Peptide YY3–36 and Glucagon-Like Peptide-17–36 Inhibit Food Intake Add
http://www.100md.com
《内分泌学杂志》
Department of Metabolic Medicine, Hammersmith Hospital, Imperial College London, London W12 ONN, United Kingdom
Abstract
Peptide YY (PYY) and glucagon like peptide (GLP)-1 are cosecreted from intestinal L cells, and plasma levels of both hormones rise after a meal. Peripheral administration of PYY3–36 and GLP-17–36 inhibit food intake when administered alone. However, their combined effects on appetite are unknown. We studied the effects of peripheral coadministration of PYY3–36 with GLP-17–36 in rodents and man. Whereas high-dose PYY3–36 (100 nmol/kg) and high-dose GLP-17–36 (100 nmol/kg) inhibited feeding individually, their combination led to significantly greater feeding inhibition. Additive inhibition of feeding was also observed in the genetic obese models, ob/ob and db/db mice. At low doses of PYY3–36 (1 nmol/kg) and GLP-17–36 (10 nmol/kg), which alone had no effect on food intake, coadministration led to significant reduction in food intake. To investigate potential mechanisms, c-fos immunoreactivity was quantified in the hypothalamus and brain stem. In the hypothalamic arcuate nucleus, no changes were observed after low-dose PYY3–36 or GLP-17–36 individually, but there were significantly more fos-positive neurons after coadministration. In contrast, there was no evidence of additive fos-stimulation in the brain stem. Finally, we coadministered PYY3–36 and GLP-17–36 in man. Ten lean fasted volunteers received 120-min infusions of saline, GLP-17–36 (0.4 pmol/kg·min), PYY3–36 (0.4 pmol/kg·min), and PYY3–36 (0.4 pmol/kg·min) + GLP-17–36 (0.4 pmol/kg·min) on four separate days. Energy intake from a buffet meal after combined PYY3–36 + GLP-17–36 treatment was reduced by 27% and was significantly lower than that after either treatment alone. Thus, PYY3–36 and GLP-17–36, cosecreted after a meal, may inhibit food intake additively.
Introduction
THE GUT HORMONES peptide YY (PYY) and glucagon-like peptide (GLP)-1 are cosecreted from endocrine L cells of the small and large bowel. Endogenous concentrations of both hormones are low in the fasting state and rise within 30 min of a meal (1). However, whereas these hormones have been found to inhibit food intake individually (2, 3), their combined effect has yet to be investigated.
It has been reported that peripheral administration of PYY3–36 inhibits appetite in rodents (4, 5) and man (2). Peripheral administration of GLP-17–36 inhibits food intake in rodents (6, 7) and man (8). GLP-17–36 and the GLP-1 receptor agonist exendin-4 also stimulate insulin release and inhibit circulating glucagon (6). In contrast, no changes in insulin and glucagon concentrations occur after PYY3–36 administration (9).
PYY and GLP-1 may mediate these peripheral effects by altering central nervous system appetite circuits. After ip administration of PYY3–36, an increase in c-fos, a marker of neuronal activation, has been reported in the hypothalamic arcuate nucleus (ARC) (2). In contrast, GLP-1 is reported to act primarily on brain stem neurons. After iv exendin-4, c-fos activation was observed in the area postrema (AP) (10), which, like the ARC, has access to the peripheral circulation. c-fos immunoreactivity was also detected in the ARC and paraventricular nucleus (PVN). This hypothalamic activation could be mediated via neuronal projections from the brain stem to the PVN (11) or directly onto ARC neurons.
Both PYY3–36 and GLP-17–36 have been administered to humans. Intravenous infusion of PYY3–36 was associated with a 36% reduction in energy intake at a free buffet meal (2). PYY3–36 has also been reported to inhibit food intake in obese subjects (12). A recent metaanalysis reported that infusion of GLP-17–36 was associated with a small dose-dependent reduction in energy intake in both lean and obese subjects (8). Interestingly, when PYY and GLP-1 were coadministered to man, an additive inhibitory effect on pentagastrin-induced gastric acid secretion was observed (13), although their combined effect on appetite was not examined.
Much work has been performed investigating the effect of exogenous administration of individual gut hormones on appetite to determine their potential physiological role in the control of food intake (4, 14, 15). However, after a meal not one but several gut hormones are elevated, and it may be that the interactions between the changes in levels of different hormones influence food intake more than each hormone individually.
