Leptin Regulation of the Anorexic Response to Glucagon-Like Peptide ...

3 downloads 33 Views 243KB Size Report
peritoneal injection of either GLP-1 or Exendin-4 (Ex4; a. GLP-1 receptor agonist). Leptin pretreatment strongly enhanced anorexia and weight loss induced by ...
Original Article Leptin Regulation of the Anorexic Response to Glucagon-Like Peptide-1 Receptor Stimulation Diana L. Williams,1 Denis G. Baskin,1,2,3 and Michael W. Schwartz1

Leptin reduces food intake in part by enhancing satiety responses to gastrointestinal signals produced in response to food consumption. Glucagon-like peptide 1 (GLP-1), secreted by the intestine when nutrients enter the gut, is one such putative satiety signal. To investigate whether leptin enhances the anorexic effects of GLP-1, rats received either saline or a subthreshold dose of leptin before intraperitoneal injection of either GLP-1 or Exendin-4 (Ex4; a GLP-1 receptor agonist). Leptin pretreatment strongly enhanced anorexia and weight loss induced by GLP-1 or Ex4 over 24 h. Conversely, fasting attenuated the anorexic response to GLP-1 or Ex4 treatment via a leptin-dependent mechanism, as demonstrated by our finding that the effect of fasting was reversed by physiological leptin replacement. As expected, Ex4 induced expression of c-Fos protein, a marker of neuronal activation, in hindbrain areas that process afferent input from satiety signals, including the nucleus of the solitary tract and area postrema. Unexpectedly, leptin pretreatment blocked this response. These findings identify physiological variation of plasma leptin levels as a potent regulator of GLP-1 receptor-mediated food intake suppression and suggest that the underlying mechanism is distinct from that which mediates interactions between leptin and other satiety signals. Diabetes 55: 3387–3393, 2006

G

lucagon-like peptide 1 (GLP-1), a product of the preproglucagon gene, is secreted by L-cells in the distal hindgut in response to nutrient ingestion (1). In addition to its incretin effect, GLP-1 is hypothesized to function as a “satiety signal,” promoting reduced food intake and meal termination. Peripheral administration of GLP-1 or long-acting GLP-1 receptor (GLP-1-R) agonists, such as Exendin-4 (Ex4), reduce blood glucose and food intake in rodents and humans, and chronic treatment results in loss of body weight (2– 4).

From the 1Division of Metabolism, Endocrinology, and Nutrition, Department of Medicine, University of Washington, Seattle, Washington; the 2VA Puget Sound Health Care System, Seattle, Washington; and the 3Department of Biological Structure, University of Washington, Seattle, Washington. Address correspondence and reprint requests to Diana L. Williams, University of Washington, Harborview Medical Center, 325 9th Ave., Box 359675, Seattle, WA 98104. E-mail: [email protected]. Received for publication 25 April 2006 and accepted in revised form 22 August 2006. M.W.S. is a member on an advisory panel for or a committee of Amylin, Abbott, and Takeda Pharmaceuticals and has received consulting fees from Phenomix, Merck, Cypress Bioscience, Bristol Myers Squibb, and Amgen. CCK, cholecystokinin; CNS, central nervous system; c-FLI, c-Fos–like immunoreactivity; Ex4, Exendin-4; GLP-1, glucagon-like peptide 1; GLP-1-R, GLP-1 receptor; NTS, nucleus of the solitary tract. DOI: 10.2337/db06-0558 © 2006 by the American Diabetes Association. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

DIABETES, VOL. 55, DECEMBER 2006

Among medications that have been approved for the treatment of type 2 diabetes in humans, Ex4 is unique in its capacity to induce weight loss while improving blood glucose control (3,5,6). GLP-1-Rs are expressed in a variety of peripheral tissues, including vagal afferent fibers (7). Because subdiaphragmatic vagotomy prevents GLP-1–induced anorexia in rats (8) and capsaicin treatment blocks Ex4-induced intake suppression in mice (9), GLP-1 may reduce intake by activating vagal afferent fibers that terminate primarily in the hindbrain nucleus of the solitary tract (NTS), as does cholecystokinin (CCK) (10). Like CCK, systemic administration of Ex4 in rats induces c-Fos expression in NTS neurons (11,12). GLP-1-Rs are also expressed within feeding-relevant central nervous system (CNS) regions, and GLP-1 is synthesized by a small population of neurons in the caudal NTS (13–15). Central injection of GLP-1 or agonists of its receptor suppress food intake (16,17), and although the physiological role played by neuronal GLP-1 in the control of food intake remains uncertain, central GLP-1-Rs have been implicated in the anorexic response to noxious stimuli such as LiCl and lipopolysaccharide (18 – 21). Leptin, produced by adipose tissue in direct proportion to its mass, plays an important role in the maintenance of energy balance through its effects to reduce food intake (22). Studies demonstrating that leptin treatment decreases meal size (23,24) support a model in which leptin signaling in the brain interacts with meal-related signals that promote satiety. Further support for this hypothesis comes from the findings that leptin pretreatment enhances the anorexic response to both gastrointestinal nutrient infusion and to satiety-inducing peptides such as CCK and bombesin (25–27). Conversely, food deprivation, which lowers plasma leptin levels, attenuates CCK-induced anorexia, and leptin replacement during fasting restores the effectiveness of CCK (28). Here, we evaluated the hypothesis that leptin interacts with GLP-1 in a similar manner by asking whether changes in ambient leptin levels impact the ability of GLP-1 and/or Ex4 to reduce food intake. Because leptin pretreatment increases the effect of intragastric nutrient infusion, CCK, or bombesin to induce expression of c-Fos in several brain areas, including the NTS and area postrema (25,27), we also asked whether we would see a similar interaction for the activation of hindbrain neurons by Ex4 in nuclei involved in meal-size regulation. RESEARCH DESIGN AND METHODS Naı¨ve male lean (Fak/Fak) and obese (fak/fak) Koletsky rats (Vassar College, Poughkeepsie, NY) were generated from serial backcrosses (N10 equivalent) of the fak mutation (also known as Koletsky, faf, f, or cp) to the inbred rat 3387

