Detection of extracellular glucose by GLUT2 contributes to ...

Am J Physiol Endocrinol Metab 298: E1078–E1087, 2010. First published February 23, 2010; doi:10.1152/ajpendo.00737.2009.

Detection of extracellular glucose by GLUT2 contributes to hypothalamic control of food intake Emilie Stolarczyk,1,2,3,4 Christophe Guissard,5 Aurélien Michau,1,2,3,4 Patrick C. Even,6 Alexandra Grosfeld,1,2,3,4 Patricia Serradas,1,2,3,4 Anne Lorsignol,5 Luc Pénicaud,5 Edith Brot-Laroche,1,2,3,4 Armelle Leturque,1,2,3,4 and Maude Le Gall1,2,3,4 1

Centre de Recherche des Cordeliers, Unité Mixte de Recherche (UMR) S872; 2Institut National de la Santé et de la Recherche Médicale, U872; 3Université Pierre et Marie Curie Paris 06, UMR S872; 4Université Paris Descartes Paris 05, UMR S872, Paris; 5UMR 5241 Centre National de la Recherche Scientifique-Université Paul Sabatier, Toulouse; and 6Institut National de Recherche Agronomique, AgroParisTech, UMR 914 Nutrition Physiology and Ingestive Behavior, Centre de Recherche en Nutrition Humaine Ile de France, Paris, France Submitted 11 December 2009; accepted in final form 22 February 2010

Stolarczyk E, Guissard C, Michau A, Even PC, Grosfeld A, Serradas P, Lorsignol A, Pénicaud L, Brot-Laroche E, Leturque A, Le Gall M. Detection of extracellular glucose by GLUT2 contributes to hypothalamic control of food intake. Am J Physiol Endocrinol Metab 298: E1078 –E1087, 2010. First published February 23, 2010; doi:10.1152/ajpendo.00737.2009.—The sugar transporter GLUT2, present in several tissues of the gut-brain axis, has been reported to be involved in the control of food intake. GLUT2 is a sugar transporter sustaining energy production in the cell, but it can also function as a receptor for extracellular glucose. A glucose-signaling pathway is indeed triggered, independently of glucose metabolism, through its large cytoplasmic loop domain. However, the contribution of the receptor function over the transporter function of GLUT2 in the control of food intake remains to be determined. Thus, we generated transgenic mice that express a GLUT2-loop domain, blocking the detection of glucose but leaving GLUT2-dependent glucose transport unaffected. Inhibiting GLUT2-mediated glucose detection augmented daily food intake by a mechanism that increased the meal size but not the number of meals. Peripheral hormones (ghrelin, insulin, leptin) were unaffected, leading to a focus on central aspects of feeding behavior. We found defects in c-Fos activation by glucose in the arcuate nucleus and changes in the amounts of TRH and orexin neuropeptide mRNA, which are relevant to poorly controlled meal size. Our data provide evidence that glucose detection by GLUT2 contributes to the control of food intake by the hypothalamus. The sugar transporter receptor, i.e., “transceptor” GLUT2, may constitute a drug target to treat eating disorders and associated metabolic diseases, particularly by modulating its receptor function without affecting vital sugar provision by its transporter function. glucose transporter 2; sugar sensing; SLC2A2

central and peripheral signals that regulate feeding behavior to adapt nutrient ingestion to energy expenditure. Peripheral hormones secreted by the digestive tract (ghrelin, glucagon-like peptide-1, peptide YY, and cholecystokinin), the pancreas (insulin), and the adipose tissue (leptin, adiponectin) are translated by the hypothalamus into the production of orexigenic [neuropeptide Y (NPY), Agouti-related protein (AgRP), orexin, melanin-concentrating hormone (MCH)] or anorexigenic [proopiomelanocortin (POMC), cocaine- and amphetamine-regulated tran-

THE HYPOTHALAMUS RECEIVES AND INTEGRATES

Address for reprint requests and other correspondence: M. Le Gall, UMR S 872, Centre de Recherche des Cordeliers, 15 rue de l’Ecole de médecine, Paris, F-75006 France (e-mail: [email protected]). E1078

script (CART), thyrotropin-releasing hormone (TRH), corticotropin-releasing hormone (CRH)] peptides (reviewed in Refs. 3 and 10). The resulting arcuate and paraventricular hypothalamic neuronal activity leads to appropriate meal number and size (3). In addition to the integration of hormonal signals, the brain can also directly sense nutrients, and glucose plays a particular role as main energy source for brain cells. The glucostatic theory for the regulation of food intake (37) has postulated that energy needs are monitored by “glucoreceptors.” The control of food intake has since been located in the gut-brain axis (reviewed in Ref. 44), and several glucoreceptors have been suggested. Glucose-sensitive neurons are localized in brain areas involved in energy homeostasis, mainly the hypothalamus and hindbrain (reviewed in Refs. 31, 38, and 41). At least four types of neurons, excited or inhibited by low or high glucose concentrations, can modulate their firing rate in response to variations in glucose concentrations (31, 38, 41). Glucoseinduced excitation of hypothalamic neurons requires glucose phosphorylation, ATP production by glucose oxidation, and subsequent closure of ATP-sensitive potassium (KATP) channels that induces a membrane depolarization (reviewed in Refs. 4, 31, and 41). A similar mechanism triggers insulin secretion in pancreatic ␤-cells (38). GLUT1 and GLUT3 transporters generate glucose fluxes in brain cells; nevertheless, these transporters get saturated at low glucose concentrations (6). By contrast, GLUT2, a high-capacity, high-Km transporter isoform (6), when associated with an efficient glucokinase, sustains fluxes of higher glucose concentrations. GLUT2 is present in brain areas involved in the control of food intake, whereas the exact nature of GLUT2-expressing cells remains unclear since GLUT2 has been identified in neurons, astrocytes, and tanycytes (1, 2, 17, 28, 36). Nevertheless, it is not known whether GLUT2 is present in the glucose-sensitive neurons expressing orexigenic or anorexigenic peptides (26). GLUT2 has been involved in the control of food intake. Indeed, reduced amounts of GLUT2 in the brain have abolished the peripheral insulin response to central glucose (29) and prevented the increase of food intake caused by glucopenia (49). Moreover, in GLUT2-null mice rescued by the pancreatic expression of GLUT1 (ripglut1; glut2⫺/⫺ mice), accelerated feeding initiation and delayed meal termination have increased the daily food intake (5). This occurred in conjunction with abnormal regulation of orexigenic and anorexigenic neuropeptide expression in the hypothalamus (5). GLUT2, glucokinase,

