Acute but Not Chronic Central Administration of the Neuropeptide 26RFa (QRFP) Improves Glucose Homeostasis in Obese/Diabetic Mice

Introduction: The aim of the study is to investigate whether acute or chronic central administration of the hypothalamic neuropeptide 26RFa may ameliorate the glycemic control of obese/diabetic mice. Methods: Mice were treated for 4 months with a high-fat (HF) diet and received a single i.c.v. injection of 26RFa (3 µg) or a chronic i.c.v. administration of the peptide during 28 days via osmotic minipumps (25 µg/day). i.p. and oral glucose (GLU) tolerance tests, insulin (INS) tolerance test, glucose-stimulated insulin secretion (GSIS), food/water intake, horizontal/vertical activity, energy expenditure, meal pattern, and whole-body composition were monitored. In addition, 26RFa and GPR103 mRNA expressions as well as plasma 26RFa levels were evaluated by RT-QPCR and radioimmunoassay. Results: Acute administration of 26RFa in HF mice induced a robust antihyperglycemic effect by enhancing INS secretion, whereas chronic administration of the neuropeptide is unable to improve glucose homeostasis in these obese/diabetogenic conditions. By contrast, chronic 26RFa treatment induced an increase of the body weight accompanied with an enhanced food intake and a decreased energy expenditure. Finally, we show that the HF diet does not alter the hypothalamic expression of the 26RFa/GPR103 neuropeptidergic system nor the levels of circulating 26RFa. Conclusion: Our data indicate that the central beneficial effect of 26RFa on glucose homeostasis, by potentiating GSIS, is preserved in HF mice. However, chronic administration of the neuropeptide is unable to balance glycemia in these pathophysiological conditions, suggesting that the hypothalamic 26RFa/GPR103 neuropeptidergic system mainly affects short-term regulation of glucose metabolism.

QPCR and radioimmunoassay. Results: Acute administration of 26RFa in HF mice induced a robust antihyperglycemic effect by enhancing INS secretion, whereas chronic administration of the neuropeptide is unable to improve glucose homeostasis in these obese/diabetogenic conditions. By contrast, chronic 26RFa treatment induced an increase of the body weight accompanied with an enhanced food intake and a decreased energy expenditure. Finally, we show that the HF diet does not alter the hypothalamic expression of the 26RFa/GPR103 neuropeptidergic system nor the levels of circulating 26RFa. Conclusion: Our data indicate that the central beneficial effect of 26RFa on glucose homeostasis, by potentiating GSIS, is preserved in HF mice. However, chronic administration of the neuropeptide is unable to balance glycemia in these pathophysiological conditions, suggesting that the hypothalamic 26RFa/GPR103 neuropeptidergic system mainly affects short-term regulation of glucose metabolism.

Introduction
Accumulated studies during the last decade provided the evidence that glucose (GLU) homeostasis is driven by a complex interplay between the pancreatic islets and the brain [1,2]. This concept is illustrated, for instance, by the fact that the pancreatic islets are richly innervated by both sympathetic and parasympathetic fibers, with the former capable of powerfully inhibiting glucose-stimulated insulin secretion (GSIS) and the latter having the opposite effect [3,4]. The crucial role of the brain in the control of glucose metabolism is supported by the observations that the nuclei of the hypothalamus, i.e., the arcuate nucleus (Arc), the paraventricular nucleus, the ventromedial hypothalamic nucleus (VMN), and the lateral hypothalamic area, all contain specific glucose-sensing neuronal populations that have a key role in initiating the glucose counter-regulatory response to hypo-or hyperglycemia [3,5]. Indeed, photoactivation of VMN SF1 neurons induces a diabetogenic-like effect characterized by hyperglycemia and glucose intolerance partially mediated by inhibition of insulin (INS) secretion and enhanced glucagon release [6]. It was also found that chemogenetic activation of Arc AgRP neurons causes peripheral INS resistance and increased plasma INS [7]. These observations and many others raise the hypothesis that aberrant activity of the hypothalamic glucoregulatory circuits may be involved in the pathogenesis of type 2 diabetes (T2D) and suggest that correcting these underlying defects should normalize glycemia in diabetic animals. Supporting this notion, it was recently reported that leptin is able to normalize glycemia in mice with severe INS deficiency, and that a single i.c.v. injection of FGF1 can induce remission of hyperglycemia in rodent models of T2D for weeks or months [8][9][10]. However, our understanding of the hypothalamic glucoregulatory neurocircuitry and its contribution to T2D pathogenesis is still in its infancy and deserves the identification of novel peptidergic neuronal systems regulating glucose homeostasis within the hypothalamus.