The aim of our study was to investigate whether the combined administration of PYY3–36 and GLP-17–36 was more effective in inhibiting food intake than either peptide alone. First, we examined the effects of coadministration of PYY3–36 and GLP-17–36 in lean and obese rodents. Second, to investigate potential mechanisms, we quantified fos-like immunoreactivity (FLI) in the hypothalamus and brain stem after ip PYY3–36 and GLP-17–36 alone and PYY3–36 + GLP-17–36 coadministration. Finally, we performed a randomized, double-blind, placebo-controlled, crossover trial in healthy volunteers to compare the effects of PYY3–36, GLP-17–36, and PYY3–36 with GLP-17–36 on energy intake in man.
Materials and Methods
Materials
Synthetic human PYY3–36 and GLP-17–36 were purchased from Phoenix Pharmaceuticals (Belmont, CA). Unless otherwise stated, all chemicals were purchased from Merck (Poole, Dorset, UK).
Animals
All animal procedures were approved by the British Home Office Animals (Scientific Procedures) Act 1986 (project license no. 70/5516 and 70/5281). Male mice and rats were maintained in individual cages under controlled temperature (21–23 C) and light (12-h light, 12-h dark cycle; lights on at 0700 h) with ad libitum access to standard chow (RM1 diet, SDS Ltd., Witham, Essex, UK) and water. All animals were handled daily for 7 d before the first study, particularly in the light of recent evidence that peripheral PYY3–36 does not inhibit food intake in stressed animals unaccustomed to their experimental conditions but does so in well-handled, acclimatized animals (5). During the acclimatization period, each animal received two saline injections to minimize stress on the study days.
Intraperitoneal injections
Intraperitoneal injections were administered to mice via a 0.5 ml syringe with a 29-gauge needle (maximum injection volume, 0.1 ml) and rats via a 1 ml syringe with a 25-gauge needle (maximum volume, 0.5 ml). In each study animals received one injection only of equivalent volume for each treatment group. For the combination treatment, GLP-17–36 and PYY3–36 were drawn up in a single syringe. In all feeding studies, food intake was measured 1, 2, 4, 8, and 24 h after injection as previously described (7).
Feeding studies in lean mice
C57BL/6 mice (20–25 g) were injected after a 20-h fast. To determine suitable doses for the combined study, dose-response studies were performed for PYY3–36 and GLP-17–36 individually. For the PYY3–36 study, mice were ip injected with 30, 100, 150, and 300 nmol/kg PYY3–36 (n = 11–12). For the GLP-17–36 study, mice were injected with 30, 100, 300, and 900 nmol/kg GLP-17–36 (n = 10). For the combined study, mice were injected with saline, 100 nmol/kg PYY3–36, 200 nmol/kg PYY3–36, 100 nmol/kg GLP-17–36, 200 nmol/kg GLP-17–36, or 100 nmol/kg PYY3–36 + 100 nmol/kg GLP-17–36 (n = 8–9).
Feeding studies in ob/ob and db/db mice
To investigate these effects in obese rodents, PYY3–36 + and GLP-17–36 were coadministered to ob/ob and db/db mice (both on a C57BL/6 background) in a four-way crossover design. After a 20-h fast, ob/ob mice (mean weight 54.5 ± 1.4 g) and their wild-type (WT) littermates (32.5 ± 1.0 g) and db/db mice (41.8 ± 1.12 g) and their WT littermates (31.0 ± 0.5 g) were ip injected with saline, 200 nmol/kg PYY3–36, 200 nmol/kg GLP-17–36, or 100 nmol/kg PYY3–36 + 100 nmol/kg GLP-17–36. Four studies were performed with 72 h between each study so that each mouse received all four treatments (n = 10 per treatment group).
Feeding studies in rats
To examine the effect of PYY3–36 and GLP-17–36 in another species and during another feeding phase, these peptides were coadministered in rats in the dark phase. Rodents are nocturnal feeders and eat the majority of their food during the night (16). Freely feeding rats (150–200 g) were injected in the hour before lights off. For the PYY3–36 dose-response study, animals received ip saline or 1, 3, or 10 nmol/kg PYY3–36 (n = 12–13). For GLP-17–36, rats received saline or 3, 10, 30, or 100 nmol/kg GLP-17–36 (n = 12–13). To investigate the combined effect of low doses, 1 nmol/kg PYY3–36 and 10 nmol/kg GLP-17–36 were chosen for the joint study (see Results). On two evenings, 72 h apart, animals were randomized and injected ip with saline, 10 nmol/kg GLP-17–36, 20 nmol/kg GLP-17–36, 1 nmol/kg PYY3–36, 2 nmol/kg PYY3–36, or 1 nmol/kg PYY3–36 + 10 nmol/kg GLP-17–36 (total n = 19–20 per group) before the dark phase.