0193-1849/10 $8.00 Copyright © 2010 the American Physiological Society

http://www.ajpendo.org

GLUCOSE TRANSCEPTOR GLUT2 REGULATES FOOD INTAKE

and KATP channels have thus been proposed as major brain actors in glucodetection (41). However, glucokinase and GLUT2 do not always colocalize in brain neurons (26, 32), suggesting an alternative GLUT2-dependent glucose-sensing pathway. Detection mechanisms, independent of KATP channels, have been reported to be involved in central glucose sensing (16, 18, 19). A sodium-glucose cotransporter activity is involved in membrane polarization of glucose-responsive orexin neurons (19, 40). Recently, sweet-taste G protein-coupled receptors, first identified in mammalian taste buds (39), were found in the brain (43). These taste receptors have been involved in the detection of sugar in the gut, where they control hormone secretion (45), gene expression (27, 35), and sugar transport (27, 33). The contribution of sweet-taste receptors in the control of food intake has been studied in terms of reward in mice lacking a functional sweet-taste transduction pathway by genetic knockout of the TRPM5 ion channel (12). However, no changes in body weight and food intake were ever reported in mice lacking components of the sweet-taste pathway (12, 35). Moreover, glucose concentrations found in the brain may be too low to activate sweet-taste G protein-coupled receptors (39). Overall, several membrane-bound ATP-independent glucoreceptors have been identified in the brain, but their physiological relevance and contribution to the control of food intake remain unknown. In addition to being a sugar transporter fueling metabolism, GLUT2 is also an extracellular glucose receptor in hepatocytes, in ␤-pancreatic cells, and in enterocytes (8, 22, 27). The concept of a transporter-receptor, or “transceptor”, allowing nutrient detection is well established in primitive eukaryotes such as yeast (48), but it is more recent in higher eukaryotes (25, 30). In GLUT2-expressing cells, large expression of a peptide corresponding to the GLUT2-cytoplasmic loop blocked GLUT2-dependent glucose detection but preserved GLUT2-dependent glucose transport and metabolism (8, 22, 27). The GLUT2 loop, present in excess, might sequester a partner of endogenous GLUT2 involved in an alternative glucose-signaling pathway independent of glucose metabolism. This dominant negative tool allowed us to distinguish between GLUT2 receptor and transporter functions in vitro. GLUT2-sugar detection-deficient (GLUT2-SDD) mice have been generated as transgenic mice ubiquitously expressing the GLUT2-loop domain (47). GLUT2-SDD mice display a different phenotype to GLUT2-null mice lacking both GLUT2 functions; whereas GLUT2-null mice develop diabetes and die during weaning (21), GLUT2-SDD mice exhibit profound alterations of glucose homeostasis but are not diabetic (47). In GLUT2-SDD mice, glucose is slowly cleared from the blood due to low insulin production by the pancreas and despite massive urinary glucose excretion (47). Defects in pancreatic insulin secretion are associated with a greater number of small islets. However, the severity of the alterations in pancreatic secretion is highly dependent on the level of transgene expression, and one of the transgenic lines was almost preserved from deficient insulin secretion (47). These data demonstrate the physiological importance of the receptor function of GLUT2 in the regulation of glucose homeostasis. The question addressed in this study is to decipher whether, in the absence of GLUT2-mediated sugar detection, the control food intake by the brain is still sensitive to glucose. The AJP-Endocrinol Metab • VOL

E1079

inhibition of GLUT2-mediated glucose detection increases daily food intake by a mechanism involving increased meal size but not meal number. This modification in feeding behavior is correlated with defects in arcuate c-Fos activation in response to glucose and abnormal levels of orexin and TRH neuropeptide mRNA in the fed state. MATERIALS AND METHODS

Animals. GLUT2-SDD mice have been described previously (47); briefly, they expressed a DNA fragment encoding the cytoplasmic loop of the rat GLUT2 (amino acids 237–301), which is driven by an actin promoter (22). Wild-type mice from nontransgenic littermates of a similar genetic background (C57BL/6J crossed with B6CBA) were used as controls. The animals were bred in the transgenic animal facilities of Centre de Recherche des Cordeliers Paris. All experiments were done with 10- to 14-wk-old animals. Animal procedures complied with published recommendations for the use of laboratory animals by the French government. Experiments with mice received approval from the ethics committee of Université Pierre et Marie Curie for animal use (registration no. p3/2008/042). Food intake measurements. Mice were housed in individual cages and adapted to a standard powder diet for 2 days before the experiments in a room controlled for temperature (22 ⫾ 1°C) under a 12:12-h light-dark cycle (lights on 7 AM to 7 PM). Food (23% protein, 51% carbohydrate, 3% lipid; obtained from Dietex, SaintGratien, France) and water were available ad libitum. Daily food intake analysis was performed using metabolic cages from INRAAgroParisTech, INA-PG UMR 914 Research Unit (34), equipped with weighed food cups (sensitivity 0.01 g). Data acquisition was performed continuously at 5-s intervals throughout the day-light cycle and was controlled by a computer running a program developed specifically for the laboratory. The structure of the meal pattern was analyzed; a meal was defined as the period during which a minimum of 0.01 g (detection limit of the scale) was consumed over a 20-s period preceded and followed by an interval of ⱖ15 min with no feeding. Ingestive bouts separated by ⬍15 min were pooled within one single meal. Hourly food intake throughout the 24-h light-dark cycle was established by pooling food intake at 1-h intervals. Food intake was also determined under glucose abundance (glucose 3.6 g/kg body wt ip) in mice that were fasted for 24 h. Food consumption was then recorded every hour, and experiments were performed between 7 AM (lights on) and 1 PM. Body weight and total fat. Body weight and total fat were determined by dual-energy X-ray absorptiometry (Lunar PIXImus mouse densitometer; GE Lunar). Mice were anesthetized with Avertin (0.02 ml/g body wt) before the analysis (47). Hormonal profiles. Blood was collected in the presence of protease inhibitors (Roche). After centrifugation, the plasma samples were stored at ⫺80°C until the hormone assay. Levels of leptin, resistin, adiponectin, insulin, and glucagon were measured in duplicate using the mouse endocrine hormone multiplex panel (Lincoplex Multiplex Assays; Linco-Millipore) and Luminex technology at the core facility of the Saint Antoine hospital in Paris (IRSSA; Inserm IFR65). Insulin levels were determined with ELISA (Rat/Mouse Insulin Kit; LincoMillipore) and ghrelin using RIA assay (Linco). Hypothalamic analysis. Mice fasted for 24 h or fed were euthanized at 7 AM (lights on) by cervical dislocation, and whole brains were rapidly isolated. The hypothalamus was dissected following its clear morphological boundaries and includes lateral and paraventricular nuclei. Resulting tissue samples were immediately frozen in liquid nitrogen and stored at ⫺80°C until RNA extraction or crude membrane preparation. Total RNA from the hypothalamus was extracted with Tri Reagent. Reverse transcription was carried out with 1 ␮g of total RNA and messenger RNA quantified with a Light Cycler system (Roche Molecular Biochemicals I primer; Roche Molecular Biochemicals, Indianapolis, IN). Using ProbeFinder version 2.45