In this context, the neuropeptide 26RFa (also referred to as QRFP) and its receptor GPR103, discovered by us and others, are of particular interest [11][12][13][14][15][16]. Indeed, 26RFa-and GPR103-expressing neurons are primarily localized in the VMN and the lateral hypothalamic, and i.c.v. administration of the neuropeptide stimulates food intake by modulating the NPY/POMC system in the Arc [11,15,[17][18][19][20]. The 26RFa/GPR103 neuropeptidergic system is also involved in the control of glucose homeostasis at the periphery [21][22][23][24]. 26RFa and GPR103 are strongly expressed by the β cells of the pancreatic islets and in the gut [21][22][23]. 26RFa attenuates drastically glucose-induced hyperglycemia by acting as an incretin and by increasing INS sensitivity, and 26RFa mutant mice exhibit an impaired regulation of glucose homeostasis [22,24]. In obese/diabetic mice, a loss of the incretin activity of 26RFa and a marked reduction of its INS-sensitive effect are observed [25]. Interestingly, this 26RFa resistance is associated with a downregulation of the 26RFa receptor in the pancreatic islets and INS target tissues [25].
Finally, we promoted, very recently, the evidence that the 26RFa neuronal system of the hypothalamus plays a key role in the central regulation of glucose homeostasis by lowering glucose-induced hyperglycemia and by mediating central INS-induced peripheral INS secretion [26]. With regard to all of these observations, the aim of the present study was to investigate whether acute or chronic central administration of 26RFa may ameliorate the glycemic control of obese/diabetic mice.

Animals
Male C57Bl/6 mice (Janvier Laboratory, Le Genest-Saint-Isle, France), weighing 22-25 g, were housed at 5 animals per cage, with free access to standard diet (U.A.R., Villemoisson-sur-Orge, France) and tap water. They were kept in a ventilated room at a temperature of 22

Surgery
Mice were anesthetized with isoflurane and placed in a stereotaxic frame. A stainless steel 26-gauge guide cannula (Phymep, Paris, France) was inserted into the right lateral ventricle (0.8 mm lateral to bregma and 2.2 mm ventral to dura mater). A stainless dummy small cap for 26-gauge cannula (Phymep) was inserted to prevent occlusion of the guide cannula and leakage of the cerebrospinal fluid. The cannula placement was secured with dental acrylic cement. At the end of the surgery, animals were treated with a nonsteroidal anti-inflammatory drug, buprenorphine, after guide cannula placement. Following the surgery, all animals were allowed to recover for 2 weeks and were handled 3 times in order to be habituated to handling by researchers. Correct cannula posi-DOI: 10.1159/000522287 tioning was confirmed by histological examination after trypan blue injection.
For chronic i.c.v. infusion, mice were implanted with a cannula to the right lateral ventricle using a brain infusion kit 3 (Alzet ® , DURECT, Cupertino, CA, USA) at stereotaxic coordinates: 0.85 mm lateral, 0.22 mm posterior to bregma, and 2.5 mm below the skull surface. The cannula was connected to a subcutaneous 28-day osmotic minipump (Alzet ® , DURECT, Cupertino, CA, USA) filed with either vehicle (HEPES buffer) or 26RFa (25 µg/ day). After implantation, mice were treated with buprenorphine for 3 days.
