRH neurons residing in the PVN has been proposed (35, 36, 37), evidence that ablation of the arcuate nucleus prevents alterations in TRH gene expression during fasting and after leptin administration to fasting animals (38) suggests that the major effect of leptin on the hypophysiotropic TRH neurons may be mediated indirectly through arcuate nucleus neurons (38). This is further supported by the observation of Perello et al. (39) showing that the stimulatory effect of central leptin administration on TRH gene expression in the PVN can be completely reversed by administration of a melanocortin receptor antagonist. Two main neuronal populations of arcuate nucleus neurons are involved in mediating the effects of leptin on the central nervous system, including NPY neurons that cosynthesize AGRP and -MSH neurons that cosynthesize CART (18). These two antagonistic neuron populations innervate hypophysiotropic TRH neurons (11, 12, 40, 41, 42) and have opposing effects in the regulation of TRH mRNA (8, 10, 11, 12).
As observed in the current study, leptin inhibited AGRP and NPY mRNA synthesis and increased the gene expression of both stimulatory peptides, CART and -MSH, completely reversing the effect of fasting on these two neuronal populations. These findings are in agreement with other investigators showing that in fasting animals, leptin has inhibitory effects on the gene expression of AGRP and NPY and stimulatory effects on CART and -MSH mRNA in arcute nucleus neurons (43, 44, 45, 46). Leptin also reversed fasting-induced effects on circulating levels of thyroid hormones, increasing circulating T4 levels toward normal and even increasing circulating T3 levels above normal. We presume that the explanation for the inability of leptin to fully restore T4 levels to normal is a result of fasting-induced suppression of the T4-binding protein, transthyretin, which is not regulated by leptin administration (1, 47). The rise in T3 may be secondary to an increase in the conversion of T4 to T3 in BAT as a result of stimulation of D2 by leptin. This explanation may also account for the increase in the T3/T4 ratio in the leptin-treated fasting animals compared with the fed controls. Central leptin administration also influenced the UCP1 mRNA level in BAT, in support of the ability for leptin to centrally induce mitochondrial heat production in this tissue (48). Similar observations regarding the effects of central leptin administration on the T3/T4 ratio, BAT D2 activity, and UCP1 mRNA levels were reported by Cettour-Rose et al. (49).
Insulin is thought to have similar roles in the regulation of energy homeostasis as leptin. Both hormones decline in association with fasting, exert anorexic effects, and increase energy expenditure (29). In addition, both hormones have similar effects on NPY- and -MSH-producing neurons in the arcuate nucleus (20, 23, 43). It seemed reasonable to assume, therefore, that insulin might also have a similar role in the regulation of the HPT axis as leptin. Indeed, streptozotocin-induced diabetes reduces the hypothalamic secretion of TRH and circulating thyroid hormone levels (50). In contrast to leptin, however, central administration of insulin had no significant effects on any parameters of the HPT axis. TRH mRNA, TSH, and thyroid hormone levels were not significantly different from levels in fasting rats receiving aCSF. Thus, we presume that in rats, insulin may not contribute in an important way to the development of central hypothyroidism during fasting or recovery of thyroid function with refeeding. Because a daily intracerebroventricular dose of 8 mU porcine insulin has been reported sufficient to exert an inhibitory effect on food intake and a stimulatory effect on POMC gene expression (23), it seems unlikely that the absence of a response of the HPT axis to insulin in fasting animals can be attributed to the administration of an insufficient amount of insulin. This is further supported by the observation that insulin prevented the fasting-induced increase in NPY mRNA and increased POMC gene expression, as observed by other groups (20, 22, 23). In addition, insulin treatment also prevented the fasting-induced inhibition of UCP1 mRNA levels in the BAT, presumably mediated through central activation of the sympathetic nervous system (51). Thus, despite the overlapping second messenger pathways for leptin and insulin (52), insulin would appear to only partially recapitulate the central effects of leptin, a concept consistent with the observation that the central insulin receptor-deficient mouse has a milder phenotype than the leptin receptor-deficient ob/ob mouse (53). Nevertheless, the possibility that insulin could potentiate the effects of leptin on the HPT axis has not been excluded by these studies.
Based on our earlier data that exogenous -MSH administration prevents the fasting-induced inhibition of the pro-TRH mRNA levels in hypophysiotropic neurons and NPY inhibits TRH synthesis in the same neuronal population (10, 11), one might assume that insulin-induced changes in NPY and POMC gene expression in arcuate nucleus neurons should have been sufficient to increase pro-TRH gene expression. The possibility that only a subpopulation of NPY- or POMC-producing neurons were responsive to insulin, leaving a nidus of unresponsive cells that retain the activating or inhibitory effects of fasting, respectively, and project to TRH neurons in the PVN, is made unlikely by the absence of any apparent qualitative regional difference in the arcuate nucleus in each experimental group or quantitative difference by statistical analysis of individual neurons in the arcuate nucleus for the presence of bimodality in response to leptin, insulin, or glucose. Contrary to leptin administration in fasting animals, which prevents the rise in AGRP and increases CART gene expression, however, insulin administration had no effect on either mRNA. It may be presumed, therefore, that the persistent increase in AGRP gene expression simultaneously with suppression in CART gene expression was enough to overcome the stimulatory effects of increased -MSH and reduced NPY on hypophysiotropic TRH neurons. This hypothesis is supported by the finding that when AGRP is administered exogenously to fed animals, TRH mRNA is inhibited in hypophysiotropic neurons despite persistent elevations in POMC gene expression (8). Alternatively, other factors may be involved in the regulation of the HPT axis during fasting that have not yet been characterized.
The glucose-sensitive neurons of the brain also have a pivotal role in the regulation of energy homeostasis. Lesion of these neurons by gold-thio-glucose results in an obese phenotype (54). When administered directly into the central nervous system, glucose has been reported to activate the electrophysiological properties of POMC neurons (19), reduce food intake (15), and block counterregulation during systemic hypoglycemia (55). In addition, central glucopenia induced by 2-deoxy-D-glucose administration increases food intake (54) and induces a peripheral counterregulatory response including hepatic glucose production and an increase in plasma epinephrine, norepinephrine, insulin, glucagon, and cortisol levels (56).
In our study, similar to insulin treatment, glucose administration to fasted animals had no stimulatory effects on the HPT axis, making it unlikely that the fall in glucose levels during fasting is responsible for inhibition of the HPT axis. In fact, a tendency for decreased pro-TRH gene expression was observed in the hypophysiotropic neurons of the glucose-treated rats even below that of fasting rats receiving aCSF but was significant only when subjected to statistical analysis using the Student’s t test. However, glucose did have a significant effect on NPY gene expression in arcuate nucleus neurons, preventing the fasting-induced increase in NPY mRNA. Additional studies are required to clarify whether glucose sensing by hypothalamic neurons are indeed coupled to suppression of hypophysiotropic TRH.
We conclude that insulin and glucose only partially recapitulate the effects of leptin on the hypothalamus, having selective effects on the gene expression of POMC and/or NPY in arcuate nucleus neurons but not AGRP and CART. In contrast to leptin, changes in insulin and glucose do not appear to have an important role in fasting-induced central hypothyroidism.
Footnotes
This work was supported by Grants NIH DK-37021, OTKA T046492, and Sixth European Union Program (LSHM-CT-2003-503041).
First Published Online October 6, 2005
Abbreviations: aCSF, Artificial cerebrospinal fluid; AGRP, agouti-related protein; BAT, brown adipose tissue; CART, cocaine and amphetamine-regulated transcript; D2, type 2 iodothyronine deiodinase; HPT, hypothalamic-pituitary-thyroid; NPY, neuropeptide Y; POMC, proopiomelanocortin; PVN, paraventricular nucleus; UCP1, uncoupling protein-1.