298 • MAY 2010 •

www.ajpendo.org

E1080


(Roche), specific primers were designed for POMC (forward 5=-AGT GCC AGG ACC TCA CCA-3=, reverse 5=-CAG CGA GAG GTC GAG TTT G-3=), CART (forward 5=-CGA GAA GAA GTA CGG CCA AG-3=, reverse 5=-CTG GCC CCT TTC CTC ACT-3=), AgRP (forward 5=-TTT GTC CTC TGA AGC TGT ATG C-3=, reverse 5=-GCA TGA GGT GCC TCC CTA-3=), TRH (forward 5=-AAG ACC TCC AGC GTG TGC-3=, reverse 5=-CCT CCT TCT CCT CCC TTT TG-3=), NPY (forward 5=-AGA AAA CGC CCC CAG AAC-3=, reverse 5=-GTC GGG AGA ACA AGT TTC ATT T-3=), orexin (forward 5=-TTG GAC CAC TGC ACT GAA GA-3=, reverse 5=-CCC AGG GAA CCT TTG TAG AAG-3=), CRH (forward 5=-AGG AGG CAT CCT GAG AGA AGT-3=, reverse 5=-CAT GTT AGG GGC GCT CTC-3=), MCH (forward 5=-TTG GGG ATG AAG AAA ACT CAG-3=, reverse 5=-CCC AGC ATA CAC CTG AGC A-3=), and cyclophilin (forward 5=-ACG CCA CTG TCG CTT TTC-3=, reverse 5=-CTG CAA ACA GCT CGA AGG A-3=). GLUT2 and L19 primers were described previously (47). All primer pairs amplified a single amplicon, as indicated by the unique melting temperature of the PCR product. Moreover, the size of the amplicon was verified by agarose gel electrophoresis. Crude membrane preparations from the hypothalamus were obtained as described previously (47). The membrane proteins were resolved on 12% SDS-PAGE and transferred to nitrocellulose membranes for subsequent immunoblotting with GLUT2 antibody against the intracellular loop produced by Eurogentec (47) and actin (MAB1501 Chemicon). Surgery and intracarotid injections. Surgery and intracarotid injections were performed as described previously (15, 24). Briefly, mice were anesthetized by an ip injection of ketamine-xylazine (100 and 10 mg/kg body wt). A silastic catheter (VWR) was implanted into the left internal carotid artery, with the proximal end directed toward the brain. The catheter was filled with heparin in 0.9% NaCl. The distal end of the catheter was placed under the skin toward the back of the neck, where it was plugged. All mice were allowed to recover from surgical procedures for 1 wk before further experimentation. Mice were handled for several minutes every day to minimize stressrelated effects on c-Fos expression. On the morning of experiment (between 1000 and 1200), freely moving mice were injected for 60 s with 60 ␮l of glucose solution (50 mg/kg body wt) or isotonic NaCl through the catheter. c-Fos immunohistochemistry and quantification. Two hours after glucose or NaCl injection, mice were anesthetized by pentobarbital injection into the catheter. Mice received an intracardiac perfusion of 3.6% formaldehyde-0.2% picric acid solution before brains were removed, postfixed at 4°C overnight, sunk in 30% sucrose for 2 days, and frozen at ⫺50°C in isopentane. The hypothalamus was cut into 30-␮m serial sections on a cryostat (Leica), and free-floating sections were collected in PBS (pH 7.4). After quenching peroxidase activity (0.3% H2O2, 30 min), sections were incubated in 3% normal goat serum (Jackson ImmunoResearch)-PBS-0.3% Triton X-100 (Sigma) for 6 h at room temperature and then in the rabbit antiserum against Fos protein (c-Fos:sc-52, 1/5,000; Santa Cruz Biotechnology) overnight at 4°C. After washing, sections were incubated in biotinylated goat anti-rabbit IgG (1/2,000; The Jackson Laboratories, Bar Harbor, ME) for 45 min at room temperature and finally incubated in streptavidin-peroxidase conjugate (1/1,000; The Jackson Laboratories) for 1 h at room temperature. Peroxidase activity was revealed by diaminobenzidine. c-Fos-immunopositive cells were counted using computerized image analysis (Image J). Statistical analysis. Results are presented as means ⫾ SE. GraphPad and StatEL software packages were used. According to sample distribution, unpaired t-test or Mann-Whitney tests were used to compare two sets of data. ANOVA or Kruskal-Wallis tests were used to compare more than two sets of data. AJP-Endocrinol Metab • VOL

RESULTS

Characterization of GLUT2-SDD mice. GLUT2-SDD transgenic mice have been described previously (47). These mice express the large cytoplasmic GLUT2 loop driven by an actin promoter in all tissues tested, including the hypothalamus (Fig. 1A). These mice were characterized by a defect in sugar sensing restricted to GLUT2-expressing tissues. From the three independent transgenic mouse lines we generated, the GLUT2SDD mouse line presenting only minor defects in insulin secretion (line TgG in Ref. 47) was used in this study to focus on the central control of food intake. The amount of GLUT2 protein was not reduced in the hypothalamus of GLUT2-SDD mice compared with wild-type littermates, allowing the normal transport of glucose (Fig. 1B). Daily food intake. Food intake was measured along the light-dark cycle. Food intake in wild-type mice was maximal immediately after the lights went off (7 PM), and it gradually decreased over the dark period (Fig. 2A). The food intake profile of GLUT2-SDD mice followed a similar pattern (Fig. 2A), but the overall daily food intake was significantly higher (28%) than that of wild-type mice (5.5 ⫾ 0.4 vs. 4.3 ⫾ 0.2 g, P ⬍ 0.03; Fig. 2B). Food intake was unaltered during the day,

Fig. 1. GLUT2-sugar detection-deficient (GLUT2-SDD) mice displayed normal levels of the membrane GLUT2 transporter. A: RT-PCR analysis of GLUT2 loop transgene and L19 (as a control) mRNA in wild-type (WT) and GLUT2-SDD mice. B: full-length GLUT2 protein and actin levels in membranes prepared from hypothalamus of WT or GLUT2-SDD mice.