Quantitative RT-PCR Total RNA from m-HypoA-59 cells and hypothalami was isolated as previously described [24]. Relative expression of the 26RFa, GPR103, insulin receptor, and IL6 genes was quantified by real-time PCR with appropriate primers (Table 1). β-Actin was used as internal control for normalization. PCR was carried out using Gene Expression Master Mix 2X Assay (Applied Biosystems, Courtaboeuf, France) in an ABI Prism 7900 HT Fast Real-time PCR System (Applied Biosystems). The purity of the PCR products was assessed by dissociation curves. The amount of target cDNA was calculated by the comparative threshold method and expressed by means of the 2 −ΔΔCt method.

Blood Glucose and INS Measurements in Mice
For acute experiments, all metabolic tests were performed on 16-h fasting mice with free access to water, and test substances including 26RFa (3 µg) and INS (10 mIU) were dissolved in HEPES buffer and injected i.c.v. in a 2 µL volume. The injections were done using a Hamilton syringe (outer diameter 0.5 mm; Hamilton Co., Reno, NV, USA) attached to polyethylene tubing. The i.c.v. injections were performed under anesthesia consisting of an i.p. administration of diazepam (5 mg/kg) and ketamine (100 mg/kg). For i.p. glucose tolerance test, mice were treated i.p. with glucose (2 g/kg).
For chronic experiment, all metabolic tests were performed on 6-h fasting mice with free access to water to avoid a too much impact of fasting on the phenotype of obese mice. For i.p. glucose tolerance test and oral glucose tolerance test, mice received 2 g/kg of i.p. or oral glucose, respectively. For INS tolerance test, mice were injected with 0.75 IU/kg of human INS (Eli Lilly, Neuilly-sur-Seine, France).
In all experiments, plasma glucose concentrations were measured from tail vein samplings at various times using an Accuchek Performa Glucometer (Roche Diagnostic, Saint-Egreve, France) or Xpress 2 glucometer (Nova Biomedical, Les Ullis, France) when glycemia was higher than 6 g/L. Plasma INS concentrations (from tail vein samplings) were determined using an ultrasensitive mouse INS AlphaLISA detection kit (cat number AL204C) from PerkinElmer.

Experiments during Chronic i.c.v. Infusion of 26RFa
Eight-week-old mice were submitted to a HF diet for 10 weeks and then divided in two groups: one group that received 26RFa via an osmotic pump for 28 days (25 µg/day) and the other one that received HEPES via the same procedure. After implantation of the pumps, mice were placed in a combined indirect calorimetry system for habituation (PhenoMaster, TSE Systems GmbH, Bad Homburg, Germany). Ten days after pump implantation, food/ water intake, horizontal/vertical activity, and energy expenditure were measured for 5 days. At the end of these calorimetric experiments, mice were placed in individual BioDAQ cages (Research Diets, Inc., New Brunswick, NJ, USA), acclimated for 3 days, and meal pattern was evaluated for 3 days. After this period, mice were placed into standard individual cages until euthanasia. During the 4th and last week of treatment, mice were submitted to metabolic assays. Throughout the treatment, the body weight was measured each day and 6-h fasting glycemia and insulinemia were measured each week. At the end of the treatment, whole-body composition was assessed on vigil animals, using MiniSpec LF110 (Bruker Wissembourg, France), a fast nuclear magnetic resonance method, and mice were euthanized. Tissues and plasma samples were collected. Metabolic biomarkers were measured on plasma samples using IDEXX Catalyst One technology (Laboratories, Inc., Westbrook, ME, USA).
26RFa Radioimmunoassay Quantification of 26RFa in plasma samples was carried out using a specific radioimmunoassay setup in the laboratory [17]. For the radioimmunoassay procedure, each plasma sample was diluted (1:1) in a solution of water/trifluoroacetic acid (99.9:0.1; vol/vol). Diluted plasmas were pumped at a flow rate of 1.5 mL/min through one Sep Pak C18 cartridge. Bound material was eluted with acetonitrile/water/trifluoroacetic acid (50:49.9:0.1; vol/vol/vol), and acetonitrile was evaporated under reduced pressure. Finally, the dried extracts were resuspended in 0.1 M PBS and assayed for 26RFa.