Immunocytochemistry
The protocols for c-fos staining and quantification were similar to those described previously (17). Briefly, rats (240–265 g) were each handled for 5 min/d for 1 wk to minimize the stress of the procedure. After an overnight fast, rats received ip injections of saline, 10 nmol/kg GLP-17–36, 1 nmol/kg PYY3–36, or 1 nmol/kg PYY3–36 + 10 nmol/kg GLP-17–36 (n = 5–6 per group). Ninety minutes later, they were terminally anesthetized with ip sodium pentobarbitone (200 mg) (Rhne Merieux, Harlow, UK) and transcardially perfused with 0.1 M PBS followed by 4% phosphate-buffered (PB)-formalin. The brains were removed and postfixed overnight in PB-formalin and then transferred to PB-sucrose (20% wt/vol) for 48 h. Serial 40-μm coronal brain sections were cut on a freezing microtome and stained for FLI using the avitin-biotin-peroxidase method. Fos-positive nuclei were counted using a light microscope (Eclipse E800; Nikon, Tokyo, Japan) by an experienced member of the research team with reference to the Rat Brain Atlas (18), who was unaware of which treatment had been given. The nuclei examined were the PVN, ARC, ventromedial and dorsomedial of the hypothalamus, and the nucleus of the solitary tract and the AP of the brain stem.
Behavioral studies
To determine whether alterations in behavior were associated with treatments of peptides used in the feeding and c-fos studies, an observation study was performed (19). After an overnight fast, rats (150–170 g) were injected ip with saline, 10 nmol/kg GLP-17–36, 1 nmol/kg PYY3–36, 1 nmol/kg PYY3–36 + 10 nmol/kg GLP-17–36, 10 nmol/kg exendin-4, or 1 M lithium chloride (positive control) (n = 10 per group) and their food replaced. Their behavior was assessed for the subsequent 60 min by two observers who were unaware of which treatment had been given. Each observer recorded behaviors of 30 animals (n = 5 per group per observer). Each rat was observed for 15 sec in every 5 min. The 15-sec periods were subdivided into three 5-sec periods. In each or these periods, the observer selected the behavior that most closely resembled what the rat was doing. These options were feeding, drinking, locomotion, grooming, head down, sleeping, and rearing.
In a further experiment, freely feeding rats were ip injected before the dark phase with saline, 10 nmol/kg GLP-17–36, 1 nmol/kg PYY3–36, 1 nmol/kg PYY3–36 + 10 nmol/kg GLP-17–36, or 10 nmol/kg exendin-4. Infrared beam breaks in two planes x-tot (horizontal movement) and z-tot (vertical movement) were counted for 1 h using an Opto-Max system (Columbus Instruments, Columbus, OH) to assess locomotion.
Human randomized, double-blind, controlled trial
The study was approved by the Hammersmith, Queen Charlotte’s, and Chelsea Ethics Committee (reference no. 2002/6261) and performed in accordance with the principles of the Declaration of Helsinki. Healthy lean volunteers were recruited through advertisement. Exclusion criteria were substance abuse, pregnancy, significant medical or psychiatric illness, or regular medication except the oral contraceptive pill. Twelve subjects were recruited, each of whom gave their informed consent to participate in the study. One subject was withdrawn due to violation of the protocol and another was withdrawn due to nausea on a study morning. Thus, 10 subjects (four men and six women) completed the study. Their ages ranged from 22 to 29 with a mean age of 25.6 ± 0.7 yr, and body mass index ranged from 20.5 to 26.4 with a mean of 23.0 ± 0.7 kg/m2.
Each volunteer received four infusions, saline, GLP-17–36, PYY3–36, or PYY3–36 with GLP-17–36, each on a separate day at least 5 d apart. Subjects fasted from 2100 h on the evening before infusions. The infusion days were run as previously described (20).