Accepted for publication September 23, 2005.
References
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
van Haasteren GA, Linkels E, Klootwijk W, van Toor H, Rondeel JM, Themmen AP, 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 (proTRH) mRNA and the hypothalamic release of proTRH-derived peptides: role of the adrenal gland. J Endocrinol 145:143–153
Rondeel JM, Heide R, de Greef WJ, van Toor H, van Haasteren GA, Klootwijk W, Visser TJ 1992 Effect of starvation and subsequent refeeding on thyroid function and release of hypothalamic thyrotropin-releasing hormone. Neuroendocrinology 56:348–353
de Jesus LA, Carvalho SD, Ribeiro MO, Schneider M, Kim SW, Harney JW, 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
Lanni A, Moreno M, Lombardi A, de Lange P, Goglia F 2001 Control of energy metabolism by iodothyronines. J Endocrinol Invest 24:897–913
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
Lechan RM, Fekete C 2004 Feedback regulation of thyrotropin-releasing hormone (TRH): mechanisms for the nonthyroidal illness syndrome. J Endocrinol Invest 27:105–119
Fekete C, Sarkar S, Rand WM, Harney JW, Emerson CH, Bianco AC, Lechan RM 2002 Agouti-related protein (AGRP) has a central inhibitory action on the hypothalamic-pituitary-thyroid (HPT) axis: comparisons between the effect of AGRP and neuropeptide Y on energy homeostasis and the HPT axis. Endocrinology 143:3846–3853
Fekete C, Marks DL, Sarkar S, Emerson CH, Rand WM, Cone RD, Lechan RM 2004 Effect of Agouti-related protein in regulation of the hypothalamic-pituitary-thyroid axis in the melanocortin 4 receptor knockout mouse. Endocrinology 145:4816–4821
Fekete C, Kelly J, Mihaly E, Sarkar S, Rand WM, Legradi G, Emerson CH, Lechan RM 2001 Neuropeptide Y has a central inhibitory action on the hypothalamic-pituitary-thyroid axis. Endocrinology 142:2606–2613
Fekete C, Legradi G, Mihaly E, Huang QH, Tatro JB, Rand WM, Emerson CH, Lechan RM 2000 -Melanocyte-stimulating hormone is contained in nerve terminals innervating thyrotropin-releasing hormone-synthesizing neurons in the hypothalamic paraventricular nucleus and prevents fasting-induced suppression of prothyrotropin-releasing hormone gene expression. J Neurosci 20:1550–1558
Fekete C, Mihaly E, Luo LG, Kelly J, Clausen JT, Mao Q, Rand WM, Moss LG, Kuhar M, Emerson CH, Jackson IM, Lechan RM 2000 Association of cocaine- and amphetamine-regulated transcript-immunoreactive elements with thyrotropin-releasing hormone-synthesizing neurons in the hypothalamic paraventricular nucleus and its role in the regulation of the hypothalamic-pituitary-thyroid axis during fasting. J Neurosci 20:9224–9234
Diengott D, Mirsky IA, Perisutti G 1957 Effect of fasting on insulinase activity and on hypoglycemic response to insulin. Endocrinology 60:303–307
Fukumoto H, Seino Y, Koh G, Takeda J, Tsuji K, Kurose T, Kitano N, Tsuda K, Taminato T, Imura H 1986 Effects of oral glucose administration on preproinsulin mRNA in rats in vivo. Biochem Biophys Res Commun 141:1201–1206
Davis JD, Wirtshafter D, Asin KE, Brief D 1981 Sustained intracerebroventricular infusion of brain fuels reduces body weight and food intake in rats. Science 212:81–83
Porte Jr D, Woods SC 1981 Regulation of food intake and body weight in insulin. Diabetologia 20(Suppl):274–280
Campfield LA, Smith FJ, Guisez Y, Devos R, Burn P 1995 Recombinant mouse OB protein: evidence for a peripheral signal linking adiposity and central neural networks. Science 269:546–549
Schwartz MW, Woods SC, Porte Jr D, Seeley RJ, Baskin DG 2000 Central nervous system control of food intake. Nature 404:661–671
Ibrahim N, Bosch MA, Smart JL, Qiu J, Rubinstein M, Ronnekleiv OK, Low MJ, Kelly MJ 2003 Hypothalamic proopiomelanocortin neurons are glucose responsive and express KATP channels. Endocrinology 144:1331–1340
Schwartz MW, Sipols AJ, Marks JL, Sanacora G, White JD, Scheurink A, Kahn SE, Baskin DG, Woods SC, Figlewicz DP 1992 Inhibition of hypothalamic neuropeptide Y gene expression by insulin. Endocrinology 130:3608–3616
Muroya S, Yada T, Shioda S, Takigawa M 1999 Glucose-sensitive neurons in the rat arcuate nucleus contain neuropeptide Y. Neurosci Lett 264:113–116
Wang J, Leibowitz KL 1997 Central insulin inhibits hypothalamic galanin and neuropeptide Y gene expression and peptide release in intact rats. Brain Res 777:231–236
Benoit SC, Air EL, Coolen LM, Strauss R, Jackman A, Clegg DJ, Seeley RJ, Woods SC 2002 The catabolic action of insulin in the brain is mediated by melanocortins. J Neurosci 22:9048–9052
Chan YY, Steiner RA, Clifton DK 1996 Regulation of hypothalamic neuropeptide-Y neurons by growth hormone in the rat. Endocrinology 137:1319–1325
Bagnol D, Lu XY, Kaelin CB, Day HE, Ollmann M, Gantz I, Akil H, Barsh GS, Watson SJ 1999 Anatomy of an endogenous antagonist: relationship between Agouti-related protein and proopiomelanocortin in brain. J Neurosci 19:RC26
Eberwine JH, Jonassen JA, Evinger MJ, Roberts JL 1987 Complex transcriptional regulation by glucocorticoids and corticotropin-releasing hormone of proopiomelanocortin gene expression in rat pituitary cultures. DNA 6:483–492
Couceyro PR, Koylu EO, Kuhar MJ 1997 Further studies on the anatomical distribution of CART by in situ hybridization. J Chem Neuroanat 12:229–241
Curcio C, Baqui MM, Salvatore D, Rihn BH, Mohr S, Harney JW, Larsen PR, Bianco AC 2001 The human type 2 iodothyronine deiodinase is a selenoprotein highly expressed in a mesothelioma cell line. J Biol Chem 276:30183–30187
Seeley RJ, Woods SC 2003 Monitoring of stored and available fuel by the CNS: implications for obesity. Nat Rev Neurosci 4:901–909
Ahima RS, Saper CB, Flier JS, Elmquist JK 2000 Leptin regulation of neuroendocrine systems. Front Neuroendocrinol 21:263–307
Levin BE, Routh VH, Kang L, Sanders NM, Dunn-Meynell AA 2004 Neuronal glucosensing: what do we know after 50 years Diabetes 53:2521–2528
Mobbs CV, Kow LM, Yang XJ 2001 Brain glucose-sensing mechanisms: ubiquitous silencing by aglycemia vs. hypothalamic neuroendocrine responses. Am J Physiol Endocrinol Metab 281:E649–E654
Levin BE 2002 Metabolic sensors: viewing glucosensing neurons from a broader perspective. Physiol Behav 76:397–401
Isse T, Ueta Y, Serino R, Noguchi J, Yamamoto Y, Nomura M, Shibuya I, Lightman SL, Yamashita H 1999 Effects of leptin on fasting-induced inhibition of neuronal nitric oxide synthase mRNA in the paraventricular and supraoptic nuclei of rats. Brain Res 846:229–235