Statistical Analysis
Statistical analysis was performed with GraphPad Prism (version 6). A Kruskal-Wallis test with Dunn's multiple comparison test or ordinary one-way ANOVA with Sidak's multiple comparison was used for comparisons between different groups. ANOVA two ways were used for repeated measures. A post hoc comparison using a Bonferroni, Tukey, or Sidak test was applied according to ANOVA results. All data represent means ± SEM. Statistical significance was set up at p < 0.05.

Effects of Acute i.c.v. Administration of 26RFa on the Glycemic Control of Obese/Diabetic Mice
The HF diet challenge induced a significant (p < 0.01) weight gain right from the first month of treatment (Fig. 1a). Plasma glucose was also significantly increased (p < 0.01) after 1 month of HF diet (Fig. 1b). After 4 months, the HF mice exhibited a fasting glycemia of 2.37 ± 0.17 g/L, whereas that of the SD mice at the same time was 1.65 ± 0.11 g/L (p < 0.01; Fig. 1b). By contrast, plasma 26RFa levels were not modified all along the HF treatment as compared to SD mice (Fig. 1c). However, it is noteworthy that, in the two groups of mice (SD and HF), plasma 26RFa increased gradually with the age of the animals to reach a significancy at the end of the experimental protocol (4 months; p < 0.05) (Fig. 1c).
The impact of i.c.v. administration of 26RFa was investigated on basal glycemia in HF mice. As illustrated in Figure 1d, 26RFa did not affect significantly basal plasma glucose levels during the 90-min period of the test. By contrast, an IPGTT revealed that 26RFa significantly attenuated (p < 0.01) the hyperglycemia induced by an i.p. glucose challenge in HF mice (Fig. 1e). Concurrently, an acute GSIS (AGSIS) test indicated that i.c.v. injection of 26RFa allowed to maintain high level of INS production as compared to the mice that did not receive 26RFa and exhibited a progressively decreased INS production during the test (Fig. 1f). Consequently, the ratio glucose/INS (during the IPGTT) was significantly lower (p < 0.01) in 26RFa-injected mice as compared to HEPES-injected mice (Fig. 1g).
We investigated whether the expression of 26RFa and its receptor was impaired in a diabetogenic environment using in vitro and in vivo complementary approaches. First, we used the hypothalamic neuronal cell line m-Hy-poA-59 that expresses 26RFa and its receptor GPR103 as a model of 26RFa-producing neurons, and we treated these cells with palmitate to mimic conditions of lipotoxicity. Our data revealed that the palmitate treatment did not alter the expression of 26RFa and GPR103 in the m-HypoA-59 neuronal cells (Fig. 2a). The increased expression of IL6 in these cells indicates that palmitate-induced lipotoxicity was effective (Fig. 2a). The hypothalamic expression of 26RFa and GPR103 was also followed during the 4-month HF challenge of the mice. A significant increase of the expression of 26RFa (p < 0.05) and GPR103 (p < 0.05-0.01) was observed 1 and 2 months after the beginning of the protocol in the HF as well as in the SD mice (Fig. 2b, c). However, the HF treatment did not alter the expression of the neuropeptide and its receptor in the hypothalamus as compared to the SD treatment (Fig. 2b,  c). Then, we compared the effects of 26RFa and INS (injected i.c.v.) during a glucose tolerance test in HF mice. 26RFa induced a robust antihyperglycemic effect, whereas the INS-induced attenuation of the glycemia was abolished in HF mice (Fig. 2d). Co-administration of 26RFa and INS induced an antihyperglycemic effect similar in terms of amplitude and kinetic to that of 26RFa alone (Fig. 2d). Concurrently, we evaluated the impact of i.c.v. injection of INS on the hypothalamic expression of 26RFa and that of its receptor (insulin receptor) in HF mice and compared it to that of SD mice. We found that INS slightly increased 26RFa expression in SD mice but not in HF animals (Fig. 2e). By contrast, the hormone does not impact the expression of its own receptor whatever the diet challenge given (Fig. 2e).