Both PYY3–36 and GLP-17–36 were sterile on culture, and limulus amoebocyte lysate assay tests for pyrogen were negative. Two vials were used for each infusion (saline + saline, saline + GLP-17–36, saline + PYY3–36, and PYY3–36 + GLP-17–36). Both vials were dissolved in 2.5 ml hemacel (Beacon, Tunbridge Wells, Kent, UK) to minimize peptide adsorption, diluted in 0.9% saline (Bayer, Newbury, Berkshire, UK) and drawn up in a single syringe (total volume 50 ml). The infusion rates chosen were based on the results of previous studies. PYY3–36 at 0.8 pmol/kg·min was associated with a 36% reduction in food intake (2). The lowest rate of GLP-17–36 that has been shown to inhibit food intake is 0.75 pmol/kg·min (21). We aimed for a submaximal effect for each individual peptide and so selected rates of 0.4 pmol/kg·min for both GLP-17–36 and PYY3–36 infusions and a rate of 0.4 pmol/kg·min PYY3–36 + 0.4 pmol/kg·min GLP-17–36 for the combined infusion. Infusions were given for 120 min.
After 90 min infusion, a buffet lunch was served of preweighed food in excess consisting of chicken or vegetable curry, rice, fruit salad, and sweets. Thirty minutes later, the remaining food was removed and weighed and the infusion discontinued. Blood samples (10 ml) were taken 30 min before the start of the infusion, immediately before the infusion, and then every 30 min until 2 h after the meal. Blood was collected into heparin-coated tubes containing 2000 Kallikrein inhibitor units (0.2 ml) of aprotonin (Bayer). After centrifugation, plasma was separated and stored immediately at –70 C until RIA. Subjects were asked to score subjective nausea on a visual analog scales (0–100 mm) (22) at each time point. Pulse and blood pressure were also measured at these time points. Subjects were allowed to leave 2 h after the meal and instructed to eat ad libitum and complete a food diary for the following 24 h.
RIAs
All samples were assayed in duplicate and single assays to eliminate the effects of interassay variation. PYY, GLP-1, and insulin concentrations were quantified using established in-house RIAs and antibodies (23, 24, 25). Glucose concentrations were measured using a YSI-2300STAT analyzer (Yellow Springs Instruments, Yellow Springs, OH).
Statistical analysis
Results are shown as mean values ± SEM. Data from the animal feeding and immunocytochemistry and behavior studies were analyzed using a one-way ANOVA with a Dunnett’s two-sided post hoc test. For the human infusions, a random effect, repeated-measures ANOVA was performed and within-subject effect examined under a Greenhouse-Geisser correction. In all cases P < 0.05 was considered to be statistically significant.
Results
Coadministration of high-dose PYY3–36 and GLP-1 in lean mice
In the PYY3–36 dose-response study, food intake in the first hour was significantly inhibited by all doses of PYY3–36 administered [food intake 0–1 h (grams): saline, 0.67 ± 0.05; PYY3–36, 30 nmol/kg, 0.49 ± 0.04, 100 nmol/kg, 0.44 ± 0.03, 150 nmol/kg, 0.47 ± 0.02, 300 nmol/kg, 0.47 ± 0.04] (P < 0.05 for all doses vs. saline) (Fig. 1A). PYY appeared to cause maximal feeding inhibition from doses of 100 nmol/kg and above, and this dose was selected for the combined study. Inhibition in cumulative feeding was sustained with all doses for up to 8 h after injection (data not shown).
In the GLP-17–36 dose-response study, food intake in the first hour was significantly inhibited by 100, 300, and 900 nmol/kg GLP-17–36 but not significantly by the 30 nmol/kg dose [food intake 0–1 h (grams): saline, 0.85 ± 0.05; GLP-17–36, 30 nmol/kg, 0.72 ± 0.07, 100 nmol/kg, 0.45 ± 0.03, 300 nmol/kg, 0.31 ± 0.05, 900 nmol/kg, 0.20 ± 0.05] (P < 0.05 for 100, 300, and 900 nmol/kg GLP-17–36 vs. saline) (Fig. 1B). In contrast with the PYY study, greater inhibition of feeding was observed with increasing doses of GLP-17–36 (P < 0.05 for 100 vs. 30 nmol/kg GLP-17–36 and for 300 vs. 100 nmol/kg GLP-17–36). A dose of 100 nmol/kg GLP-17–36 was selected for the combined study, which achieved an approximately 50% reduction in food intake in the first hour. Inhibition in cumulative feeding was sustained up to 8 h after injection after 100, 300, and 900 nmol/kg GLP-17–36.