Guo F, Bakal K, Minokoshi Y, Hollenberg AN 2004 Leptin signaling targets the thyrotropin-releasing hormone gene promoter in vivo. Endocrinology 145:2221–2227
Nillni EA, Vaslet C, Harris M, Hollenberg A, Bjorbak C, Flier JS 2000 Leptin regulates prothyrotropin-releasing hormone biosynthesis. Evidence for direct and indirect pathways. J Biol Chem 275:36124–36133
Huo L, Munzberg H, Nillni EA, Bjorbaek C 2004 Role of signal transducer and activator of transcription 3 in regulation of hypothalamic TRH gene expression by leptin. Endocrinology 145:2516–2523
Legradi G, Emerson CH, Ahima RS, Rand WM, Flier JS, Lechan RM 1998 Arcuate nucleus ablation prevents fasting-induced suppression of proTRH mRNA in the hypothalamic paraventricular nucleus. Neuroendocrinology 68:89–97
Perello M, Stuart RC, Rosapour T, Nillni EA The role of leptin in the stimulation of the proTRH neurons through direct and indirect pathways. Program of the 87th Annual Meeting of The Endocrine Society, San Francisco, CA, 2005, p 219 (Abstract P1-209)
Toni R, Jackson IM, Lechan RM 1990 Neuropeptide-Y-immunoreactive innervation of thyrotropin-releasing hormone-synthesizing neurons in the rat hypothalamic paraventricular nucleus. Endocrinology 126:2444–2453
Legradi G, Lechan RM 1998 The arcuate nucleus is the major source for neuropeptide Y-innervation of thyrotropin-releasing hormone neurons in the hypothalamic paraventricular nucleus. Endocrinology 139:3262–3270
Legradi G, Lechan RM 1999 Agouti-related protein containing nerve terminals innervate thyrotropin-releasing hormone neurons in the hypothalamic paraventricular nucleus. Endocrinology 140:3643–3652
Ahima RS, Kelly J, Elmquist JK, Flier JS 1999 Distinct physiologic and neuronal responses to decreased leptin and mild hyperleptinemia. Endocrinology 140:4923–4931
Mizuno TM, Kleopoulos SP, Bergen HT, Roberts JL, Priest CA, Mobbs CV 1998 Hypothalamic pro-opiomelanocortin mRNA is reduced by fasting and in ob/ob and db/db mice, but is stimulated by leptin. Diabetes 47:294–297
Mizuno TM, Mobbs CV 1999 Hypothalamic agouti-related protein messenger ribonucleic acid is inhibited by leptin and stimulated by fasting. Endocrinology 140:814–817
Hahn TM, Breininger JF, Baskin DG, Schwartz MW 1998 Coexpression of Agrp and NPY in fasting-activated hypothalamic neurons. Nat Neurosci 1:271–272
Emerson CH, Cohen 3rd JH, Young RA, Alex S, Fang SL 1990 Gender-related differences of serum thyroxine-binding proteins in the rat. Acta Endocrinol (Copenh) 123:72–78
Cusin I, Zakrzewska KE, Boss O, Muzzin P, Giacobino JP, Ricquier D, Jeanrenaud B, Rohner-Jeanrenaud F 1998 Chronic central leptin infusion enhances insulin-stimulated glucose metabolism and favors the expression of uncoupling proteins. Diabetes 47:1014–1019
Cettour-Rose P, Burger AG, Meier CA, Visser TJ, Rohner-Jeanrenaud F 2002 Central stimulatory effect of leptin on T3 production is mediated by brown adipose tissue type II deiodinase. Am J Physiol Endocrinol Metab 283:E980–E987
Rondeel JM, de Greef WJ, Heide R, Visser TJ 1992 Hypothalamo-hypophysial-thyroid axis in streptozotocin-induced diabetes. Endocrinology 130:216–220
Thomas SA, Palmiter RD 1997 Thermoregulatory and metabolic phenotypes of mice lacking noradrenaline and adrenaline. Nature 387:94–97
Niswender KD, Schwartz MW 2003 Insulin and leptin revisited: adiposity signals with overlapping physiological and intracellular signaling capabilities. Front Neuroendocrinol 24:1–10
Bruning JC, Gautam D, Burks DJ, Gillette J, Schubert M, Orban PC, Klein R, Krone W, Muller-Wieland D, Kahn CR 2000 Role of brain insulin receptor in control of body weight and reproduction. Science 289:2122–2125
Bergen HT, Monkman N, Mobbs CV 1996 Injection with gold thioglucose impairs sensitivity to glucose: evidence that glucose-responsive neurons are important for long-term regulation of body weight. Brain Res 734:332–336
Miller JC, Gnaedinger JM, Rapoport SI 1987 Utilization of plasma fatty acid in rat brain: distribution of [14C]palmitate between oxidative and synthetic pathways. J Neurochem 49:1507–1514
Molina PE, Williams P, Abumrad NN 1997 Histaminergic contribution to the metabolic effects of neuroglucopenia. Am J Physiol 272:R1918–R1924(Csaba Fekete, Praful S. Singru, Edith Sa)
Molecular Medicine, New England Medical Center, Boston, Massachusetts 02111
Department of Endocrine Neurobiology (C.F.), Institute of Experimental Medicine, Hungarian Academy of Sciences, 1083 Budapest, Hungary
Division of Endocrinology, Diabetes
Hypertension (M.A.C., A.C.B.), Brigham and Women’s Hospital, Boston, Massachusetts 02115
Division of Endocrinology (R.S.R.), Universidade Federal de Sao Paulo, 04039-032 Sao Paulo, Brazil
Departments of Community Health (W.M.R.) and Neuroscience (R.M.L.), Tufts University School of Medicine, Boston, Massachusetts 02111
Division of Endocrinology (C.H.E.), University of Massachusetts Medical School, Worcester, Massachusetts 01655
Abstract
The reductions in circulating levels of leptin, insulin, and glucose with fasting serve as important homeostasis signals to neurons of the hypothalamic arcuate nucleus that synthesize neuropeptide Y (NPY)/agouti-related protein (AGRP) and -MSH/cocaine and amphetamine-regulated transcript. Because the central administration of leptin is capable of preventing the inhibitory effects of fasting on TRH mRNA in hypophysiotropic neurons primarily through effects on the arcuate nucleus, we determined whether the continuous administration of 30 mU/d insulin or 648 μg/d glucose into the cerebrospinal fluid by osmotic minipump might also have similar effects on the hypothalamic-pituitary-thyroid axis. As anticipated, the intracerebroventricular infusion of leptin reduced fasting-induced elevations in NPY and AGRP mRNA and increased proopiomelanocortin and cocaine and amphetamine-regulated transcript mRNA in the arcuate nucleus. In addition, leptin prevented fasting-induced reduction in pro-TRH mRNA levels in the paraventricular nucleus and in circulating thyroid hormone levels. In contrast, whereas insulin increased proopiomelanocortin mRNA and both insulin and glucose reduced NPY mRNA in arcuate nucleus neurons, neither prevented the fasting-induced suppression in hypophysiotropic TRH mRNA or circulating thyroid hormone levels. We conclude that insulin and glucose only partially replicate the central effects of leptin and may not be essential components of the hypothalamic-pituitary-thyroid regulatory system during fasting.