LEPTIN AND GLP-1 INTERACT TO AFFECT FEEDING

strain, LA/N. Naı¨ve male Wistar rats (mean weight 340 g except where otherwise noted) were obtained from Charles River Laboratories (Wilmington, MA). All subjects were individually housed in Plexiglas cages in a temperature-controlled room under a 12-h light-dark cycle. Water and standard rat chow (PMI Nutrition International) were available ad libitum except where otherwise noted. All subjects were handled daily and habituated to intraperitoneal injection of 1 ml saline and measurement of food intake throughout the dark phase on at least three occasions before the experiments began. Body weight and food intake were measured daily. Study procedures were approved by the Animal Care Committee at the University of Washington and conformed to standards described in the Guide for the Care and Use of Laboratory Animals (National Research Council, 1996). Drugs. GLP-1 was obtained from Bachem (Torrance, CA), Ex4 was obtained from California Peptide Research (Napa, CA), and both were dissolved in sterile 0.9% saline. Recombinant mouse leptin (Dr. Parlow, National Hormone & Peptide Program, National Institute of Diabetes and Digestive and Kidney Diseases) was dissolved in 0.01 mol/l NaOH. GLP-1 response in leptin receptor– deficient Koletsky rats. Obese leptin receptor– deficient Koletsky fak/fak rats (mean body wt, 681 g) and their lean FaK/FaK littermates (mean body wt, 403 g) were assigned to one of three intraperitoneal treatment groups: saline, 10 ␮g GLP-1, or 33.3 ␮g GLP-1 (n ⫽ 6 – 8/group). GLP-1 doses were chosen based on pilot studies undertaken in our laboratory (data not shown). On the day of the experiment, food was removed 4 h before the dark phase. Immediately before dark onset, rats received intraperitoneal injections of saline or GLP-1, and food was returned. Food intake was measured at 30 and 60 min after injections. Effect of peripheral leptin pretreatment on GLP-1– or Ex4-induced anorexia. In the first study, rats were assigned to one of four weight-matched groups: saline/saline, leptin/saline, saline/GLP-1, and leptin/GLP-1 (n ⫽ 6 –7/ group). On the day of the experiment, all rats were weighed, and food was removed 4 h before the onset of the dark phase. Intraperitoneal injections of either a low dose of leptin (0.5 mg/kg, 1 ml, shown in preliminary studies to be subthreshold for independent feeding effects) or saline were administered 1 h before the start of the dark cycle. Immediately before dark onset, rats were intraperitoneally injected with saline or GLP-1 at a dose (100 ␮g/kg, 1 ml) shown in preliminary studies to significantly suppress food intake. After the second injections, preweighed food was returned to all subjects. Food intake was measured 0.5, 1, 2, 3, 4, and 24 h later. In a separate experiment of the same design, rats were assigned to one of four weight-matched groups: saline/saline, leptin/saline, saline/Ex4, and leptin/Ex4 (n ⫽ 5– 6/group). Ex4 was administered at a dose (1 ␮g/kg, 1 ml) shown in preliminary studies to reduce food intake for at least 4 h after administration. The study protocol was otherwise identical to that described above. Effect of third intracerebroventricular leptin pretreatment on Ex4induced anorexia. At least 1 week after stereotaxic implantation of a 26-gauge cannula (Plastics One, Roanoke, VA) aimed at the third cerebral ventricle (third i.c.v.) (28), rats (mean body wt, 475 g) were assigned to one of four groups: saline/saline, leptin/saline, saline/Ex4, and leptin/Ex4 (n ⫽ 5–7/group). The study design was otherwise identical to those described above, with third intracerebroventricular injections of leptin (2.5 ␮g/2 ␮l) or saline administered 45 min before the start of the dark cycle and intraperitoneal injection of saline or 1 ␮g/kg Ex4 immediately before dark onset. Cannula placement was verified before the start of the experiment through the measurement of a sympathetically mediated increase in plasma glucose 60 min after third intracerebroventricular injection of 210 ␮g 5-thio-D-glucose (29). Effect of fasting on the anorexic response to GLP-1 or Ex4. In the first study, rats were divided into two weight-matched groups: one was allowed ad libitum access to food before intraperitoneal injections, and the other was fasted for 24 h before intraperitoneal treatment. Each group was further subdivided into two groups of ad libitum–fed rats and two groups of fasted rats (n ⫽ 6 –7/group), each receiving an intraperitoneal injection of either saline or 33.3 ␮g GLP-1. Food was removed from the ad libitum groups 4 h before intraperitoneal injections, which were administered immediately before dark onset. After intraperitoneal treatment, preweighed food was returned to all subjects, and food intake was measured 30 and 60 min later. In a separate study of the same design, ad libitum–fed or fasted rats received intraperitoneal injections of saline, 0.1 ␮g Ex4, or 0.33 ␮g Ex4 (n ⫽ 9 –11/group). Preweighed food was then returned to all subjects, and food intake was measured 2 and 4 h later. Effect of leptin replacement on fasting-induced attenuation of Ex4induced anorexia. Rats were separated into three weight-matched groups: 1) ad libitum feeding/saline minipump (n ⫽ 11); 2) 24-h fasted before intraperitoneal treatment, with leptin replacement via minipump during the fast (n ⫽ 12); and 3) 24-h fasted before intraperitoneal treatment/saline minipump (n ⫽ 11). As previously described (28), all rats were implanted subcutaneously with 3388