The doses for the coadministration study were as follows: saline, 100 nmol/kg PYY3–36, 200 nmol/kg PYY3–36, 100 nmol/kg GLP-17–36, 200 nmol/kg GLP-17–36 and 100 nmol/kg PYY3–36 + 100 nmol/kg GLP-17–36. All peptides inhibited feeding in the first hour, compared with saline. Coadministration of PYY3–36 with GLP-17–36 led to significantly greater feeding inhibition than all other groups [food intake 0–1 h (grams): saline, 0.74 ± 0.07; 100 nmol/kg GLP-17–36, 0.33 ± 0.05; 200 nmol/kg GLP-17–36, 0.33 ± 0.07; 100 nmol/kg PYY3–36, 0.60 ± 0.03; 200 nmol/kg PYY3–36, 0.51 ± 0.04; 100 nmol/kg PYY3–36 + 100 nmol/kg GLP-17–36, 0.07 ± 0.07: P < 0.05 for combined groups vs. all other groups] (Fig. 2). Feeding inhibition after the combination treatment was sustained up to 24 h after injection [food intake 0–24 h (grams): saline, 7.0 ± 0.2; 100 nmol/kg GLP-17–36, 6.5 ± 0.2; 200 nmol/kg GLP-17–36, 7.2 ± 0.2; 100 nmol/kg PYY3–36, 6.9 ± 0.2; 200 nmol/kg PYY3–36, 6.8 ± 0.2; 100 nmol/kg PYY3–36 + 100 nmol/kg GLP-17–36, 6.1 ± 0.3: P < 0.05 for combined group vs. saline].
Coadministration of PYY3–36 and GLP-17–36 in ob/ob and db/db mice
Mice were randomized to receive saline, 200 nmol/kg GLP-17–36, 200 nmol/kg PYY3–36, or 100 nmol/kg PYY3–36 + 100 nmol/kg GLP-17–36. Coadministration of PYY3–36 and GLP-17–36 in ob/ob mice and their WT littermates led to significant inhibition of feeding, compared with saline and both peptides alone [food intake 0–1 h (grams), ob/ob mice: saline, 0.48 ± 0.09; 200 nmol/kg GLP-17–36, 0.25 ± 0.03; 200 nmol/kg PYY3–36, 0.32 ± 0.05; 100 nmol/kg PYY3–36 + 100 nmol/kg GLP-17–36, 0.14 ± 0.03; P < 0.05 for combined group vs. all other groups; food intake 0–1 h (grams), WT mice: saline, 1.2 ± 0.08; 200 nmol/kg GLP-17–36, 0.68 ± 0.09; 200 nmol/kg PYY3–36, 0.78 ± 0.069; 100 nmol/kg PYY3–36 + 100 nmol/kg GLP-17–36, 0.33 ± 04; P < 0.05 for combined group vs. all other groups]. The percentage reduction in food intake with the coadministration treatment, compared with saline, was similar in ob/ob (–74 ± 4%) and WT mice (–70 ± 6%) (Fig. 3A). Reduction in cumulative food intake was sustained for up to 24 h in the WT mice and up to 8 h in the ob/ob mice (data not shown).
Coadministration of PYY3–36 and GLP-17–36 was also associated with significant reductions in food intake in db/db mice and their WT littermates, compared with saline and both peptides alone [food intake 0–1 h (grams), db/db mice: saline, 0.59 ± 0.10; 200 nmol/kg GLP-1, 0.45 ± 0.05; 200 nmol/kg PYY3–36, 0.36 ± 0.04; 100 nmol/kg PYY3–36 + 100 nmol/kg GLP-1, 0.20 ± 0.04; P < 0.05 for combined group vs. all other groups; food intake 0–1 h (grams), WT mice: saline, 1.17 ± 0.09; 200 nmol/kg GLP-17–36, 0.99 ± 0.05; 200 nmol/kg PYY3–36, 0.76 ± 0.07; 100 nmol/kg PYY3–36 + 100 nmol/kg GLP-17–36, 0.41 ± 0.07; P < 0.05 for combined group vs. all other groups]. The percentage reduction in food intake with the coadministration treatment, compared with saline, was similar in db/db (–65 ± 7%) and WT mice (–63 ± 6%) (Fig. 3B). Reduction in cumulative food intake was sustained for up to 8 h in both WT and db/db mice (data not shown).