298 • MAY 2010 •

www.ajpendo.org


E1081

Fig. 2. Inhibition of GLUT2 sugar detection increased the daily food intake through larger meal sizes. Daily food intake (A and B) and meal size (C) were measured in WT and GLUT2-SDD mice during the day (lights on), during the night (lights off), or over 24 h in metabolic cages. Results are means ⫾ SE (n ⫽ 6). Statistical analysis by ANOVA and post hoc pairwise analysis (A) or Student’s t-test (B and C). *P ⬍ 0.05; ns (nonsignificant) comparison between WT and GLUT2-SDD.

but it was significantly greater in GLUT2-SDD mice at night (Fig. 2B). Meal pattern analysis showed that the increase in food intake, measured overnight, was due to an increase in meal size (Fig. 2C), whereas the meal frequency was unaffected (9.1 ⫾ 1.9 meals/12 h in GLUT2-SDD vs. 8.9 ⫾ 1.6 meals/12 h in wild-type mice, P ⫽ 0.84). These results indicate abnormal feeding termination in GLUT2-SDD mice. Although GLUT2-SDD mice presented an increased food intake, they exhibited normal body wt (23.9 ⫾ 1.7 vs. 22.7 ⫾ 0.8 g, P ⫽ 0.65) and normal fat mass (12.93 ⫾ 0.70 vs. 12.97 ⫾ 0.49%, P ⫽ 0.96) as determined by dual-energy X-ray absorptiometry. Regulation of food intake in response to glucose abundance. To evaluate whether the suppression glucose detection was responsible for the food intake disorder observed in GLUT2SDD mice, we measured food intake after a standardized ip injection of glucose in fasted mice to mimic plethora of sugar. AJP-Endocrinol Metab • VOL

Spontaneous 3-h food intake, measured after an overnight fast, was slightly yet not significantly higher in GLUT2SDD than in wild-type mice (1.08 ⫾ 0.13 vs. 0.82 ⫾ 0.08 g/3 h, P ⫽ 0.11; Fig. 3A). An injection of glucose (3.6 g/kg body wt) reduced food intake in wild-type mice by 37% (from 0.82 ⫾ 0.08 to 0.51 ⫾ 0.08 g/3 h, P ⬍ 0.05); however, this decrease in food intake was not observed in GLUT2SDD mice (from 1.08 ⫾ 0.13 to 1.00 ⫾ 0.09 g/3 h, P ⫽ 0.63; Fig. 3A). These findings suggest that GLUT2-SDD mice did not detect the presence of glucose and did not reduce their food intake. Blood glucose and plasma insulin levels were measured after a similar 3.6 g/kg glucose ip injection in fasted mice (Fig. 3, B and C). No differences were observed between wild-type and GLUT2-SDD mice, suggesting that insulin secretion and peripheral glucose clearance were not responsible for the altered feeding behavior observed in GLUT2-SDD mice.

298 • MAY 2010 •

www.ajpendo.org

E1082


Fig. 3. Inhibition of GLUT2 sugar detection prevented food intake control by glucose abundance. Fasted WT or GLUT2-SDD mice received an ip injection of 0.9% NaCl (Gluc ⫺) or 3.6 g/kg glucose (Gluc ⫹) solutions, and spontaneous food intake (A), blood glucose (B), and plasma insulin (C) levels were measured over 2 or 3 h. Results are means ⫾ SE (n ⬎ 10 for WT and n ⬎ 5 GLUT2-SDD). Statistical analysis by ANOVA (P ⬍ 0.001) and Newman and Keuls posttests for A, Kruskal-Wallis for B, and Mann-Whitney for C. P ⬎ 0.05. *P ⬍ 0.05 when comparing Gluc ⫺ and Gluc ⫹ in WT mice. #P ⬍ 0.05 when comparing WT and GLUT2-SDD mice in response to glucose, ns when comparing WT and GLUT2-SDD mice. AUC, area under the curve.

Hormonal profiles. Because hormones other than insulin also drive feeding behavior, we measured the profiles of glucagon, leptin, resistin, and ghrelin in GLUT2-SDD mice. No significant differences in leptin, resistin, and ghrelin plasma levels were detected in wild-type or in GLUT2-SDD mice during a 3-h time course after the ip injection of glucose (data not shown). Therefore, mice were fasted or fed for 24 h to assess possible variations in hormonal profiles over a longer period of time (Fig. 4). Insulin, leptin, and resistin plasma levels were significantly greater (P ⬍ 0.01) in both wild-type and GLUT2-SDD fed

Fig. 4. Inhibition of GLUT2 sugar detection did not affect plasma hormone concentrations measured during the fasted-to-refed transition. Plasma insulin, resistin, leptin, and ghrelin levels in fasted and fed WT and GLUT2-SDD mice. Samples were obtained after 24 h in fasted (white and gray bars) or fed mice (hatched bars). Results are means ⫾ SE (n ⬎ 5). Statistical analysis by ANOVA or Kruskal-Wallis (P ⬍ 0.001) tests and corresponding posttests. **P ⬍ 0.01 for comparisons between fasted and fed mice of the same genotype; #P ⬍ 0.05; ns for comaprisons between WT and GLUT2-SDD in the same metabolic state. P ⬎ 0.05 for comparisons between WT and GLUT2SDD. AJP-Endocrinol Metab • VOL

mice than in corresponding fasted mice. Ghrelin plasma levels were significantly (P ⬍ 0.01) and similarly lower in fed wild-type and GLUT2-SDD mice than in 24-h-fasted mice (Fig. 4). There were no variations in plasma glucagon levels (data not shown). These results suggest that leptin, resistin, and ghrelin levels cannot explain the increase of spontaneous food intake observed in GLUT2-SDD mice during the 24-h lightdark cycle. Neuropeptide profiles. We investigated hypothalamic mRNA levels of neuropeptides involved in the control of feeding behavior to shed light on possible central mechanisms that might trigger the increase in food intake observed in GLUT2-SDD mice. No variations in neuropeptide mRNA levels were observed during the first 3 h following ip injection of glucose (data not shown). The accumulation of neuropeptide mRNA was detectable only over a longer period of time; therefore, we compared mRNA levels in fed and 24-h-fasted mice (Fig. 5). Orexigenic peptides NPY and AgRP, as well as anorexigenic peptides POMC and CART, displayed identical mRNA expression profiles in wild-type and GLUT2-SDD mice. NPY and AgRP mRNA decreased significantly (Fig. 5A), whereas POMC and CART mRNA increased in fed vs. fasted mice (Fig. 5B). No significant differences were observed between wild-type and GLUT2-SDD mice in either the fasted state or the fed state (Fig. 5, A and B). On the contrary, mRNA levels of TRH, CRH, and orexin were affected in GLUT2-SDD mice compared with wild-type mice. In the fasted state, CRH mRNA levels were increased in GLUT2-SDD vs. wild-type mice (Fig. 5D). More interestingly, in the fed state, orexin mRNA levels were increased (Fig. 5C) and TRH mRNA levels decreased (Fig. 5D) in GLUT2-SDD compared with wild-type mice (P ⬍ 0.05). Orexigenic MCH mRNA levels did not change significantly during the time of this experiment (Fig. 5C). Thus, in the fed state, orexin and TRH mRNA levels are dependent on the GLUT2 detection function, and their abnormal regulation, in favor of orexigenic peptides, may contribute to the defective control of food intake and meal size in GLUT2-SDD mice. c-Fos activation. To assess whether brain activity was involved in the food intake disorders observed in GLUT2-SDD mice, we measured c-Fos activation in various brain areas 2 h after a cerebral load of glucose (Fig. 6). The injection of a glucose solution (50 mg/kg body wt) into the carotid artery toward the brain generates a local hyperglycemia without changes in the peripheral blood concentration (15). We quan-