osmotic minipumps that were attached to Lynch coils delivering saline for 6 days after surgery, allowing for complete recovery of food intake and body weight before the experiment. On the 6th day after surgery, food was removed from the fasting/saline and fasting/leptin groups, and the fasting/leptin group began to receive leptin at a concentration of 30 ␮g/day, a dose previously determined to achieve plasma leptin levels in fasted rats comparable with those of controls fed ad libitum (28). At this time, the ad libitum/saline and fasting/saline groups continued to receive saline. To determine the success of our leptin replacement protocol, a tail-blood sample (50 ␮l) was taken from each rat for assessment of prefasting plasma leptin levels, and we took a second tail-blood sample 22 h into the fast. At that time, food was removed from the ad libitum/saline group. Immediately before the onset of the dark cycle, 24 h after the fast began for the fasting/saline and fasting/leptin groups, each rat was injected intraperitoneally with 0.33 ␮g Ex4 or saline. Preweighed food was immediately returned to all subjects, and food intake was measured at 2 and 4 h after injections. Plasma leptin levels were determined using a mouse leptin ELISA (Crystal Chem, Downers Grove, IL) that is 90% cross-reactive with rat leptin, with a detection range of 0.2–12.8 ng/ml. Samples were measured in duplicate, and the mean value for each was used for data analysis. Effect of leptin pretreatment on Ex4-induced c-Fos in the caudal brainstem. Rats were divided into four weight-matched groups: saline/saline, leptin/saline, saline/Ex4, and leptin/Ex4 (n ⫽ 3– 4/group). On the experiment day, food was removed 4 h before dark phase onset. Intraperitoneal injections of either leptin (0.5 mg/kg) or saline were administered 1 h before the dark cycle, and rats were intraperitoneally injected with Ex4 (1 ␮g/kg) or saline immediately before dark onset. Ninety minutes after the second intraperitoneal injection, rats were deeply anesthetized (180 mg/kg ketamine and 30 mg/kg xylazine i.p.) and then transcardially perfused with PBS followed by 4% paraformaldehyde. Brains were removed and placed in 25% sucrose/PBS overnight. The brains were then frozen in isopentane at ⫺37°C. Coronal cryostat sections (14 ␮m) through the NTS and area postrema were slide mounted and stored at ⫺80°C. c-Fos immunohistochemical staining. The technique that we used for c-Fos staining of anatomically matched sections through the NTS at the level of the area postrema and rostral to the area postrema was recently published (30). Our primary antibody was rabbit polyclonal anti– c-Fos (Calbiochem, San Diego, CA) diluted 1:5,000 in 0.1% BSA in 10 mmol/l PBS, and the secondary antibody, donkey anti-rabbit IgG–Cy-3 (Jackson Immunoresearch, West Grove, PA), was diluted 1:200 in 0.1% BSA in 10 mmol/l PBS. Control sections incubated with normal serum did not show staining of hindbrain cells. Quantitative analysis of immunostaining. Slides were examined using a 10⫻ objective on a Nikon Eclipse E600 fluorescence microscope, and digital RGB images were acquired with a Diagnostic Images SPOT RT Color camera and SPOT software. NIH Image software was used to count cells positive for c-Fos–like immunoreactivity (c-FLI). Labeling in the NTS was counted bilaterally on four to six sections per rat at two different anatomical levels: one rostral to the area postrema (between ⫺12.80 and ⫺13.24 mm from bregma), and one at the mid–area postrema level (between ⫺13.68 and ⫺13.80 mm from bregma) (31). Labeling in the area postrema was counted on four to six sections per rat. For each brain nucleus and anatomical level, mean values were determined for each subject. Statistical analysis. Statistical comparisons between and within groups were made by two-way ANOVA. Post hoc comparisons were made with Tukey’s honestly significant difference test, and pairwise comparisons were

FIG. 1. Dose response for effects of intraperitoneal GLP-1 on food intake in lean FaK/FaK and obese fak/fak rats. Food intake was measured at 30 and 60 min after intraperitoneal injection of either saline or GLP-1 at the indicated doses. Data are means ⴞ SE, *P < 0.05 relative to vehicle. DIABETES, VOL. 55, DECEMBER 2006

D.L. WILLIAMS, D.G. BASKIN, AND M.W. SCHWARTZ

FIG. 2. A: Cumulative food intake measured during the first 4 h after pretreatment injections of either saline or leptin (0.5 mg/kg i.p.) followed 1 h later by saline or GLP-1 (100 ␮g/kg i.p.). 䡺, saline/saline; E, leptin/saline; f, saline/GLP-1; F, leptin/GLP-1. Food intake (B) and body weight (C) were measured in the same four groups over the 24 h after treatment. Data are means ⴞ SE. *P < 0.05. LEP, leptin; SAL, saline. made with Bonferroni-adjusted t tests. P values ⬍0.05 were considered to be statistically significant.