Coadministration of low-dose PYY3–36 and GLP-17–36 in rats
PYY3–36 and GLP-17–36 dose-response studies were performed in freely feeding rats in the dark phase. In the PYY3–36 study, food intake in the first hour of the dark phase was significantly inhibited by 3 and 10 nmol/kg PYY3–36 but not significantly by the 1 nmol/kg dose [food intake 0–1 h (grams): saline, 4.9 ± 0.2; PYY3–36, 1 nmol/kg, 4.5 ± 0.5; 3 nmol/kg, 3.8 ± 0.3; 10 nmol/kg, 3.5 ± 0.3; P < 0.05 for 3 and 10 nmol/kg PYY3–36 vs. saline] (Fig. 4A). In the GLP-17–36 study, food intake in the first hour was significantly inhibited by 30 and 100 nmol/kg GLP-17–36 [food intake 0–1 h (grams): saline, 5.0 ± 0.3; GLP-17–36, 3 nmol/kg 4.9 ± 0.5; 10 nmol/kg, 4.5 ± 0.3; 30 nmol/kg, 4.1 ± 0.3; 100 nmol/kg, 3.8 ± 0.3; P < 0.05 for 30 and 100 nmol/kg GLP-17–36 vs. saline] (Fig. 4B).
The doses of peptides selected for the combined study were 3-fold lower than the minimum doses found to inhibit food intake significantly in the first hour. Rats received ip injections of saline, 1 nmol/kg PYY3–36, 2 nmol/kg PYY3–36, 10 nmol/kg GLP-17–36, 20 nmol/kg GLP-17–36, or 1 nmol/kg PYY3–36 + 10 nmol/kg GLP-17–36. Food intake was significantly inhibited only in the PYY3–36 + GLP-17–36 group (Fig. 5) [food intake 0–1 h (grams): saline, 4.1 ± 0.2; 10 nmol/kg GLP-17–36, 3.9 ± 0.3; 20 nmol/kg GLP-17–36, 4.2 ± 0.3; 1 nmol/kg PYY3–36, 4.2 ± 0.3; 2 nmol/kg PYY3–36, 4.1 ± 0.2; 1 nmol/kg PYY3–36 + 10 nmol/kg GLP-17–36, 2.8 ± 0.2]. There was significant inhibition of food intake with the combined treatment, compared with all other groups (P < 0.02). This effect was most marked in the first hour of the dark phase. Cumulative food intake was inhibited in the coadministration group for up to 4 h, compared with saline (P < 0.05) [food intake 0–4 h (grams): saline, 9.8 ± 0.3; 10 nmol/kg GLP-17–36, 9.7 ± 0.4; 20 nmol/kg GLP-17–36, 10.0 ± 0.2; 1 nmol/kg PYY3–36, 9.9 ± 0.5; 2 nmol/kg PYY3–36, 10.1 ± 0.4; 1 nmol/kg PYY3–36 + 10 nmol/kg GLP-17–36, 8.9 ± 0.4]. At these low doses, no differences in food intake were observed between groups at later time points (data not shown).
Immunocytochemistry
To elucidate a possible mechanism for the additive inhibitory feeding effect of PYY and GLP-1, c-fos was quantified after low dose individual peptide and combined peptide administration. Doses of PYY and GLP-1 that had been found not to alter food intake individually but had led to significant reduction in food intake after coadministration were chosen. Rats received ip injections of saline, 10 nmol/kg GLP-17–36, 1 nmol/kg PYY3–36, or 1 nmol/kg PYY3–36 + 10 nmol/kg GLP-17–36. In the hypothalamus, there were no significant differences in the number of FLI counts in the PVN, and ventromedial and dorsomedial of the hypothalamus between treatment groups (data not shown). At these low doses, no significant differences were seen in the ARC with either PYY3–36 or GLP-17–36 alone, compared with saline. However, an 85% increase in the number of FLI counts was observed in the ARC after the coadministration treatment, compared with saline (P < 0.05) (Fig. 6). Moreover, after the combined PYY3–36 + GLP-17–36 treatment, there were significantly more FLI than after PYY3–36 alone (P < 0.05) and a trend for more FLI after GLP-17–36 alone (P = 0.05) (FLI-positive neurons in the ARC per section: saline, 35.4 ± 6.18; GLP-17–36, 42.9 ± 1.2; PYY3–36, 36.2 ± 4.4; PYY3–36 + GLP-17–36, 65.6 ± 10.0).