298 • MAY 2010 •

www.ajpendo.org


tified c-Fos protein levels by immunodetection in brain slices through the arcuate nucleus, since this GLUT2-positive nucleus is known to play a role in brain glucose sensing (28). Basal c-Fos activation levels in the arcuate nucleus (ARC) 2 h after saline injection were higher in GLUT2-SDD than in wild-type mice (Fig. 6A). In another GLUT2-positive nucleus, i.e., the paraventricular nucleus of the thalamus, basal c-Fos activation was also higher in GLUT2-SDD than in wild-type

E1083

mice (Fig. 6A). c-Fos activation in the somatosensory cortex, an area devoid of GLUT2, was identical in wild-type and GLUT2-SDD mice. Thus, basal increase in c-Fos activation occurred only in GLUT2-positive areas (Fig. 6A). Cerebral hyperglycemia induced a significant increase in c-Fos activation in the ARC of wild-type mice, whereas it was reduced in GLUT2-SDD mice (Fig. 6B). This suggests that glucose detection in the ARC is defective, possibly contributing to the poor control of food intake in GLUT2-SDD mice. DISCUSSION

The inhibition of GLUT2-mediated sugar detection increased food intake in GLUT2-SDD mice (line TgG) by enlarging meal sizes without changes in meal frequency. The decision to stop eating is delayed in conjunction with defects in arcuate c-Fos activation in response to glucose and changes in orexin and TRH neuropeptide mRNA levels in the hypothalamus of fed mice. Thus, GLUT2 receptor function, independent of sugar transport and metabolism, is involved in controlling feeding behavior. GLUT2 plays a key role in glucodetection, which controls the feeding behavior in mice. Indeed, daily food intake is similarly increased in GLUT2-SDD (line TgG) and in GLUT2null rescued (ripglut1;glut2⫺/⫺) mice (5), and in both mouse models, provision of glucose failed to reduce food intake. GLUT2-dependent sugar transport (except in pancreatic ␤-cells) and sugar detection were invalidated in GLUT2-null mice; by contrast, GLUT2-SDD mice lack only the receptor function, and thus sugar transport was still able to trigger metabolic signaling upon glucose addition. Thus, the regulation of feeding behavior appeared to be reliant mainly on GLUT2 receptor function. GLUT2-mediated sugar detection affected meal size rather than meal frequency. Therefore, the decision to stop eating, i.e., the occurrence of satiation, may have been linked to the concentration of extracellular glucose detected by the GLUT2 receptor. This is an important point because it may be easier to target a membrane-bound receptor to modulate behavior than to alter a metabolic pathway. Despite increased food intake, GLUT2-SDD mice (line TgG) exhibited normal body weight and fat mass. We have previ˙ O 2, ously measured, by indirect calorimetry, heat production, V ˙ CO2, and respiratory quotient in the same GLUT2-SDD TgG V line (47). In response to a mixed meal, GLUT2-SDD mice (line TgG) display an increased lipid oxidation over glucose oxidation, and a massive urinary glucose loss contributes to get rid

Fig. 5. Inhibition of GLUT2 sugar detection modified regulation of hypothalamic neuropeptide mRNA levels from second-order neurons during the fasted-to-refed transition. Neuropeptide mRNAs were measured by quantitative RT-PCR in hypothalamus of WT (white bars) or GLUT2-SDD (gray bars) mice, which were fasted for 24 h (white and gray bars) or fed (hatched bars). A: orexigenic neuropeptide Y (NPY) and Agouti-related protein (AgRP). B: anorexigenic proopiomelanocortin (POMC) and cocaine- and amphetamineregulated transcript (CART). C: orexigenic orexin and melanin-concentrating hormone (MCH). D: anorexigenic corticotropin-releasing hormone (CRH) and thyrotropin-releasing hormone (TRH). Results are means ⫾ SE (n ⬎ 8). Statistical analysis by Kruskal-Wallis (P ⬍ 0.001) tests and the corresponding posttests. **P ⬍ 0.01 for comparisons between fasted and fed mice of the same genotype. #P ⬍ 0.05, ns for comparisons between WT and GLUT2-SDD in the same metabolic state. P ⬎ 0.05 for comparisons between WT and GLUT2SDD in the same metabolic state. AJP-Endocrinol Metab • VOL

298 • MAY 2010 •

www.ajpendo.org

E1084


Fig. 6. Inhibition of GLUT2 sugar detection decreased c-Fos activation in response to intracarotid glucose infusion in the hypothalamic arcuate nucleus. A, left: representative photomicrographs of arcuate nucleus (ARC), paraventricular nucleus of the thalamus (PVT), and somatosensorial cortex (SSC) sections immunostained with c-Fos of fed WT or GLUT2-SDD mice. A, right: quantification of the c-Fos-positive cells in the ARC, PVT, and SSC. B, left: representative photomicrographs of ARC sections immunostained with c-Fos of WT or GLUT2-SDD mice 2 h after intracarotid glucose infusion (50 mg/kg body wt). B, right: quantification of the c-Fos-positive cells in the ARC. Results are expressed as the number of c-Fospositive cells/␮m2, means ⫾ SE (n ⬎ 4 mice). Statistical analysis by unpaired Student’s t-test. *P ⬍ 0.05, ns for comparisons between WT and GLUT2-SDD. #P ⬍ 0.05 for comparisons between NaCl and glucose infusions. P ⬎ 0.05 for comparisons between WT and GLUT2-SDD.