Introduction
FASTING RESULTS IN a number of adaptive responses to decrease energy expenditure. Included is the development of central hypothyroidism, characterized by decreased synthesis of TRH in the hypothalamic paraventricular nucleus (PVN) and low TSH and thyroid hormone levels in the bloodstream (1, 2, 3). Because thyroid hormone stimulates mitochondrial oxygen consumption and increases thermogenesis (4, 5), the reduction in circulating levels of thyroid hormone during fasting is presumed to be an important mechanism to conserve energy.
Data by Ahima et al. (6) and from our laboratories (1) have demonstrated that the amount of leptin circulating in the bloodstream serves as an important regulatory signal to the brain that modulates the responsiveness of the hypothalamic-pituitary-thyroid (HPT) axis to circulating levels of thyroid hormone. Thus, fasting-induced inhibition of the HPT axis can be completely reversed if circulating levels of leptin are increased by exogenous administration, despite continuation of the fast (1, 6).
The effect of leptin on the HPT axis is mediated primarily through monosynaptic connections to TRH neurons in the PVN from two leptin-sensitive neuronal populations in the arcuate nucleus (7) that have opposing effects on TRH biosynthesis. These include neuropeptide Y (NPY)/agouti-related protein (AGRP)-synthesizing neurons that are inhibitory to hypophysiotropic TRH neurons (8, 9, 10), and -MSH/cocaine and amphetamine-regulated transcript (CART)-synthesizing neurons that are stimulatory (11, 12). In addition to changes in circulating leptin levels, however, fasting also reduces circulating levels of insulin and glucose (13, 14) that may also serve as regulatory signals to the central nervous system. Like leptin, the central administration of insulin or glucose inhibits food intake (15, 16, 17) and has similar effects on the expression of proopiomelanocortin (POMC) and NPY mRNA in arcuate nucleus neurons (18, 19, 20, 21, 22, 23). We hypothesized, therefore, that insulin and/or glucose may also contribute to the regulation of the HPT axis during fasting through actions on the arcuate nucleus. Accordingly, we compared the effects of the central administration of leptin on the HPT axis during fasting to that of insulin and glucose.
Materials and Methods
The experiments were carried out on adult male Sprague-Dawley rats, weighing 200–250 g. Animals were housed under standard environmental conditions (light from 0600–1800 h; temperature, 22 ± 1 C; rat chow and water ad libitum). All experimental protocols were reviewed and approved by the Institutional Animal Care and Use Committee of the Tufts-New England Medical Center.
Adult rats (n = 36) were implanted with a 22-gauge stainless steel guide cannula (Plastics One, Inc., Roanoke, VA) into the lateral cerebral ventricle under stereotaxic control (coordinates from bregma: anterior-posterior, –0.8; lateral, 1.2; dorsal-ventral, 3.5) through a burr hole in the skull. The cannula was secured to the skull with three stainless steel screws and dental cement and was temporarily occluded with a dummy cannula. Bacitracin ointment was applied to the interface of the cement and the skin after surgery. One week after intracerebroventricular cannulation, an osmotic minipump (Alzet model 1003D; Alza Corp., Palo Alto, CA) was implanted under sodium pentobarbital anesthesia (35 mg/kg body weight ip) intradermally between the scapulas and connected with durameter vinyl tubing (Scientific Comodities Inc., Lake Havasu City, AZ) to a 28-gauge needle that was permanently inserted into and extended 1 mm below the external guide cannula. The animals were divided into five groups. The first group (group 1) had free access to food, whereas in the remaining groups (groups 2–5), food was withdrawn over the 3 d of experimentation. Osmotic minipumps delivered either artificial cerebrospinal fluid (aCSF) (140 mM NaCl, 3.35 mM KCl, 1.15 mM MgCl2, 1.26 mM CaCl2, 1.2 mM Na2HPO4, 0.3 mM NaH2PO4, 0.1% BSA, pH 7.4) (groups 1 and 2), 10 μg/24 h mouse leptin (Lilly Pharmaceutical Co, Indianapolis, IN) in aCSF (group 3), 30 mU porcine insulin (Lilly) in aCSF (group 4), or 648 μg glucose in aCSF (group 5) for 3 d at a rate of 1 μl/h.
At the completion of the experiment, the animals were anesthetized with sodium pentobarbital between 0900 and 1200 h, CSF was taken from the cisterna magna for mouse leptin, porcine insulin, and glucose measurements; brown fat was dissected from the interscapular region; blood was taken from the inferior vena cava for measurement of serum TSH, thyroid hormone, and glucose levels; and the animals were immediately perfused with fixative as described below. Blood was collected into polypropylene tubes and centrifuged for 15 min at 4000 rpm, and the plasma was stored at –80 C until assayed. The brown fat was snap-frozen in dry ice and stored at –80 C until processed for type 2 iodothyronine deiodinase (D2) enzymatic activity and uncoupling protein-1 (UCP1) mRNA.
Tissue preparation for in situ hybridization histochemistry
Under sodium pentobarbital anesthesia, the animals were perfused transcardially with 20 ml 0.01 M PBS, pH 7.4, containing 15,000 U/liter heparin sulfate, followed by 150 ml 4% paraformaldehyde in PBS. The brains were removed and postfixed by immersion in the same fixative for 2 h at room temperature. Tissue blocks containing the hypothalamus were cryoprotected in 20% sucrose in PBS at 4 C overnight and then frozen on dry ice. Serial 18-μm-thick coronal sections through the rostrocaudal extent of the PVN and the arcuate nucleus were cut on a cryostat (Leica CM3050 S; Leica Microsystems, Nussloch GmbH, Germany) and adhered to SuperFrost Plus glass slides (Fisher Scientific, Pittsburgh, PA) to obtain four sets of slides, each set containing every fourth section through the PVN and the arcuate nucleus. Cannula placement was confirmed by light microscopic examination. The tissue sections were desiccated overnight at 42 C and were stored at –80 C until prepared for in situ hybridization histochemistry.
In situ hybridization histochemistry
Every fourth section of the PVN was hybridized with a 1241-bp single-stranded [35S]UTP-labeled cRNA probe for pro-TRH, and every fourth section through the arcuate nucleus was hybridized with single-stranded [35S]UTP-labeled cRNA probe for NPY (24), AGRP (25), POMC (26), and CART (27) as previously described (10). Hybridizations were performed under plastic coverslips in a buffer containing 50% formamide, a 2-fold concentration of standard sodium citrate, 10% dextran sulfate, 0.5% sodium dodecyl sulfate, 250 μg/ml denatured salmon sperm DNA, and 6 x 105 cpm radiolabeled probe for 16 h at 56 C. Slides were dipped into Kodak NTB2 autoradiography emulsion (Eastman Kodak Co., Rochester, NY), and the autoradiograms were developed after 2–4 d of exposure at 4 C.
Image analysis
Autoradiograms were visualized under dark-field illumination using a COHU 4910 video camera (COHU, Inc., San Diego, CA). The images were analyzed with a Macintosh G4 computer (Apple Computers, Cupertino, CA) using Scion Image (Scion Corp., Frederick, MD). Background density points were removed by thresholding the image, and integrated density values (density x area) of hybridized neurons in the same region of each side of the PVN or arcuate nucleus were measured in five to six consecutive sections for each animal depending upon the radiolabeled probe (TRH = 6, NPY = 5, AGRP = 5, CART = 5, and POMC = 5). Nonlinearity of radioactivity in the emulsion was evaluated by comparing density values with a calibration curve created from autoradiograms of known dilutions of the radiolabeled probes immobilized on glass slides in 1.5% gelatin fixed with 4% formaldehyde and exposed and developed simultaneously with the in situ hybridization autoradiograms.