In the brain stem, no significant differences in FLI counts were observed between these low-dose treatments in the nucleus of the solitary tract or AP (data not shown).
Behavioral studies
To determine whether behavioral alteration were associated with doses of PYY3–36 and GLP-17–36 at which an additive feeding effect was observed, an observation study was performed. Fasted rats were ip injected with saline, 10 nmol/kg GLP-17–36, 1 nmol/kg PYY3–36, 1 nmol/kg PYY3–36 + 10 nmol/kg GLP-17–36, 10 nmol/kg exendin-4, or 1 M lithium chloride (positive control) (n = 10 per group). The groups treated with exendin-4 and lithium chloride had significantly fewer feeding episodes and significantly more head-down episodes, compared with saline (P < 0.05) (Fig. 7). However, no significant differences in behavior were observed after PYY3–36, GLP-17–36, or PYY3–36 + GLP-17–36.
In further study, locomotion was assessed in freely feeding rats before the dark phase after ip saline, 10 nmol/kg GLP-17–36, 1 nmol/kg PYY3–36, 1 nmol/kg PYY3–36 + 10 nmol/kg GLP-17–36, or 10 nmol/kg exendin-4 (n = 9–10 per group). No significant differences in horizontal movement (x-tot) were observed after the combined PYY3–36 + GLP-17–36 treatment, but there was a slight reduction in vertical movement (z-tot), compared with saline (P = 0.04) (Table 1). There were significant reductions in both horizontal and vertical movement after exendin-4, compared with saline (P < 0.01 for both comparisons).
Coadministration of PYY3–36 and GLP-17–36 in human volunteers
Infusion of PYY3–36 (0.4 pmol/kg·min) was associated with a 15% reduction in energy intake at the buffet meal, compared with saline, although this reduction achieved only borderline statistical significance (P = 0.07). There was no significant difference in energy intake with GLP-17–36 infusion (0.4 pmol/kg·min), in keeping with previously published results (8). With the combined PYY3–36 and GLP-17–36 infusion, caloric intake at the buffet meal was reduced by 27% (buffet meal energy intake (kilojoules): saline, 4174 ± 285; GLP-17–36, 3978 ± 340; PYY3–36, 3551 ± 441; PYY3–36 + GLP-17–36, 3033 ± 279; percentage change from saline day: GLP-17–36, –5 ± 5%; PYY3–36, –15 ± 9%; PYY3–36 + GLP-17–36, –27 ± 3%) (Fig. 8). There was a statistical difference in energy intake among the four infusion groups (P < 0.01). The buffet caloric intake on the PYY3–36 + GLP-17–36 day was significantly less than on all other infusion days (P < 0.05).
Cumulative energy intake over 24 h (buffet + food diary) was significantly reduced with the combined PYY3–36 + GLP-17–36, compared with all the other groups (cumulative 24-h energy intake (megajoules): saline, 13.7 ± 0.6; GLP-17–36, 13.6 ± 1.3; PYY3–36, 13.5 ± 0.9; PYY3–36 + GLP-17–36, 12.0 ± 0.7, P < 0.05 for PYY3–36 + GLP-17–36 vs. all other groups; percentage change from saline day: GLP-17–36, –2 ± 7%; PYY3–36, –1 ± 5%; PYY3–36 + GLP-17–36, –13 ± 5%). Analysis of the visual analog scales revealed no significant differences in nausea scores between infusion days [nausea score after 90 min infusion (millimeters): saline, 9.4 ± 0.83; GLP-17–36, 1.2 ± 0.05; PYY3–36, 3.8 ± 0.18; PYY3–36 + GLP-17–36, 6.3 ± 2.2, P=NS].
There were significant increases in plasma PYY levels with PYY3–36 and PYY3–36 + GLP-17–36 infusions and plasma GLP-1 levels with GLP-17–36 and PYY3–36 + GLP-17–36 infusions (Tables 2 and 3). Interestingly, our GLP-17–36 infusion inhibited circulating PYY (PYY picomoles per liter at 90 min: saline infusion, 10.3 ± 1.2; GLP-17–36 infusion, 7.7 ± 0.8; P < 0.05). However, no significant changes in circulating GLP-1 were observed during PYY3–36 infusion.