RESULTS

GLP-1 response in leptin receptor– deficient Koletsky rats. We observed a significant interaction between genotype and GLP-1 treatment at 30 and 60 min after injections [30 min: F (2,34) ⫽ 3.68, P ⬍ 0.05; 60 min: F (2,34) ⫽ 4.39, P ⬍ 0.05]. Both doses of GLP-1 significantly suppressed food intake in lean FaK/FaK rats (P ⬍ 0.05), but obese fak/fak rats showed no response to GLP-1 (Fig. 1). Effect of leptin pretreatment on GLP-1– or Ex4induced anorexia. Leptin pretreatment strongly enhanced the anorexic effect of GLP-1 (Fig. 2). At 0.5 h after injections, GLP-1 effectively reduced food intake regardless of pretreatment condition (P ⬍ 0.001). The effect of GLP-1 alone was gone by 3 and 4 h after injections; however, leptin/GLP-1 treatment substantially suppressed food intake [interaction at 3 h: F (1,22) ⫽ 9.19, P ⬍ 0.01;

4 h: F (1,22) ⫽ 16.04, P ⬍ 0.001, post hoc P ⬍ 0.05). At 24 h after treatment, food intake was similar across groups, but body weight was significantly reduced by leptin/GLP-1 treatment [interaction: F (1,21) ⫽ 6.17, P ⬍ 0.05, post hoc P ⬍ 0.01]. The same profile of results was observed for our leptin/ Ex4 experiment (Fig. 3). At 1 and 2 h after intraperitoneal injections, Ex4 significantly suppressed food intake regardless of pretreatment condition (P ⬍ 0.001), with no significant interactions between leptin and Ex4. At 3 h after injections, there was a near-significant interaction between leptin and Ex4 [F (1,18) ⫽ 4.17, P ⫽ 0.055], and a significant interaction was observed at 4 h after injections [F (1,18) ⫽ 6.70, P ⬍ 0.05]. Relative to the saline/saline treatment, Ex4 reduced 4-h food intake by 30% when delivered after a saline injection (P ⬍ 0.01). Leptin had no effect on feeding at this dose when delivered with saline, but rats given leptin/Ex4 suppressed food intake by 59% relative to saline/saline-treated rats (P ⬍ 0.001). Intake after leptin/Ex4 was significantly lower than after saline/

FIG. 3. A: Cumulative food intake measured during the first 4 h after pretreatment injections of either saline or leptin (0.5 mg/kg i.p.) followed 1 h later by saline or Ex4 (1.0 ␮g/kg i.p.). 䡺, saline/saline; E, leptin/saline; f, saline/Ex4; F, leptin/Ex4. Food intake (B) and body weight (C) were measured in the same four groups over the 24 h after treatment. Data are means ⴞ SE. *P < 0.05 vs. saline/saline and leptin/saline, #P < 0.05 vs. saline/saline, leptin/saline, and saline/Ex4. LEP, leptin; SAL, saline. DIABETES, VOL. 55, DECEMBER 2006

3389

LEPTIN AND GLP-1 INTERACT TO AFFECT FEEDING

FIG. 4. A: Food intake measured 4 h after third intracerebroventricular pretreatment with either saline or leptin (2.5 ␮g) followed 45 min later by intraperitoneal injection of either saline or Ex4 (1.0 ␮g/kg). Food intake (B) and body weight (C) were measured in the same four groups over the 24 h after treatment. Data are means ⴞ SE. *P < 0.05. LEP, leptin; SAL, saline.

Ex4 (P ⬍ 0.01). At 24 h after treatment, we observed significant interactions between leptin and Ex4 effects on food intake and body weight change [food intake: F (1,18) ⫽ 5.37, P ⬍ 0.05; body weight: F (1,18) ⫽ 8.12, P ⬍ 0.05]. Cumulative 24-h intake was not suppressed by leptin or Ex4 delivered alone, but the combination of leptin and Ex4 resulted in a significant 20% intake reduction (P ⬍ 0.05). Leptin/Ex4-treated rats also lost a small but significant amount of body weight compared with each of the other groups, which gained weight during those 24 h (P ⬍ 0.05). Effect of third intracerebroventricular leptin pretreatment on Ex4-induced anorexia. Although the interaction between intracerebroventricular leptin and Ex4 did not reach significance for 4-h food intake [F (1,19) ⫽ 2.36, P ⫽ 0.14], pairwise comparisons showed that the leptin/Ex4 group ate significantly less than all other groups at that time (P ⬍ 0.001) (Fig. 4A). We did observe a significant interaction between central leptin and Ex4 [F (1,19) ⫽ 6.93, P ⬍ 0.05] for 24-h intake. Leptin/Ex4-treated rats significantly reduced their food intake compared with all other groups (P ⬍ 0.01), and the leptin/saline and saline/Ex4 groups did not differ from vehicle (Fig. 4B). The same pattern of results was obtained for body weight change during this 24-h period [F (1,16) ⫽ 5.93, P ⬍ 0.05, excluding three statistical outliers], in which only the