of excess glucose and thus reduced fat deposits (47). GLUT2 is also proposed to control food intake in humans. Indeed, individuals with a GLUT2 allelic variant (Thr110Ile) from two separate Canadian populations have a higher daily intake of sugars (13). Nevertheless, it remains to be established whether functional defects are associated with this variant form of GLUT2. Insulin, glucagon-like peptide-1 (GLP-1), leptin, and cholecystokinin (CCK) are satiety hormones that decrease food intake by modulating the function of neurotransmitter systems and neuronal circuitry (14, 42, 50). Rescue of GLUT2-null mice by GLUT1 expression in pancreatic ␤-cells preserves the second phase of insulin secretion (20), suggesting that the first phase of insulin secretion is involved in controlling food intake. The first phase of insulin secretion inhibits hepatic glucose production (9); therefore, we can speculate that glucose-inhibited neurons are activated via a decreased blood AJP-Endocrinol Metab • VOL

glucose level. To circumvent the contribution of peripheral insulin secretion to feeding behavior, we used, in the present study, the GLUT2-SDD transgenic mouse line in which insulin response to intraperitoneal glucose was unaffected (line TgG; Ref. 47). GLP-1 is also a sugar-dependent satiation signal (50). It is secreted during the meal by enteroendocrine L cells, and its secretion may be mediated by GLUT2 signaling. Indeed, a defect in GLP-1 secretion in response to a sugar load has been reported in rescued GLUT2-null mice (7). This may be relevant to the abnormal feeding termination in GLUT2-null mice (5). However, gut hormone secretion is not induced during intraperitoneal injections of glucose (10) and thus could not contribute to the defective feeding response observed in GLUT2-SDD mice. Plasma ghrelin and leptin levels were not affected in GLUT2-SDD mice (line TgG), which is consistent with the observation that stomach and adipose tissue do not express GLUT2. The gut hormone CCK is known to induce

298 • MAY 2010 •

www.ajpendo.org


satiation by decreasing meal size and meal number (42). CCK was not suspected in this study because CCK secretion is not induced by sugar, but it is induced by meals containing protein and fat. Thus, the GLUT2-SDD mouse TgG line allowed us to investigate the role of central inhibition of sugar detection on the control of food intake with only minor contribution of the peripheral hormones. The amounts of orexigenic (NPY, AgRP) and anorexigenic (POMC, CART) neuropeptides in fasted and fed conditions were similar in GLUT2-SDD and wild-type mice. By contrast, mRNA levels of these neuropeptides were altered significantly in rescued GLUT2-null mice (5), suggesting that glucose transport via GLUT2 is required for their regulated expression. Accordingly, mice with heterozygous invalidation of the glucokinase gene also increased NPY and reduced POMC gene expression (51), confirming that mRNA levels of these neuropeptides are regulated by glucose metabolism. Glucose metabolism reflects various pathways, i.e., glycolysis, oxidative phosphorylation, pentose pathway, hexosamine pathway, and AMP-activated protein kinase phosphorylation. It has been reported, in a clonal hypothalamic neuronal cell line, that inhibition of AgRP expression by glucose is mediated by metabolites provided through glycolysis but not oxidative phosphorylation (11). The molecular regulation by glucose metabolism of other neuropeptides from first-order neurons remains to be determined. We demonstrate in this study that sugar detection by GLUT2 contributes to control orexin, CRH, and TRH mRNA levels independently of intracellular glucose fluxes. In the fasted state, mRNA levels of the anorexigenic CRH neuropeptide are increased in GLUT2-SDD mice, suggesting that detection of sugar concentration by GLUT2 inhibits its expression. In the fed state, increased amounts of the orexigenic neuropeptide orexin, together with reduced amounts of anorexigenic TRH, in fed GLUT2-SDD mice may sustain increased food intake. The c-Fos activation in response to glucose was significantly lower in GLUT2-SDD mice than in wild-type mice, suggesting that the arcuate nucleus fails to detect glucose appropriately. This occurred despite adequate amounts of POMC, CART, NPY, or AgRP mRNA. These data indicate that poorly stimulated arcuate neurons cannot transmit glucose signals to the neurons of the lateral and paraventricular nuclei in the hypothalamus. Accordingly, the altered expression of TRH and orexin observed in GLUT2-SDD mice may be due to neuronal blindness in response to glucose. We observed that the basal c-Fos activation was higher in the arcuate nucleus of the hypothalamus and in the paraventricular nucleus of the thalamus from GLUT2-SDD mice than from wild-type mice. This was also reported in the ventromedial hypothalamic nucleus of rescued GLUT2-null mice (36). Feeding increased lateral periarcuate c-Fos immunoreactivity in wild-type mice, particularly in POMC-positive neurons (46). Both GLUT2-SDD (line TgG) and GLUT2-null mice exhibit elevated food intake, resulting in the increased c-Fos signal observed in the basal ad libitum fed state. Moreover, basal c-Fos activation in GLUT2-negative regions such as the somatosensorial cortex remained unaffected in GLUT2-SDD mice. This suggests that feeding-induced c-Fos immunoreactivity is dependent on GLUT2 sugar detection. To characterize the molecular mechanisms underlying the blocking effect of the GLUT2 loop peptide on glucose signalAJP-Endocrinol Metab • VOL

E1085

ing triggered by the GLUT2 receptor, we identified karyopherin-␣2 as a partner of endogenous GLUT2 loop (8, 23). Moreover, we demonstrated that karyopherin-␣2 was involved in glucose signaling by importing proteins, e.g., components of the glucose-sensitive transcription machinery, into the nucleus (8, 23). In cells or tissues devoid of GLUT2, glucose detection occurred through mechanisms that are not affected by expression of GLUT2 loop. Accordingly, GLUT2-SDD mice, expressing GLUT2-loop peptide, displayed normal insulin sensitivity in tissues expressing GLUT4 (muscle and adipose tissues) but not GLUT2 (47). In the hypothalamus, GLUT2 was recovered in neurons, astrocytes, or tanycytes (1, 2, 17, 28, 36). A hypothesis is that glucose-sensitive neurons involved in the control of food intake express GLUT2; in that case, expression of the GLUT2 loop will directly block sugar-sensitive mechanisms (gene transcription or exocytosis). Another hypothesis is that those glucose-sensitive neurons do not express GLUT2 and might receive messengers from close GLUT2-expressing cells (e.g., astrocytes) to control glucose-sensitive actions. In favor of this second hypothesis, reexpression of GLUT2 in glial cells is sufficient to restore the response to hypoglycemia in GLUT2-null mice (36). Whatever the exact location of GLUT2 within the brain cells, food intake is affected by expression of the GLUT2 loop and thus appears dependent on GLUT2 sugar detection. In conclusion, our findings provide evidence that GLUT2dependent glucose detection, in addition to glucose transport, plays a major role in the control of food intake by the hypothalamus. GLUT2 detects and signals glucose to adapt to the meal size. In the arcuate nucleus of the hypothalamus, GLUT2 sugar detection is required for c-Fos activation but not for neuropeptide mRNA expression. However, GLUT2 sugar detection activates anorexic TRH production in the paraventricular nucleus and reduces production of the orexigenic peptide orexin in the lateral hypothalamus. The dual function of GLUT2 may fulfill the need for detecting acute extracellular vs. intracellular glucose concentrations. Modulating the receptor function without affecting vital sugar provision by the transporter may be a strategy to control feeding behavior. Thus, the GLUT2 transceptor constitutes a potential drug target to treat eating disorders and associated metabolic diseases. ACKOWLEDGMENTS We thank V. Frochot, A. Houllier, A. Benkouhi (UMR S 872), and the core facility of the Saint Antoine hospital in Paris (IRSSA, Inserm IFR65) for hormone assays. We are grateful to P. Guillou (Toulouse) for mouse microsurgery. Present address of E. Stolarczyk: Obesity & Metabolic Medicine Group, Division of Reproduction & Endocrinology, King’s College London, St. Thomas Hospital, London SE1 7EH, UK. Present address of L. Pénicaud: UMR 5548, CNRS-Université de Bourgogne, Dijon, F-21000, France. GRANTS Institut National de la Santé et de la Recherche Médicale, Centre National de la Recherche Scientifique, University Pierre and Marie Curie, and grants from Ministère de la Recherche, Fondation pour la Recherche Médicale, Programme National pour la Recherche sur le Diabète, and ANR Nutrisens (PNRA 004) helped to finance this study. DISCLOSURES The authors have nothing to disclose.