A 1 month in vivo protocol was set up to determine the impact of continuous administration of 26RFa on the metabolic and glycemic phenotype of the obese/diabetic mice (Fig. 3a). Chronic i.c.v. administration of 26RFa via Alzet pumps, 10 weeks after the beginning of the HF diet, induced a significant increase in body weight as compared to the mice receiving HEPES (p < 0.001; Fig. 3b). Concurrently, measurement of cumulative food intake for 5 days revealed that the mice treated with 26RFa ate more than the mice receiving the HEPES chronically (p < 0.001; Fig. 3c). This higher food consumption observed in 26RFa-treated mice was associated with a significant increase in the meal size (online suppl. Fig. S1a; see www. karger.com/doi/10.1159/000522287 for all online suppl. material), whereas the meal frequency, duration, and total mealtime were not impaired (online suppl. Fig. S1b-d). The increased food intake observed in 26RFa-treated mice was accompanied by a small increase of water intake (Fig. 3d) and a tendency to decreased energy expenditure (Fig. 3e). Concurrently, the locomotor activity (horizontal and vertical) was assessed, but no significant difference between the HF mice receiving 26RFa and those chronically injected with HEPES was observed (online suppl.   Fig. 3. Effects of chronic i.c.v. administration of 26RFa on the metabolic and glycemic phenotype of the obese/diabetic mice. a Schematic representation of the experimental protocol used to evaluate the impact of 26RFa, injected chronically (25 µg/day), on various metabolic and glycemic parameters during a 4-month HF diet. b Evolution of the body weight of the HF mice for 28 days after the implantation of the Alzet pumps filled with 26RFa (red) or the vehicle (black) (n = 7-9/groups). c-e Evolution of food intake (c), water intake (d), and energy expenditure (e) for 5 days in 26RFatreated mice (red) versus HEPES-treated mice (black) (n = 7-8/ groups). f, g Evolution of basal glycemia (f) and insulinemia (g) in HF mice during the chronic treatment with 26RFa (red) or the vehicle (black) (n = 7-8). h Fat mass/body weight ratio measured after a 28-day chronic treatment with 26RFa (red) or the vehicle (black) (n = 7-8/group). Values represent the mean ± SEM. ***, p < 0.001; ****, p < 0.0001 compared to mice injected with the vehicle. Evaluation of basal glycemia in HEPES-or 26RFatreated mice did not reveal any significant difference between the two groups during the chronic treatment (Fig. 3f). In contrast, evaluation of insulinemia at the same time showed higher (although not significant) plasma INS levels in 26RFa-treated mice (Fig. 3g). Body fat mass also tended to be higher in 26RFa chronically injected mice (Fig. 3h), whereas the biochemical parameters were similar between the 26RFa-treated mice and the HEPES-treated animals ( Table 2).
The impact of chronic i.c.v. administration of 26RFa on the "glycemic" phenotype of the HF mice was investigated using complementary in vivo tests performed during the last week of treatment. A IPGTT revealed a slight increase of glucose-induced hyperglycemia in the 26RFatreated mice (p = 0.07) that was associated with slightly higher plasma INS levels during the first 15 min of the test as compared to HEPES-treated animals (Fig. 4a, b). However, calculation of the glucose/ INS ratio during the first 30 min of the IPGTT did not reveal any difference between the two groups (Fig. 4c). ITT showed that INS tolerance was not significantly altered in 26RFa-injected mice (Fig. 4d). An OGTT revealed that the glycemic profile following the oral glucose load was not impaired in 26RFa-treated mice versus mice receiving HEPES (Fig. 4e). Finally, a Q-PCR experiment indicated that hypothalamic expression of the 26RFa receptor was not impaired by the chronic treatment with its ligand (Fig. 4f).