For mRNAs shown to be affected by insulin or glucose above, a more detailed analysis was performed using the approach described previously. However, integrated density values were measured for each hybridized neuron in the arcuate nucleus sections at the same level of the rostral caudal extent of the median eminence and expressed as a histogram, reflecting the number of arcuate nucleus neurons with integrated density values above background ranging from 0.1–110 density units at intervals of 5 density units.
Hormone and glucose measurements
Serum T3 and T4 were measured by the Diagnostic Products Corp. (Los Angeles, CA) TKT41 and TKT31 assay systems, respectively. The intraassay coefficients of variation for these assays were 3.1 and 2.8%, respectively. The interassay coefficients of variation for these assays were 5.7 and 5.9%, respectively. Serum TSH was measured by the rat TSH assay kit RPA 554 obtained from GE Healthcare (formerly Amersham Bioscience, Piscataway, NJ). The intraassay coefficient of variation for this assay was 4.8%, and its interassay coefficient of variation was 8.5%. Leptin was measured by the Linco Research (St. Charles, MO) ML-82K mouse leptin assay. Its intraassay coefficient of variation was 3.3%, and its interassay coefficient of variation was 4.9%. Insulin was measured by the Linco PI-12K porcine insulin assay. The blood glucose levels were measured by glucometer (Medisense Precision Xtra; Abbott, Bedford, MA). CSF glucose levels were measured by the Freestyle Blood Glucose Monitoring System (Therasense, Alameda, CA).
Brown adipose tissue (BAT) measurements
D2 activity was measured as previously described (28). Approximately 250 μg total BAT lysate protein was incubated for 2 h in the presence of 1 nM [125I]5'T4, 20 mM dithiothreitol, and 1 mM propylthiouracil. Specific T4 to T3 conversion was calculated by subtracting nonspecific deiodination in tubes containing the same amount of lysate protein obtained from human embryonic kidney cells. The background activity of these samples was less than 2%. Deiodinase activity was expressed as femtomoles of T4 deiodinated per minute per milligram of protein. Total RNA was extracted using TRIzol Reagent (Invitrogen, Carlsbad, CA) and used to synthesize cDNA using SuperScript First-Strand Synthesis System for RT-PCR (Invitrogen). The generated cDNAs were used in a real-time PCR using the QuantiTect SYBR Green PCR kit in I-Cycler (Bio-Rad, Hercules, CA). Standard curves (five-point serial dilution of mixed experimental and control groups cDNA) were analyzed in each assay and used as calibrators to the relative quantification of product generated in the exponential phase of amplification curve. The r2 value was below 0.99 for all standard curves, and the amplification efficiency varied between 90 and 100%. Sample quantification was calculated by the standard curve and corrected by the internal control -actin mRNA in all experiments. Specific oligonucleotides for rat UCP1 and -actin were designed using Beacon Designer 2.06 (Premier Biosoft International, Palo Alto, CA).
Statistical analyses
Results are presented as mean ± SEM. One-way ANOVA was used to examine the effect of group on each variable of interest, followed by the Newman-Keuls post hoc test to determine whether the ANOVA showed a significant (P < 0.05) group effect. Because of skewness of the CART data, the natural log of the variables was used to test significance. To explore whether a combination of T3, T4, and TRH would be able to further separate the groups, discriminant function analysis was used, with group differences being evaluated with Hotelling’s t2 statistic. Density values of individual neurons in each experimental group for NPY and POMC were examined for biomodality by testing the kurtosis values for each animal after correcting for skewness by taking the natural log of the NPY and POMC density values. Analyses were conducted using the SPSS statistical program version 11.5 (SPSS Inc., Chicago, IL).
Results
Effect of central leptin, insulin, and glucose administration on body weight and serum and CSF hormone parameters in fasted animals
Body weight change and serum and CSF hormone levels are shown in Table 1. Fasting resulted in an approximately 15% weight loss during the 3 d of experimentation. Central leptin and insulin administration had no influence on weight loss of the animals, and although glucose appeared to diminish the effects of fasting on weight loss, the difference was not significant.
Murine leptin and porcine insulin levels were elevated in the CSF only in the leptin- and insulin-treated groups, respectively. Glucose levels tended to decrease in the CSF of fasted animals, and a further decrease was observed in the leptin-treated fasted animals. The glucose level in the CSF of glucose-treated fasted animals was significantly increased compared with fasted rats receiving aCSF but was not different when compared with the glucose levels of fed rats. Serum glucose levels of fasted animals were significantly decreased compared with fed animals, and the central administration of leptin caused an even further decline. However, the central administration of neither insulin nor glucose had any effect on peripheral glucose levels.
As expected, fasting resulted in a fall in serum T4 and T3 levels. Central leptin treatment completely reversed the effects of fasting on serum T3 levels and attenuated the effects of fasting on serum T4 levels. The central administration of either insulin or glucose, however, had no significant effects on thyroid hormone levels. Serum TSH levels were not significantly changed by any of the treatments. Leptin administration significantly increased the T3/T4 ratio in the peripheral blood, whereas fasting and the other treatments had no effect.
Effect of central leptin, insulin, and glucose administration on BAT UCP1 mRNA levels and D2 activity in fasted animals
Fasting significantly decreased UCP1 mRNA levels in BAT (Table 1). Leptin administration to fasting animals, however, increased UCP1 mRNA to levels even above fed controls, and insulin treatment prevented the fasting-induced inhibition of the UCP1 mRNA. Glucose had no significant effect. In contrast, only leptin treatment altered D2 activity. Fasting, insulin, and glucose treatment had no effect, but central leptin administration resulted in an approximately 2-fold increase in D2 activity.
Effect of central leptin, insulin, or glucose administration on gene expression of feeding-related peptides in the arcuate nucleus
In agreement with the literature (18), fasting increased the NPY and AGRP expression and decreased the POMC and CART mRNA levels in the arcuate nucleus (Figs. 1–4). All groups showed significant differences (P < 0.01) by ANOVA. Leptin, insulin, and glucose exerted similar effects on NPY-producing arcuate nucleus neurons in fasted rats (Fig. 1). All three substances significantly decreased NPY mRNA to the level observed in fed animals, despite continuation of the fast (Fig. 1). In contrast, leptin prevented the fasting-induced increase in AGRP mRNA levels, but insulin and glucose had no effect (Fig. 2). In addition, whereas both leptin and insulin significantly reversed the fasting-induced fall of the POMC mRNA levels, glucose had no effect (Fig. 3). Leptin resulted in a significant increase in CART gene expression in the arcuate nucleus compared with fasting animals, but the central administration of insulin or glucose had no effect (Fig. 4).
Regional differences were not observed in the response of the arcuate nucleus neurons to any of the treatments. Analysis of density values for individual arcuate nucleus neurons expressing the two mRNAs affected by leptin and insulin or glucose revealed that compared with fasting animals, leptin, insulin, and glucose all resulted in a significant and marked shift of the density values for NPY mRNA in fasting animals to the left (Fig. 5A), whereas leptin and insulin-treated fasting animals showed a shift of the density values for POMC mRNA to the right (Fig. 5B). Testing for bimodality of the density values for the individual cells examined was performed by computing kurtosis values. For none of the animals in either the NPY or POMC groups could the hypothesis of normality be rejected.