298 • MAY 2010 •

www.ajpendo.org

E1086


REFERENCES 1. Arluison M, Quignon M, Nguyen P, Thorens B, Leloup C, Penicaud L. Distribution and anatomical localization of the glucose transporter 2 (GLUT2) in the adult rat brain—an immunohistochemical study. J Chem Neuroanat 28: 117–136, 2004. 2. Arluison M, Quignon M, Thorens B, Leloup C, Penicaud L. Immunocytochemical localization of the glucose transporter 2 (GLUT2) in the adult rat brain. II. Electron microscopic study. J Chem Neuroanat 28: 137–146, 2004. 3. Arora S, Anubhuti. Role of neuropeptides in appetite regulation and obesity—a review. Neuropeptides 40: 375–401, 2006. 4. Ashford ML, Boden PR, Treherne JM. Glucose-induced excitation of hypothalamic neurones is mediated by ATP-sensitive K⫹ channels. Pflugers Arch 415: 479 –483, 1990. 5. Bady I, Marty N, Dallaporta M, Emery M, Gyger J, Tarussio D, Foretz M, Thorens B. Evidence from glut2-null mice that glucose is a critical physiological regulator of feeding. Diabetes 55: 988 –995, 2006. 6. Bell GI, Kayano T, Buse JB, Burant CF, Takeda J, Lin D, Fukumoto H, Seino S. Molecular biology of mammalian glucose transporters. Diabetes Care 13: 198 –208, 1990. 7. Cani PD, Holst JJ, Drucker DJ, Delzenne NM, Thorens B, Burcelin R, Knauf C. GLUT2 and the incretin receptors are involved in glucoseinduced incretin secretion. Mol Cell Endocrinol 276: 18 –23, 2007. 8. Cassany A, Guillemain G, Klein C, Dalet V, Brot-Laroche E, Leturque A. A karyopherin alpha2 nuclear transport pathway is regulated by glucose in hepatic and pancreatic cells. Traffic 5: 10 –19, 2004. 9. Caumo A, Luzi L. First-phase insulin secretion: does it exist in real life? Considerations on shape and function. Am J Physiol Endocrinol Metab 287: E371–E385, 2004. 10. Chaudhri OB, Salem V, Murphy KG, Bloom SR. Gastrointestinal satiety signals. Annu Rev Physiol 70: 239 –255, 2008. 11. Cheng H, Isoda F, Belsham DD, Mobbs CV. Inhibition of agouti-related peptide expression by glucose in a clonal hypothalamic neuronal cell line is mediated by glycolysis, not oxidative phosphorylation. Endocrinology 149: 703–710, 2008. 12. de Araujo IE, Oliveira-Maia AJ, Sotnikova TD, Gainetdinov RR, Caron MG, Nicolelis MA, Simon SA. Food reward in the absence of taste receptor signaling. Neuron 57: 930 –941, 2008. 13. Eny KM, Wolever TM, Fontaine-Bisson B, El-Sohemy A. Genetic variant in the glucose transporter type 2 is associated with higher intakes of sugars in two distinct populations. Physiol Genomics 33: 355–360, 2008. 14. Figlewicz DP, Benoit SC. Insulin, leptin, and food reward: update 2008. Am J Physiol Regul Integr Comp Physiol 296: R9 –R19, 2009. 15. Fioramonti X, Contie S, Song Z, Routh VH, Lorsignol A, Penicaud L. Characterization of glucosensing neuron subpopulations in the arcuate nucleus: integration in neuropeptide Y and pro-opio melanocortin networks? Diabetes 56: 1219 –1227, 2007. 16. Fioramonti X, Lorsignol A, Taupignon A, Penicaud L. A new ATPsensitive K⫹ channel-independent mechanism is involved in glucoseexcited neurons of mouse arcuate nucleus. Diabetes 53: 2767–2775, 2004. 17. García MA, Millán C, Balmaceda-Aguilera C, Castro T, Pastor P, Montecinos H, Reinicke K, Zúñiga F, Vera JC, Oñate SA, Nualart F. Hypothalamic ependymal-glial cells express the glucose transporter GLUT2, a protein involved in glucose sensing. J Neurochem 86: 709–724, 2003. 18. Gonzalez JA, Jensen LT, Fugger L, Burdakov D. Metabolism-independent sugar sensing in central orexin neurons. Diabetes 57: 2569 –2576, 2008. 19. Gonzalez JA, Reimann F, Burdakov D. Dissociation between sensing and metabolism of glucose in sugar sensing neurones. J Physiol 587: 41–48, 2009. 20. Guillam MT, Dupraz P, Thorens B. Glucose uptake, utilization, and signaling in GLUT2-null islets. Diabetes 49: 1485–1491, 2000. 21. Guillam MT, Hummler E, Schaerer E, Yeh JI, Birnbaum MJ, Beermann F, Schmidt A, Deriaz N, Thorens B. Early diabetes and abnormal postnatal pancreatic islet development in mice lacking Glut-2. Nat Genet 17: 327–330, 1997. 22. Guillemain G, Loizeau M, Pincon-Raymond M, Girard J, Leturque A. The large intracytoplasmic loop of the glucose transporter GLUT2 is involved in glucose signaling in hepatic cells. J Cell Sci 113: 841–847, 2000. 23. Guillemain G, Munoz-Alonso MJ, Cassany A, Loizeau M, Faussat AM, Burnol AF, Leturque A. Karyopherin alpha2: a control step of glucose-sensitive gene expression in hepatic cells. Biochem J 364: 201–209, 2002. AJP-Endocrinol Metab • VOL