Discussion/Conclusion
During the last decade, a number of studies revealed that the 26RFa/GPR103 neuropeptidergic system is a key player in the regulation of energy and glucose metabo-lism. Indeed, 26RFa produced by neurons of the hypothalamus regulates feeding behavior, and at the periphery, the neuropeptide produced by the gut and the pancreatic islets regulates glucose homeostasis by acting as an incretin [14,22,27]. However in obese/diabetic mice, the antihyperglycemic effect of 26RFa is significantly blunted due, at least in part, to a decreased expression of its receptor (GPR103) both in beta cells and in INS target tissues such as the adipose tissue and the muscles [25]. More recently, we found that the 26RFa neuronal system of the hypothalamus was also able to promote GSIS and could be considered as a mediator for central INS effects on glucose tolerance [26]. With regard to these recent data, we thought important to evaluate the central effect of the 26RFa on glucose metabolism in a diabetogenic context.
In the present study, we show for the first time that an acute central administration of 26RFa in HF mice is still able to attenuate glucose-induced hyperglycemia by enhancing GSIS. This favorable central effect of 26RFa on INS secretion in this hyperglycemic model is therefore preserved and very similar in terms of amplitude and kinetic to that observed in euglycemic mice [26]. The dose we used for the acute i.c.v. administration of 26RFa (3 µg) is very low compared to that used for peripheral administration in our previous works and does not alter glucose tolerance when administered i.p. [22]. It is thus unlikely that the effects of 26RFa after an i.c.v. administration could be due to an infusion of the neuropeptide to the systemic circulation.
We also show that the obese/diabetogenic status of the mice does not impair the hypothalamic expression of 26RFa and GPR103, contrasting to what we observed at the periphery [25]. Supporting this observation, we also found that the expression of the 26RFa and GPR103 in the neuronal hypothalamic cell line m-HypoA-59 is not al-  tered under lipotoxic culture condition. It thus appears that the hypothalamic 26RFa/GPR103 system is protected from HF-induced defects and allows to maintain the antihyperglycemic effect of 26RFa, whereas the same peptidergic system is profoundly altered at the periphery, resulting in a loss of the incretin effect during of 26RFa. However, we observed that, in these HF mice, central INS administration was not able to promote 26RFa hypothalamic expression, whereas we previously found that, in euglycemic mice, brain INS induced hypothalamic 26RFa expression and secretion during a glucose load [26]. In the same study, we also promoted the evidence that the 26RFa/GPR103 neuronal system mediates the central antihyperglycemic effect of INS [26]. Altogether, these observations support the idea that the inability of INS to trigger the 26RFa/GPR103 system in hyperglycemic condition is an additional consequence of the wellknown central INS resistance observed in obesity/diabetes and probably participates to the glucose intolerance developed by the HF mice.
In addition, while the HFD does not impair the hypothalamic 26RFA/GPR103 system, our data reveal that the age influences the expression of this peptidergic system. Indeed, plasma 26RFa as well as hypothalamic expression of GPR103 and 26RFa is upregulated over time in our protocol in both normal and HFD mice. To our knowledge, it is the first time that evolution of the 26RFa peptidergic system is described over time. By contrast, previous studies reported stable or decreasing hypothalamic expression of other neuropeptides regulating energy metabolism such as NPY or AgRP over time, but the experimental conditions were different with extreme age analysis or long fasting period [28,29]. Because of the limited length of our analysis, we speculate that the observed upregulation of the 26RFa/GPR103 system may correspond to a maturation of this neuropeptidergic system rather than a sign of aging. However, further studies with extreme age analyses are warranted to confirm this hypothesis.