Effect of central leptin, insulin, or glucose administration on TRH mRNA in hypophysiotropic neurons in the PVN
In fed animals, neurons containing pro-TRH mRNA were readily visualized by in situ hybridization histochemistry, symmetrically distributed in the medial and periventricular parvocellular subdivisions of the PVN on either side of the third ventricle (Fig. 6A). As previously recognized, fasting caused a marked reduction in the hybridization signal (Fig. 6B), whereas leptin administration completely prevented the fasting-induced inhibition of TRH mRNA (Fig. 6C). In contrast, no significant effect on the density of silver grains over pro-TRH neurons in the PVN in fasting animals was observed after insulin administration (Fig. 6D). Central glucose administration to fasted animals had a tendency to further decrease TRH mRNA levels below that observed in the fasted animals receiving aCSF (Fig. 6E), and they were significantly different when examined by t test (P = 0.028). However, the values were not significantly different when subjected to a more rigorous statistical analysis using ANOVA and Newman-Keuls testing or when the effect of glucose on the HPT axis was assessed by discriminant function analysis using combinations of T3, T4, and TRH mRNA. The results of the image analyses are shown in Fig. 6F.
Discussion
Two main theories have been proposed as to how the brain senses the status of the peripheral energy stores (29). According the lipostatic theory, the brain monitors the storage and metabolism of fat, whereas the glucostatic theory is based on the hypothesis that the brain senses the storage and use of carbohydrates (29). It is now recognized that the main adiposity signal to the brain is leptin (18), with circulating leptin levels directly proportional to the total amount of stored fat (30). The glucostatic signal is glucose itself sensed by specific populations of glucose-sensing neurons in the central nervous system (31, 32). Insulin also signals to the brain, and its peripheral level is dependent on both the amount of adipose tissue and circulating glucose levels (29). Interestingly, neurons in a discrete location in the hypothalamus, the arcuate nucleus, are a major central target for all three signaling substances, and in some instances the same arcuate nucleus neurons express leptin and insulin receptors and are glucose sensitive (33). This is the case for NPY-producing arcuate nucleus neurons that are inhibited by all three substances and -MSH-producing neurons that are activated by all three substances (18, 19, 20, 21, 22, 23). Because NPY and -MSH have critical roles in regulation of the HPT axis (7), we determined whether in addition to leptin, insulin and glucose might also affect hypophysiotropic TRH gene expression and circulating levels of TSH and thyroid hormones.
Similar to previous observations (1, 34), the central administration of leptin to fasting animals prevented fasting-induced inhibition of the TRH mRNA in hypophysiotropic neurons. Although a direct effect of leptin on TRH neurons residing in the PVN has been proposed (35, 36, 37), evidence that ablation of the arcuate nucleus prevents alterations in TRH gene expression during fasting and after leptin administration to fasting animals (38) suggests that the major effect of leptin on the hypophysiotropic TRH neurons may be mediated indirectly through arcuate nucleus neurons (38). This is further supported by the observation of Perello et al. (39) showing that the stimulatory effect of central leptin administration on TRH gene expression in the PVN can be completely reversed by administration of a melanocortin receptor antagonist. Two main neuronal populations of arcuate nucleus neurons are involved in mediating the effects of leptin on the central nervous system, including NPY neurons that cosynthesize AGRP and -MSH neurons that cosynthesize CART (18). These two antagonistic neuron populations innervate hypophysiotropic TRH neurons (11, 12, 40, 41, 42) and have opposing effects in the regulation of TRH mRNA (8, 10, 11, 12).
As observed in the current study, leptin inhibited AGRP and NPY mRNA synthesis and increased the gene expression of both stimulatory peptides, CART and -MSH, completely reversing the effect of fasting on these two neuronal populations. These findings are in agreement with other investigators showing that in fasting animals, leptin has inhibitory effects on the gene expression of AGRP and NPY and stimulatory effects on CART and -MSH mRNA in arcute nucleus neurons (43, 44, 45, 46). Leptin also reversed fasting-induced effects on circulating levels of thyroid hormones, increasing circulating T4 levels toward normal and even increasing circulating T3 levels above normal. We presume that the explanation for the inability of leptin to fully restore T4 levels to normal is a result of fasting-induced suppression of the T4-binding protein, transthyretin, which is not regulated by leptin administration (1, 47). The rise in T3 may be secondary to an increase in the conversion of T4 to T3 in BAT as a result of stimulation of D2 by leptin. This explanation may also account for the increase in the T3/T4 ratio in the leptin-treated fasting animals compared with the fed controls. Central leptin administration also influenced the UCP1 mRNA level in BAT, in support of the ability for leptin to centrally induce mitochondrial heat production in this tissue (48). Similar observations regarding the effects of central leptin administration on the T3/T4 ratio, BAT D2 activity, and UCP1 mRNA levels were reported by Cettour-Rose et al. (49).
Insulin is thought to have similar roles in the regulation of energy homeostasis as leptin. Both hormones decline in association with fasting, exert anorexic effects, and increase energy expenditure (29). In addition, both hormones have similar effects on NPY- and -MSH-producing neurons in the arcuate nucleus (20, 23, 43). It seemed reasonable to assume, therefore, that insulin might also have a similar role in the regulation of the HPT axis as leptin. Indeed, streptozotocin-induced diabetes reduces the hypothalamic secretion of TRH and circulating thyroid hormone levels (50). In contrast to leptin, however, central administration of insulin had no significant effects on any parameters of the HPT axis. TRH mRNA, TSH, and thyroid hormone levels were not significantly different from levels in fasting rats receiving aCSF. Thus, we presume that in rats, insulin may not contribute in an important way to the development of central hypothyroidism during fasting or recovery of thyroid function with refeeding. Because a daily intracerebroventricular dose of 8 mU porcine insulin has been reported sufficient to exert an inhibitory effect on food intake and a stimulatory effect on POMC gene expression (23), it seems unlikely that the absence of a response of the HPT axis to insulin in fasting animals can be attributed to the administration of an insufficient amount of insulin. This is further supported by the observation that insulin prevented the fasting-induced increase in NPY mRNA and increased POMC gene expression, as observed by other groups (20, 22, 23). In addition, insulin treatment also prevented the fasting-induced inhibition of UCP1 mRNA levels in the BAT, presumably mediated through central activation of the sympathetic nervous system (51). Thus, despite the overlapping second messenger pathways for leptin and insulin (52), insulin would appear to only partially recapitulate the central effects of leptin, a concept consistent with the observation that the central insulin receptor-deficient mouse has a milder phenotype than the leptin receptor-deficient ob/ob mouse (53). Nevertheless, the possibility that insulin could potentiate the effects of leptin on the HPT axis has not been excluded by these studies.
Based on our earlier data that exogenous -MSH administration prevents the fasting-induced inhibition of the pro-TRH mRNA levels in hypophysiotropic neurons and NPY inhibits TRH synthesis in the same neuronal population (10, 11), one might assume that insulin-induced changes in NPY and POMC gene expression in arcuate nucleus neurons should have been sufficient to increase pro-TRH gene expression. The possibility that only a subpopulation of NPY- or POMC-producing neurons were responsive to insulin, leaving a nidus of unresponsive cells that retain the activating or inhibitory effects of fasting, respectively, and project to TRH neurons in the PVN, is made unlikely by the absence of any apparent qualitative regional difference in the arcuate nucleus in each experimental group or quantitative difference by statistical analysis of individual neurons in the arcuate nucleus for the presence of bimodality in response to leptin, insulin, or glucose. Contrary to leptin administration in fasting animals, which prevents the rise in AGRP and increases CART gene expression, however, insulin administration had no effect on either mRNA. It may be presumed, therefore, that the persistent increase in AGRP gene expression simultaneously with suppression in CART gene expression was enough to overcome the stimulatory effects of increased -MSH and reduced NPY on hypophysiotropic TRH neurons. This hypothesis is supported by the finding that when AGRP is administered exogenously to fed animals, TRH mRNA is inhibited in hypophysiotropic neurons despite persistent elevations in POMC gene expression (8). Alternatively, other factors may be involved in the regulation of the HPT axis during fasting that have not yet been characterized.