leptin/Ex4 group showed significant weight loss (P ⬍ 0.05) (Fig. 4C). Effect of fasting on the anorexic response to GLP-1 or Ex4. The 24-h fast reduced body weight by an average of 14 g, whereas the ad libitum groups gained an average of 5 g during the same time period. Fasting before intraperitoneal injections increased mean baseline food intake at 2 and 4 h after treatment, but the differences between ad libitum–fed/saline and fasted/saline group intakes did not achieve statistical significance. Compared with rats fed ad libitum, fasting before GLP-1 treatment attenuated GLP-1–induced anorexia (Fig. 5). At 30 min after treatment, we observed a significant interaction between fasting state and GLP-1 [F (1,21) ⫽ 4.3, P ⬍ 0.05], where ad libitum–fed rats showed a 50% suppression of food intake after GLP-1 (P ⬍ 0.01), but fasted rats did not respond to drug treatment. We observed a similar effect of fasting on the response to Ex4 (Fig. 6). At the time point 4 h after injection, there was a significant interaction between preinjection fasting state and Ex4 [F (1,21) ⫽ 5.41, P ⬍ 0.01]. Ex4 reduced intake in ad libitum–fed rats by 48% at the 0.33-␮g dose (P ⬍ 0.01), and there was a nonsignificant tendency

FIG. 5. Effect of intraperitoneal GLP-1 (33.3 ␮g) on 30-min food intake in rats that were either fed ad libitum or fasted for 24 h before treatment. *P < 0.05. SAL, saline.

FIG. 6. Dose response for the effect of intraperitoneal Ex4 on 4-h food intake in rats that were either fed ad libitum or fasted for 24 h before Ex4 treatment. *P < 0.05.

3390

DIABETES, VOL. 55, DECEMBER 2006

D.L. WILLIAMS, D.G. BASKIN, AND M.W. SCHWARTZ

FIG. 7. A: Plasma leptin levels before and after a 24-h period, during which rats were either fed ad libitum or fasted and received a continuous subcutaneous infusion of either saline or leptin (30 ␮g/day) via osmotic minipump. B: Cumulative food intake 4 h after intraperitoneal injections of either vehicle or Ex4 (0.33 ␮g/kg) in the same groups of rats. Data are means ⴞ SE. *P < 0.05.

toward intake suppression at the 0.1-␮g dose (P ⫽ 0.09). In contrast, rats that were fasted before Ex4 treatment failed to show any reduction of food intake after either dose. Effect of physiological leptin replacement on fastinginduced attenuation of Ex4-induced anorexia. Although our protocol for leptin replacement during fasting prevented the dramatic fall of plasma leptin levels observed in the fasting/saline groups, plasma leptin remained significantly below prefasting values in leptin-infused animals (Fig. 7A). ANOVA revealed a significant interaction between group and time [F (2,28) ⫽ 4.27, P ⬍ 0.05]. Baseline leptin levels were the same across all groups, but differences were apparent for the second sample, 22 h into the fast with or without leptin replacement. Ad libitum/ saline rats showed no change in plasma leptin between the two sample points, whereas fasted/saline rats showed a decrease of 87% in plasma leptin levels obtained after fasting (P ⬍ 0.01), as expected. Fasted/leptin rats also showed significantly reduced plasma leptin in response to the fast (P ⬍ 0.01), but the magnitude of this reduction was small by comparison with that seen in the fasted/ saline group. After fast, leptin levels were significantly higher in the fasted/leptin group compared with fasted/ saline rats (P ⬍ 0.01). We again found that prior fasting eliminated the Ex4 effect on feeding (Fig. 7B) and that among saline-treated animals, there was a nonsignificant tendency for 4-h intake to be increased by prior fasting. For 4-h intake, there was a significant interaction between fasting/leptin replacement group and Ex4 [F (2,28) ⫽ 4.21, P ⬍ 0.05]. Despite the fact that our leptin replacement dose was insufficient to maintain plasma leptin at prefast levels, it completely reversed the effect of fasting to prevent Ex4-induced anorexia. Ex4 suppressed intake by 45–50% relative to saline in both the ad libitum/saline and fasted/leptin groups (P ⬍ 0.01), but fasted/saline rats showed no response to Ex4 injection. Effect of leptin pretreatment on Ex4-induced c-Fos in the caudal brainstem. Although Ex4 treatment after saline significantly increased the number of neurons with c-FLI in both the NTS and the area postrema (P ⬍ 0.05), leptin pretreatment attenuated the c-Fos response to Ex4 in both areas (Fig. 8). We observed significant interactions between leptin and Ex4 for each hindbrain area examined [rostral NTS: F (1,10) ⫽ 6.69, P ⬍ 0.05; caudal NTS: F DIABETES, VOL. 55, DECEMBER 2006

(1,10) ⫽ 7.48, P ⬍ 0.05; area postrema: F (1,10) ⫽ 58.27, P ⬍ 0.01]. Leptin alone had no effect on hindbrain c-Fos, but when delivered before Ex4, leptin blocked Ex4-induced c-Fos at both levels of the NTS and in the area postrema (P ⬍ 0.05). The amount of c-Fos observed in the NTS and area postrema in leptin/Ex4-treated rats did not significantly differ from that seen in rats treated with saline/saline. DISCUSSION

The experiments presented here provide strong evidence that variation in plasma leptin levels within the physiological range potently influences the feeding response to