Given the beneficial effects of acute administration of 26RFa on glucose metabolism in HF mice, it was tempting to evaluate the impact of chronic administration of the peptide on metabolic parameters in an obese/hyperglycemic pathophysiological context. By using osmotic minipumps, 26RFa was chronically administered i.c.v. for 28 days and the stability of the peptide was checked all along the protocol by coupling HPLC analysis with mass spectrometry detection (data not shown). In 26RFa-treated mice, a significant increase of the body weight was observed during the experimental protocol that was accompanied with an enhanced food intake and a slight decrease of energy expenditure. Despite this increase of body weight, fasting blood glucose levels of the 26RFa-treated mice were not significantly different as compared to mice that did not receive the peptide. In addition, fasting insulinemia tended to be higher all along the protocol in the 26RFa-treated mice. However, the IPGTT and ITT tests revealed a slightly more pronounced glucose intolerance in 26RFa-treated mice, whereas GSIS was not improved, in contrast to what we observed after an acute administration of 26RFa. Considering the weight gain, our data are in agreement with previous results observed with the elongated form of 26RFa, 43RFa [19]. Indeed, this study reports that central chronic infusion of 43RFa for 13 days is associated with a significant increase of body weight and fat mass. Similarly to what we found, in this study, food intake was enhanced and energy expenditure was decreased because of a reduction of thermogenesis [19]. Considering the metabolic parameters, our results are not consistent with those of Moriya et al. [19], as these authors report higher fasting glucose levels as well as insulinemia in 43RFa-treated mice. However, no data concerning INS sensitivity or GSIS were available in this study. Supporting our findings, a previous study reports that acute administration of another incretin, GLP-1, enhances GSIS, whereas a 3-day i.c.v. infusion of the peptide does not affect INS secretion stimulation [30].
Altogether, these observations indicate that central chronic infusion of 26RFa or 43RFa stimulates appetite, leading to a significant increase of body weight and adiposity. In addition, the two peptides are able to promote fasting INS secretion that may also represent a major determinant for adiposity. Our results also suggest that fasting hyperinsulinemia results from a direct effect of the peptide on INS secretion rather than from an indirect effect simply due to the INS resistance caused by 26RFa-induced adiposity. Indeed, we did not find any significant difference in INS sensitivity during ITT between the 26RFa-treated mice and the control group. Nevertheless, it appears that, in the context of chronic central infusion of 26RFa in HF mice, the sustained orexigenic activity of the neuropeptide system leading to a significant weight gain exceeds the expected benefits of the peptide on glucose metabolism which is altered, especially considering the GSIS results. A decrease of GPR103 activity may be discarded as hypothalamic expression of the 26RFa receptor is not impaired in 26RFa-treated mice. However, further studies are warranted to understand whether the attenuation of the GSIS associated with chronic 26RFa treatment is linked to a central dysfunction of the 26RFa pathway rather than a beta cell failure. An alternative explanation, which has been suggested for GLP-1, would be that brain 26RFa might be specifically involved in the shortterm regulation of glucose metabolism (especially postprandial), whereas its effects on the long-term control DOI: 10.1159/000522287 of glucose homeostasis would be mainly related to changes in body weight [30].
In conclusion, the present study provides the first evidence that in obese/diabetic condition, the 26RFa/ GPR103 neuropeptidergic system is not altered in the hypothalamus, in contrast to the periphery. Indeed, the beneficial effect of 26RFa on glucose homeostasis, by potentiating GSIS, is preserved in HF mice. However, chronic central administration of the neuropeptide is unable to balance glycemia in these pathophysiological conditions, suggesting that the hypothalamic 26RFa/GPR103 neuropeptidergic system mainly affects short-term regulation of glucose metabolism.

Statement of Ethics
Animal manipulations were performed according to the European Community Council Directive of November 24, 1986 (86:609: EEC), and were conducted by authorized investigators. Our Institutional Animal Use and Care Committee (CENOMAX, Agreement of the Ministry of Research No. 54) approved the experimental procedures (Agreement No.: #11752).