The glucose-sensitive neurons of the brain also have a pivotal role in the regulation of energy homeostasis. Lesion of these neurons by gold-thio-glucose results in an obese phenotype (54). When administered directly into the central nervous system, glucose has been reported to activate the electrophysiological properties of POMC neurons (19), reduce food intake (15), and block counterregulation during systemic hypoglycemia (55). In addition, central glucopenia induced by 2-deoxy-D-glucose administration increases food intake (54) and induces a peripheral counterregulatory response including hepatic glucose production and an increase in plasma epinephrine, norepinephrine, insulin, glucagon, and cortisol levels (56).
In our study, similar to insulin treatment, glucose administration to fasted animals had no stimulatory effects on the HPT axis, making it unlikely that the fall in glucose levels during fasting is responsible for inhibition of the HPT axis. In fact, a tendency for decreased pro-TRH gene expression was observed in the hypophysiotropic neurons of the glucose-treated rats even below that of fasting rats receiving aCSF but was significant only when subjected to statistical analysis using the Student’s t test. However, glucose did have a significant effect on NPY gene expression in arcuate nucleus neurons, preventing the fasting-induced increase in NPY mRNA. Additional studies are required to clarify whether glucose sensing by hypothalamic neurons are indeed coupled to suppression of hypophysiotropic TRH.
We conclude that insulin and glucose only partially recapitulate the effects of leptin on the hypothalamus, having selective effects on the gene expression of POMC and/or NPY in arcuate nucleus neurons but not AGRP and CART. In contrast to leptin, changes in insulin and glucose do not appear to have an important role in fasting-induced central hypothyroidism.
Footnotes
This work was supported by Grants NIH DK-37021, OTKA T046492, and Sixth European Union Program (LSHM-CT-2003-503041).
First Published Online October 6, 2005
Abbreviations: aCSF, Artificial cerebrospinal fluid; AGRP, agouti-related protein; BAT, brown adipose tissue; CART, cocaine and amphetamine-regulated transcript; D2, type 2 iodothyronine deiodinase; HPT, hypothalamic-pituitary-thyroid; NPY, neuropeptide Y; POMC, proopiomelanocortin; PVN, paraventricular nucleus; UCP1, uncoupling protein-1.
Accepted for publication September 23, 2005.
References
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
van Haasteren GA, Linkels E, Klootwijk W, van Toor H, Rondeel JM, Themmen AP, 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 (proTRH) mRNA and the hypothalamic release of proTRH-derived peptides: role of the adrenal gland. J Endocrinol 145:143–153
Rondeel JM, Heide R, de Greef WJ, van Toor H, van Haasteren GA, Klootwijk W, Visser TJ 1992 Effect of starvation and subsequent refeeding on thyroid function and release of hypothalamic thyrotropin-releasing hormone. Neuroendocrinology 56:348–353
de Jesus LA, Carvalho SD, Ribeiro MO, Schneider M, Kim SW, Harney JW, 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
Lanni A, Moreno M, Lombardi A, de Lange P, Goglia F 2001 Control of energy metabolism by iodothyronines. J Endocrinol Invest 24:897–913
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
Lechan RM, Fekete C 2004 Feedback regulation of thyrotropin-releasing hormone (TRH): mechanisms for the nonthyroidal illness syndrome. J Endocrinol Invest 27:105–119
Fekete C, Sarkar S, Rand WM, Harney JW, Emerson CH, Bianco AC, Lechan RM 2002 Agouti-related protein (AGRP) has a central inhibitory action on the hypothalamic-pituitary-thyroid (HPT) axis: comparisons between the effect of AGRP and neuropeptide Y on energy homeostasis and the HPT axis. Endocrinology 143:3846–3853
Fekete C, Marks DL, Sarkar S, Emerson CH, Rand WM, Cone RD, Lechan RM 2004 Effect of Agouti-related protein in regulation of the hypothalamic-pituitary-thyroid axis in the melanocortin 4 receptor knockout mouse. Endocrinology 145:4816–4821
Fekete C, Kelly J, Mihaly E, Sarkar S, Rand WM, Legradi G, Emerson CH, Lechan RM 2001 Neuropeptide Y has a central inhibitory action on the hypothalamic-pituitary-thyroid axis. Endocrinology 142:2606–2613
Fekete C, Legradi G, Mihaly E, Huang QH, Tatro JB, Rand WM, Emerson CH, Lechan RM 2000 -Melanocyte-stimulating hormone is contained in nerve terminals innervating thyrotropin-releasing hormone-synthesizing neurons in the hypothalamic paraventricular nucleus and prevents fasting-induced suppression of prothyrotropin-releasing hormone gene expression. J Neurosci 20:1550–1558
Fekete C, Mihaly E, Luo LG, Kelly J, Clausen JT, Mao Q, Rand WM, Moss LG, Kuhar M, Emerson CH, Jackson IM, Lechan RM 2000 Association of cocaine- and amphetamine-regulated transcript-immunoreactive elements with thyrotropin-releasing hormone-synthesizing neurons in the hypothalamic paraventricular nucleus and its role in the regulation of the hypothalamic-pituitary-thyroid axis during fasting. J Neurosci 20:9224–9234
Diengott D, Mirsky IA, Perisutti G 1957 Effect of fasting on insulinase activity and on hypoglycemic response to insulin. Endocrinology 60:303–307
Fukumoto H, Seino Y, Koh G, Takeda J, Tsuji K, Kurose T, Kitano N, Tsuda K, Taminato T, Imura H 1986 Effects of oral glucose administration on preproinsulin mRNA in rats in vivo. Biochem Biophys Res Commun 141:1201–1206
Davis JD, Wirtshafter D, Asin KE, Brief D 1981 Sustained intracerebroventricular infusion of brain fuels reduces body weight and food intake in rats. Science 212:81–83
Porte Jr D, Woods SC 1981 Regulation of food intake and body weight in insulin. Diabetologia 20(Suppl):274–280
Campfield LA, Smith FJ, Guisez Y, Devos R, Burn P 1995 Recombinant mouse OB protein: evidence for a peripheral signal linking adiposity and central neural networks. Science 269:546–549
Schwartz MW, Woods SC, Porte Jr D, Seeley RJ, Baskin DG 2000 Central nervous system control of food intake. Nature 404:661–671
Ibrahim N, Bosch MA, Smart JL, Qiu J, Rubinstein M, Ronnekleiv OK, Low MJ, Kelly MJ 2003 Hypothalamic proopiomelanocortin neurons are glucose responsive and express KATP channels. Endocrinology 144:1331–1340
Schwartz MW, Sipols AJ, Marks JL, Sanacora G, White JD, Scheurink A, Kahn SE, Baskin DG, Woods SC, Figlewicz DP 1992 Inhibition of hypothalamic neuropeptide Y gene expression by insulin. Endocrinology 130:3608–3616
Muroya S, Yada T, Shioda S, Takigawa M 1999 Glucose-sensitive neurons in the rat arcuate nucleus contain neuropeptide Y. Neurosci Lett 264:113–116