FIG. 8. Representative images demonstrating the effect of leptin, Ex4, or both treatments on the c-Fos expression in NTS and area postrema (cc, central canal): saline/saline (A), leptin/saline (B), saline/Ex4 (C), and leptin/Ex4 (D). E: Means ⴞ SE number of c-Fos–positive nuclei in the NTS rostral to the area postrema (R-NTS), the NTS at the level of the area postrema (C-NTS), and in the area postrema. *P < 0.05. 3391

LEPTIN AND GLP-1 INTERACT TO AFFECT FEEDING

GLP-1 receptor stimulation. We first showed that leptin receptor– deficient rats do not respond to GLP-1 at doses that produce significant anorexia in wild-type rats. We then demonstrated that in normal Wistar rats, leptin pretreatment (at a low dose that had no effects when given alone) substantially increased the anorexic responses to GLP-1 and Ex4. Conversely, when rats were food deprived before GLP-1 or Ex4 treatment, these agents no longer suppressed food intake. Because fasting lowers plasma leptin levels, we hypothesized that its inhibitory effect on the response to GLP-1-R stimulation was attributable to a reduced leptin signal. In direct support of this hypothesis, low-dose leptin replacement during the fast completely restored the anorexic effect of Ex4 but had no effect on baseline food intake. Our findings lend further support to a model in which changes in plasma leptin levels affect food intake by modulating the response to dynamic mealrelated signals, including CCK and now GLP-1. The hypothesis that the anorexic response to GLP-1-R stimulation is dependent on an intact leptin signal is seemingly at odds with previous reports of impressive Ex4-induced anorexia and weight loss in leptin receptor– deficient Zucker fa/fa rats and ob/ob and db/db mice (4,16,32). We note that these previous studies used 4- to 6-week treatment protocols of multiple daily Ex4 injections, with doses 2–200 times higher than those used in the present studies. A parsimonious explanation for these and our current findings is that changes in leptin signaling shift the dose-response function for Ex4, and this shift is most evident at relatively low doses of Ex4. In any case, our demonstration that fak/fak rats fail to respond to low doses of GLP-1 strongly supports our conclusion that deficient leptin signaling impairs the feeding response to GLP-1-R stimulation. Whether this is true for other GLP-1 effects is an important unanswered question. Available data suggest that systemic administration of GLP-1 or Ex4 reduces food intake at least in part through a vagal afferent pathway (8,9), similar to the mechanism of action of CCK. This peripheral mechanism is likely distinct from the well-established effect of central GLP-1 receptor activation to reduce food intake (16,17). Because of its short half-life (⬍2 min), it is uncertain whether endogenous, gut-derived GLP-1 can enter the brain in sufficient amounts to stimulate central GLP-1-R. In contrast, Ex4 is not rapidly degraded and readily crosses the blood-brain barrier (33); it therefore remains possible that distinct mechanisms contribute to anorexia induced by Ex4 versus GLP-1 and that some or all of the feeding effects that we observed with peripherally administered Ex4 are mediated by central GLP-1-R action. As a first step toward identifying the CNS sites that integrate leptin and GLP-1-R signaling, we examined c-Fos expression in the hindbrain after Ex4 with or without leptin pretreatment. Consistent with previous studies (11,12), we found that Ex4 administered alone induced c-FLI in both the NTS and area postrema. Based on the established synergistic effects of leptin and CCK on hindbrain c-Fos (25), we expected that leptin would enhance Ex4-induced neuronal activation in these regions. Our finding that leptin pretreatment not only failed to enhance Ex4-induced c-FLI in the NTS and area postrema, but also completely prevented Ex4 from activating neurons in these regions, was most surprising. We draw several conclusions from these results. First, the interaction between leptin and GLP-1-R stimulation by Ex4 must be mediated through neural pathways that differ significantly 3392

from those involved in the interaction between leptin and CCK. Second, we suggest that excitation of NTS and area postrema neurons, as indicated by c-Fos expression, is not likely to be required for the expression of an anorexic response to Ex4 treatment. We observed that rats treated with leptin plus Ex4 showed a profound and long-lasting anorexia, but in a separate study, NTS and area postrema c-Fos expression in animals receiving leptin plus Ex4 was no greater than that of saline-treated controls. Thus, hindbrain neurons expressing c-FLI in response to Ex4 are unlikely to mediate its feeding effects. These cells may contribute to some other effect of Ex4, such as a slowing of gastric emptying (34), increasing insulin or decreasing glucagon secretion (35), or effects on cardiovascular function (12,36). Although the CNS mechanisms underlying the interactive effects of leptin and Ex4 on food intake remain uncertain, we note that this interaction may rely on an inhibitory response involving hindbrain neurons that is not apparent in the present c-Fos analysis. Our results rule out a synergistic activation of NTS and area postrema neurons by leptin and Ex4 and emphasize the potential role of other CNS nuclei as targets for future research. Our data from fasted rats with or without leptin replacement demonstrate that variation of plasma leptin concentrations within the physiological range can substantially impact the ability of GLP-1-R stimulation to reduce food intake. These observations support a model in which the satiety response to GLP-1 signaling is regulated by changes in plasma leptin levels and raise the possibility that reduced sensitivity to GLP-1 contributes to fasting-induced homeostatic responses that favor the recovery of lost weight. Our findings also raise the possibility that dietinduced obesity and other states of leptin resistance may attenuate the ability of GLP-1 receptor stimulation to reduce food intake. It will be important to discern whether the efficacy of Ex4 in the treatment of human obesity is sensitive to changes of energy balance and/or leptin levels and whether the interaction between leptin and GLP-1 signaling pertains to other effects of GLP-1 and Ex4, such as glucose lowering among patients with type 2 diabetes. ACKNOWLEDGMENTS