24. Guillod-Maximin E, Lorsignol A, Alquier T, Penicaud L. Acute intracarotid glucose injection towards the brain induces specific c-fos activation in hypothalamic nuclei: involvement of astrocytes in cerebral glucosesensing in rats. J Neuroendocrinol 16: 464 –471, 2004. 25. Hundal HS, Taylor PM. Amino acid transceptors: gate keepers of nutrient exchange and regulators of nutrient signaling. Am J Physiol Endocrinol Metab 296: E603–E613, 2009. 26. Kang L, Routh VH, Kuzhikandathil EV, Gaspers LD, Levin BE. Physiological and molecular characteristics of rat hypothalamic ventromedial nucleus glucosensing neurons. Diabetes 53: 549 –559, 2004. 27. Le Gall M, Tobin V, Stolarczyk E, Dalet V, Leturque A, Brot-Laroche E. Sugar sensing by enterocytes combines polarity, membrane bound detectors and sugar metabolism. J Cell Physiol 213: 834 –843, 2007. 28. Leloup C, Arluison M, Lepetit N, Cartier N, Marfaing-Jallat P, Ferre P, Penicaud L. Glucose transporter 2 (GLUT 2): expression in specific brain nuclei. Brain Res 638: 221–226, 1994. 29. Leloup C, Orosco M, Serradas P, Nicolaidis S, Penicaud L. Specific inhibition of GLUT2 in arcuate nucleus by antisense oligonucleotides suppresses nervous control of insulin secretion. Brain Res Mol Brain Res 57: 275–280, 1998. 30. Leturque A, Brot-Laroche E, Le Gall M. GLUT2 mutations, translocation, and receptor function in diet sugar managing. Am J Physiol Endocrinol Metab 296: E985–E992, 2009. 31. Levin BE, Routh VH, Kang L, Sanders NM, Dunn-Meynell AA. Neuronal glucosensing: what do we know after 50 years? Diabetes 53: 2521–2528, 2004. 32. Li B, Xi X, Roane DS, Ryan DH, Martin RJ. Distribution of glucokinase, glucose transporter GLUT2, sulfonylurea receptor-1, glucagon-like peptide-1 receptor and neuropeptide Y messenger RNAs in rat brain by quantitative real time RT-PCR. Brain Res Mol Brain Res 113: 139–142, 2003. 33. Mace OJ, Affleck J, Patel N, Kellett GL. Sweet taste receptors in rat small intestine stimulate glucose absorption through apical GLUT2. J Physiol 582: 379 –392, 2007. 34. Makarios-Lahham L, Roseau SM, Fromentin G, Tome D, Even PC. Rats free to select between pure protein and a fat-carbohydrate mix ingest high-protein mixed meals during the dark period and protein meals during the light period. J Nutr 134: 618 –624, 2004. 35. Margolskee RF, Dyer J, Kokrashvili Z, Salmon KS, Ilegems E, Daly K, Maillet EL, Ninomiya Y, Mosinger B, Shirazi-Beechey SP. T1R3 and gustducin in gut sense sugars to regulate expression of Na⫹-glucose cotransporter 1. Proc Natl Acad Sci USA 104: 15075–15080, 2007. 36. Marty N, Dallaporta M, Foretz M, Emery M, Tarussio D, Bady I, Binnert C, Beermann F, Thorens B. Regulation of glucagon secretion by glucose transporter type 2 (glut2) and astrocyte-dependent glucose sensors. J Clin Invest 115: 3545–3553, 2005. 37. Mayer J. Regulation of energy intake and the body weight: the glucostatic theory and the lipostatic hypothesis. Ann NY Acad Sci 63: 15–43, 1955. 38. Mountjoy PD, Rutter GA. Glucose sensing by hypothalamic neurones and pancreatic islet cells: AMPle evidence for common mechanisms? Exp Physiol 92: 311–319, 2007. 39. Nelson G, Hoon MA, Chandrashekar J, Zhang Y, Ryba NJ, Zuker CS. Mammalian sweet taste receptors. Cell 106: 381–390, 2001. 40. O’Malley D, Reimann F, Simpson AK, Gribble FM. Sodium-coupled glucose cotransporters contribute to hypothalamic glucose sensing. Diabetes 55: 3381–3386, 2006. 41. Penicaud L, Leloup C, Fioramonti X, Lorsignol A, Benani A. Brain glucose sensing: a subtle mechanism. Curr Opin Clin Nutr Metab Care 9: 458 –462, 2006. 42. Raybould HE. Mechanisms of CCK signaling from gut to brain. Curr Opin Pharmacol 7: 570 –574, 2007. 43. Ren X, Zhou L, Terwilliger R, Newton SS, de Araujo IE. Sweet taste signaling functions as a hypothalamic glucose sensor. Front Integr Neurosci 3: 12, 2009. 44. Romijn JA, Corssmit EP, Havekes LM, Pijl H. Gut-brain axis. Curr Opin Clin Nutr Metab Care 11: 518 –521, 2008. 45. Rozengurt N, Wu SV, Chen MC, Huang C, Sternini C, Rozengurt E. Colocalization of the ␣-subunit of gustducin with PYY and GLP-1 in L cells of human colon. Am J Physiol Gastrointest Liver Physiol 291: G792–G802, 2006. 46. Shu IW, Lindenberg DL, Mizuno TM, Roberts JL, Mobbs CV. The fatty acid synthase inhibitor cerulenin and feeding, like leptin, activate hypothalamic proopiomelanocortin (POMC) neurons. Brain Res 985: 1–12, 2003.

298 • MAY 2010 •

www.ajpendo.org

GLUCOSE TRANSCEPTOR GLUT2 REGULATES FOOD INTAKE 47. Stolarczyk E, Le Gall M, Even P, Houllier A, Serradas P, BrotLaroche E, Leturque A. Loss of sugar detection by GLUT2 affects glucose homeostasis in mice. PLoS One 2: e1288, 2007. 48. Thevelein JM, Voordeckers K. Functioning and evolutionary significance of nutrient transceptors. Mol Biol Evol 26: 2407–2414, 2009. 49. Wan HZ, Hulsey MG, Martin RJ. Intracerebroventricular administration of antisense oligodeoxynucleotide against GLUT2 glucose transporter

AJP-Endocrinol Metab • VOL

E1087

mRNA reduces food intake, body weight change and glucoprivic feeding response in rats. J Nutr 128: 287–291, 1998. 50. Williams DL. Minireview: finding the sweet spot: peripheral versus central glucagon-like peptide 1 action in feeding and glucose homeostasis. Endocrinology 150: 2997–3001, 2009. 51. Yang XJ, Mastaitis J, Mizuno T, Mobbs CV. Glucokinase regulates reproductive function, glucocorticoid secretion, food intake, and hypothalamic gene expression. Endocrinology 148: 1928 –1932, 2007.

298 • MAY 2010 •

www.ajpendo.org