Wang J, Leibowitz KL 1997 Central insulin inhibits hypothalamic galanin and neuropeptide Y gene expression and peptide release in intact rats. Brain Res 777:231–236
Benoit SC, Air EL, Coolen LM, Strauss R, Jackman A, Clegg DJ, Seeley RJ, Woods SC 2002 The catabolic action of insulin in the brain is mediated by melanocortins. J Neurosci 22:9048–9052
Chan YY, Steiner RA, Clifton DK 1996 Regulation of hypothalamic neuropeptide-Y neurons by growth hormone in the rat. Endocrinology 137:1319–1325
Bagnol D, Lu XY, Kaelin CB, Day HE, Ollmann M, Gantz I, Akil H, Barsh GS, Watson SJ 1999 Anatomy of an endogenous antagonist: relationship between Agouti-related protein and proopiomelanocortin in brain. J Neurosci 19:RC26
Eberwine JH, Jonassen JA, Evinger MJ, Roberts JL 1987 Complex transcriptional regulation by glucocorticoids and corticotropin-releasing hormone of proopiomelanocortin gene expression in rat pituitary cultures. DNA 6:483–492
Couceyro PR, Koylu EO, Kuhar MJ 1997 Further studies on the anatomical distribution of CART by in situ hybridization. J Chem Neuroanat 12:229–241
Curcio C, Baqui MM, Salvatore D, Rihn BH, Mohr S, Harney JW, Larsen PR, Bianco AC 2001 The human type 2 iodothyronine deiodinase is a selenoprotein highly expressed in a mesothelioma cell line. J Biol Chem 276:30183–30187
Seeley RJ, Woods SC 2003 Monitoring of stored and available fuel by the CNS: implications for obesity. Nat Rev Neurosci 4:901–909
Ahima RS, Saper CB, Flier JS, Elmquist JK 2000 Leptin regulation of neuroendocrine systems. Front Neuroendocrinol 21:263–307
Levin BE, Routh VH, Kang L, Sanders NM, Dunn-Meynell AA 2004 Neuronal glucosensing: what do we know after 50 years Diabetes 53:2521–2528
Mobbs CV, Kow LM, Yang XJ 2001 Brain glucose-sensing mechanisms: ubiquitous silencing by aglycemia vs. hypothalamic neuroendocrine responses. Am J Physiol Endocrinol Metab 281:E649–E654
Levin BE 2002 Metabolic sensors: viewing glucosensing neurons from a broader perspective. Physiol Behav 76:397–401
Isse T, Ueta Y, Serino R, Noguchi J, Yamamoto Y, Nomura M, Shibuya I, Lightman SL, Yamashita H 1999 Effects of leptin on fasting-induced inhibition of neuronal nitric oxide synthase mRNA in the paraventricular and supraoptic nuclei of rats. Brain Res 846:229–235
Guo F, Bakal K, Minokoshi Y, Hollenberg AN 2004 Leptin signaling targets the thyrotropin-releasing hormone gene promoter in vivo. Endocrinology 145:2221–2227
Nillni EA, Vaslet C, Harris M, Hollenberg A, Bjorbak C, Flier JS 2000 Leptin regulates prothyrotropin-releasing hormone biosynthesis. Evidence for direct and indirect pathways. J Biol Chem 275:36124–36133
Huo L, Munzberg H, Nillni EA, Bjorbaek C 2004 Role of signal transducer and activator of transcription 3 in regulation of hypothalamic TRH gene expression by leptin. Endocrinology 145:2516–2523
Legradi G, Emerson CH, Ahima RS, Rand WM, Flier JS, Lechan RM 1998 Arcuate nucleus ablation prevents fasting-induced suppression of proTRH mRNA in the hypothalamic paraventricular nucleus. Neuroendocrinology 68:89–97
Perello M, Stuart RC, Rosapour T, Nillni EA The role of leptin in the stimulation of the proTRH neurons through direct and indirect pathways. Program of the 87th Annual Meeting of The Endocrine Society, San Francisco, CA, 2005, p 219 (Abstract P1-209)
Toni R, Jackson IM, Lechan RM 1990 Neuropeptide-Y-immunoreactive innervation of thyrotropin-releasing hormone-synthesizing neurons in the rat hypothalamic paraventricular nucleus. Endocrinology 126:2444–2453
Legradi G, Lechan RM 1998 The arcuate nucleus is the major source for neuropeptide Y-innervation of thyrotropin-releasing hormone neurons in the hypothalamic paraventricular nucleus. Endocrinology 139:3262–3270
Legradi G, Lechan RM 1999 Agouti-related protein containing nerve terminals innervate thyrotropin-releasing hormone neurons in the hypothalamic paraventricular nucleus. Endocrinology 140:3643–3652
Ahima RS, Kelly J, Elmquist JK, Flier JS 1999 Distinct physiologic and neuronal responses to decreased leptin and mild hyperleptinemia. Endocrinology 140:4923–4931
Mizuno TM, Kleopoulos SP, Bergen HT, Roberts JL, Priest CA, Mobbs CV 1998 Hypothalamic pro-opiomelanocortin mRNA is reduced by fasting and in ob/ob and db/db mice, but is stimulated by leptin. Diabetes 47:294–297
Mizuno TM, Mobbs CV 1999 Hypothalamic agouti-related protein messenger ribonucleic acid is inhibited by leptin and stimulated by fasting. Endocrinology 140:814–817
Hahn TM, Breininger JF, Baskin DG, Schwartz MW 1998 Coexpression of Agrp and NPY in fasting-activated hypothalamic neurons. Nat Neurosci 1:271–272
Emerson CH, Cohen 3rd JH, Young RA, Alex S, Fang SL 1990 Gender-related differences of serum thyroxine-binding proteins in the rat. Acta Endocrinol (Copenh) 123:72–78
Cusin I, Zakrzewska KE, Boss O, Muzzin P, Giacobino JP, Ricquier D, Jeanrenaud B, Rohner-Jeanrenaud F 1998 Chronic central leptin infusion enhances insulin-stimulated glucose metabolism and favors the expression of uncoupling proteins. Diabetes 47:1014–1019
Cettour-Rose P, Burger AG, Meier CA, Visser TJ, Rohner-Jeanrenaud F 2002 Central stimulatory effect of leptin on T3 production is mediated by brown adipose tissue type II deiodinase. Am J Physiol Endocrinol Metab 283:E980–E987
Rondeel JM, de Greef WJ, Heide R, Visser TJ 1992 Hypothalamo-hypophysial-thyroid axis in streptozotocin-induced diabetes. Endocrinology 130:216–220
Thomas SA, Palmiter RD 1997 Thermoregulatory and metabolic phenotypes of mice lacking noradrenaline and adrenaline. Nature 387:94–97
Niswender KD, Schwartz MW 2003 Insulin and leptin revisited: adiposity signals with overlapping physiological and intracellular signaling capabilities. Front Neuroendocrinol 24:1–10
Bruning JC, Gautam D, Burks DJ, Gillette J, Schubert M, Orban PC, Klein R, Krone W, Muller-Wieland D, Kahn CR 2000 Role of brain insulin receptor in control of body weight and reproduction. Science 289:2122–2125
Bergen HT, Monkman N, Mobbs CV 1996 Injection with gold thioglucose impairs sensitivity to glucose: evidence that glucose-responsive neurons are important for long-term regulation of body weight. Brain Res 734:332–336
Miller JC, Gnaedinger JM, Rapoport SI 1987 Utilization of plasma fatty acid in rat brain: distribution of [14C]palmitate between oxidative and synthetic pathways. J Neurochem 49:1507–1514
Molina PE, Williams P, Abumrad NN 1997 Histaminergic contribution to the metabolic effects of neuroglucopenia. Am J Physiol 272:R1918–R1924(Csaba Fekete, Praful S. Singru, Edith Sa)