D.L.W. has received a National Institutes of Health (NIH) individual National Research Service Award fellowship. D.G.B. is the recipient of a Department of Veterans Affairs Senior Research Career Scientist Award at the VA Puget Sound Health Care System. M.W.S. has received NIH Grants DK-52989, DK-68340, and NS-32273. This work has received support from the Diabetes Endocrinology Research Center and from the Clinical Nutrition Research Unit of the University of Washington. This material is also based on work supported by the Office of Research and Development Medical Research Service, Department of Veterans Affairs. We thank Alex Cubelo, Iaela David, Jenny Kam, and Loan Nguyen for their expert technical assistance. REFERENCES 1. Baggio LL, Drucker DJ: Clinical endocrinology and metabolism: glucagonlike peptide-1 and glucagon-like peptide-2. Best Pract Res Clin Endocrinol Metab 18:531–554, 2004 2. Gutzwiller JP, Goke B, Drewe J, Hildebrand P, Ketterer S, Handschin D, Winterhalder R, Conen D, Beglinger C: Glucagon-like peptide-1: a potent regulator of food intake in humans. Gut 44:81– 86, 1999 3. Poon T, Nelson P, Shen L, Mihm M, Taylor K, Fineman M, Kim D: Exenatide improves glycemic control and reduces body weight in subjects DIABETES, VOL. 55, DECEMBER 2006

D.L. WILLIAMS, D.G. BASKIN, AND M.W. SCHWARTZ

with type 2 diabetes: a dose-ranging study. Diabetes Technol Ther 7:467– 477, 2005 4. Gedulin BR, Nikoulina SE, Smith PA, Gedulin G, Nielsen LL, Baron AD, Parkes DG, Young AA: Exenatide (exendin-4) improves insulin sensitivity and {beta}-cell mass in insulin-resistant obese fa/fa Zucker rats independent of glycemia and body weight. Endocrinology 146:2069 –2076, 2005 5. DeFronzo RA, Ratner RE, Han J, Kim DD, Fineman MS, Baron AD: Effects of exenatide (exendin-4) on glycemic control and weight over 30 weeks in metformin-treated patients with type 2 diabetes. Diabetes Care 28:1092– 1100, 2005 6. Triplitt C, Chiquette E: Exenatide: from the Gila monster to the pharmacy. J Am Pharm Assoc (Wash DC) 46:44 –52, 2006 7. Nakagawa A, Satake H, Nakabayashi H, Nishizawa M, Furuya K, Nakano S, Kigoshi T, Nakayama K, Uchida K: Receptor gene expression of glucagonlike peptide-1, but not glucose-dependent insulinotropic polypeptide, in rat nodose ganglion cells. Auton Neurosci 110:36 – 43, 2004 8. Abbott CR, Monteiro M, Small CJ, Sajedi A, Smith KL, Parkinson JR, Ghatei MA, Bloom SR: The inhibitory effects of peripheral administration of peptide YY(3–36) and glucagon-like peptide-1 on food intake are attenuated by ablation of the vagal-brainstem-hypothalamic pathway. Brain Res 1044:127–131, 2005 9. Talsania T, Anini Y, Siu S, Drucker DJ, Brubaker PL: Peripheral exendin-4 and peptide YY(3–36) synergistically reduce food intake through different mechanisms in mice. Endocrinology 146:3748 –3756, 2005 10. Moran TH, Baldessarini AR, Salorio CF, Lowery T, Schwartz GJ: Vagal afferent and efferent contributions to the inhibition of food intake by cholecystokinin. Am J Physiol 272:R1245–R1251, 1997 11. Baggio LL, Huang Q, Brown TJ, Drucker DJ: Oxyntomodulin and glucagonlike peptide-1 differentially regulate murine food intake and energy expenditure. Gastroenterology 127:546 –558, 2004 12. Yamamoto H, Lee CE, Marcus JN, Williams TD, Overton JM, Lopez ME, Hollenberg AN, Baggio L, Saper CB, Drucker DJ, Elmquist JK: Glucagonlike peptide-1 receptor stimulation increases blood pressure and heart rate and activates autonomic regulatory neurons. J Clin Invest 110:43–52, 2002 13. Merchenthaler I, Lane M, Shughrue P: Distribution of pre-pro-glucagon and glucagon-like peptide-1 receptor messenger RNAs in the rat central nervous system. J Comp Neurol 403:261–280, 1999 14. Mercer JG, Moar KM, Findlay PA, Hoggard N, Adam CL: Association of leptin receptor (OB-Rb), NPY and GLP-1 gene expression in the ovine and murine brainstem. Regul Pept 75–76:271–278, 1998 15. Larsen PJ, Tang-Christensen M, Holst JJ, Orskov C: Distribution of glucagon-like peptide-1 and other preproglucagon-derived peptides in the rat hypothalamus and brainstem. Neuroscience 77:257–270, 1997 16. Rodriquez de Fonseca F, Navarro M, Alvarez E, Roncero I, Chowen JA, Maestre O, Gomez R, Munoz RM, Eng J, Blazquez E: Peripheral versus central effects of glucagon-like peptide-1 receptor agonists on satiety and body weight loss in Zucker obese rats. Metabolism 49:709 –717, 2000 17. Turton MD, O’Shea D, Gunn I, Beak SA, Edwards CM, Meeran K, Choi SJ, Taylor GM, Heath MM, Lambert PD, Wilding JP, Smith DM, Ghatei MA, Herbert J, Bloom SR: A role for glucagon-like peptide-1 in the central regulation of feeding. Nature 379:69 –72, 1996 18. Grill HJ, Carmody JS, Amanda Sadacca L, Williams DL, Kaplan JM: Attenuation of lipopolysaccharide anorexia by antagonism of caudal brain stem but not forebrain GLP-1-R. Am J Physiol Regul Integr Comp Physiol 287:R1190 –R1193, 2004 19. Seeley RJ, Blake K, Rushing PA, Benoit S, Eng J, Woods SC, D’Alessio D:

DIABETES, VOL. 55, DECEMBER 2006

The role of CNS glucagon-like peptide-1 (7–36) amide receptors in mediating the visceral illness effects of lithium chloride. J Neurosci 20:1616 – 1621, 2000 20. Rinaman L: A functional role for central glucagon-like peptide-1 receptors in lithium chloride-induced anorexia. Am J Physiol 277:R1537–R1540, 1999 21. Thiele TE, Seeley RJ, D’Alessio D, Eng J, Bernstein IL, Woods SC, van Dijk G: Central infusion of glucagon-like peptide-1-(7–36) amide (GLP-1) receptor antagonist attenuates lithium chloride-induced c-Fos induction in rat brainstem. Brain Res 801:164 –170, 1998 22. Schwartz MW, Woods SC, Porte D Jr, Seeley RJ, Baskin DG: Central nervous system control of food intake. Nature 404:661– 671, 2000 23. Flynn MC, Scott TR, Pritchard TC, Plata-Salaman CR: Mode of action of OB protein (leptin) on feeding. Am J Physiol 275:R174 –R179, 1998 24. Kahler A, Geary N, Eckel LA, Campfield LA, Smith FJ, Langhans W: Chronic administration of OB protein decreases food intake by selectively reducing meal size in male rats. Am J Physiol 275:R180 –R185, 1998 25. Emond M, Schwartz GJ, Ladenheim EE, Moran TH: Central leptin modulates behavioral and neural responsivity to CCK. Am J Physiol 276:R1545– R1549, 1999 26. Emond M, Ladenheim EE, Schwartz GJ, Moran TH: Leptin amplifies the feeding inhibition and neural activation arising from a gastric nutrient preload. Physiol Behav 72:123–128, 2001 27. Ladenheim EE, Emond M, Moran TH: Leptin enhances feeding suppression and neural activation produced by systemically administered bombesin. Am J Physiol Regul Integr Comp Physiol 289:R473–R477, 2005 28. McMinn JE, Sindelar DK, Havel PJ, Schwartz MW: Leptin deficiency induced by fasting impairs the satiety response to cholecystokinin. Endocrinology 141:4442– 4448, 2000 29. Ritter RC, Slusser PG, Stone S: Glucoreceptors controlling feeding and blood glucose: location in the hindbrain. Science 213:451– 452, 1981 30. Blevins JE, Schwartz MW, Baskin DG: Evidence that paraventricular nucleus oxytocin neurons link hypothalamic leptin action to caudal brain stem nuclei controlling meal size. Am J Physiol Regul Integr Comp Physiol 287:R87–R96, 2004 31. Paxinos G, Watson C: The Rat Brain in Stereotaxic Coordinates. Burlington, MA, Elsevier Science & Technology Books, 1998 32. Young AA, Gedulin BR, Bhavsar S, Bodkin N, Jodka C, Hansen B, Denaro M: Glucose-lowering and insulin-sensitizing actions of exendin-4: studies in obese diabetic (ob/ob, db/db) mice, diabetic fatty Zucker rats, and diabetic rhesus monkeys (Macaca mulatta). Diabetes 48:1026 –1034, 1999 33. Kastin AJ, Akerstrom V: Entry of exendin-4 into brain is rapid but may be limited at high doses. Int J Obes Relat Metab Disord 27:313–318, 2003 34. Chelikani PK, Haver AC, Reidelberger RD: Intravenous infusion of glucagon-like peptide-1 potently inhibits food intake, sham feeding, and gastric emptying in rats. Am J Physiol Regul Integr Comp Physiol 288:R1695– R1706, 2005 35. Knauf C, Cani PD, Perrin C, Iglesias MA, Maury JF, Bernard E, Benhamed F, Gremeaux T, Drucker DJ, Kahn CR, Girard J, Tanti JF, Delzenne NM, Postic C, Burcelin R: Brain glucagon-like peptide-1 increases insulin secretion and muscle insulin resistance to favor hepatic glycogen storage. J Clin Invest 115:3554 –3563, 2005 36. Yamamoto H, Kishi T, Lee CE, Choi BJ, Fang H, Hollenberg AN, Drucker DJ, Elmquist JK: Glucagon-like peptide-1-responsive catecholamine neurons in the area postrema link peripheral glucagon-like peptide-1 with central autonomic control sites. J Neurosci 23:2939 –2946, 2003

3393