Fasting plasma insulin was increased by GLP-17–36 infusion at 90 min, compared with saline, and no changes in insulin secretion were observed with PYY3–36 alone. Whereas the coadministration PYY3–36 + GLP-17–36 infusion enhanced insulin secretion, there was no additional secretion, compared with GLP-17–36 alone (insulin picomoles per liter at 90 min: saline, 53.2 ± 4.9; GLP-17–36, 63.0 ± 3.5; PYY3–36, 51.2 ± 5.5; PYY3–36 + GLP-17–36, 66.7 ± 4.3; P < 0.05 for GLP-17–36 and combined PYY3–36 + GLP-17–36 infusions vs. saline) (Table 4). Consistent with these changes in insulin levels, fasting plasma glucose was reduced by GLP-17–36 infusion and the joint coadministration infusion but unchanged by PYY3–36 alone (glucose millimoles per liter at 90 min: saline, 4.6 ± 0.1; GLP-17–36, 4.2 ± 0.1; PYY3–36, 4.5 ± 0.1; PYY3–36 + GLP-17–36, 4.1 ± 0.1; P < 0.01 for GLP-17–36 and combined PYY3–36+GLP-17–36 infusions vs. saline) (Table 5). Postprandial insulin and glucose levels were not measured due to the variability of energy intake at the buffet meal.
Discussion
In this paper we investigated the effects of the coadministration of PYY3–36 and GLP-17–36 on feeding in rodents and humans.
PYY3–36 and GLP-17–36 inhibited food intake individually in mice and rats. PYY3–36 appeared to cause maximal feeding inhibition at doses of 100 nmol/kg and above. In contrast, feeding inhibition associated with GLP-17–36 was similar to that seen with CCK-8, with increasing inhibition seen with increasing doses (14). In the dose-response studies with low doses, the minimum doses of PYY3–36 and GLP-17–36 found to inhibit food intake were 3 and 30 nmol/kg, respectively. Both these results are consistent with our previously published data (2, 7). In both high- and low-dose coadministration studies, PYY3–36 and GLP-17–36 administered together inhibited food intake significantly more than either peptide alone or more than double the dose of either peptide. Thus, PYY3–36 and GLP-17–36 appeared to inhibit feeding additively.
PYY3–36 and GLP-17–36 were coadministered in genetically obese mice. Examination of the food intake in the saline-injected groups revealed that the ob/ob and db/db mice ate less in the first hour of refeeding than their WT littermates. This somewhat surprising finding is consistent with previously published work (26, 27) and suggests that thermogenic adaptation to fasting reduced the subsequent energy requirement of the obese mouse models (28). Coadministration of PYY3–36 and GLP-17–36 treatment led to significantly greater reduction in feeding than twice the dose of either peptide alone in the ob/ob and db/db and WT mice. Thus, PYY3–36 and GLP-17–36 inhibited food intake additively in obese as well as lean rodents.
We quantified c-fos afterPYY3–36, GLP-17–36, and PYY3–36 + GLP-17–36 administration to determine a possible mechanism for their additive effect on feeding. There was a significant increase in c-fos in the ARC after low-dose combination PYY3–36 + GLP-17–36 treatment but no changes in the other treatment groups. Therefore, PYY3–36 and GLP-17–36 may inhibit food intake through additive stimulation of ARC neurons. However, c-fos mapping cannot identify the pathway(s) through which neuronal activation was mediated. The ARC neuronal activation observed in our study could have been mediated through the blood-brain barrier via the median eminence (29) or the vagus nerve and brain stem (30).
There were no significant differences in observed behavior in rats after PYY3–36, GLP-17–36, or PYY3–36 + GLP-17–36 at doses at which the combination inhibited feeding. In the locomotion study, no significant differences in horizontal beam breaks were observed after the combined PYY3–36 + GLP-17–36 treatment, but there were marginally fewer vertical beam breaks. In their recent publication, Talsania et al. (31) suggested that low-dose exendin-4 together with PYY3–36 may increase the suppression of food intake in rodents without inducing significant side effects. There were no differences in nausea scores between infusions in our human study. Taken together these results suggest that behavioral change is unlikely to account for the feeding inhibition observed with coadministration of PYY3–36 and GLP-17–36. Thus, these peptides may have a specific effect on feeding independent of behavioral change.
In our human study, GLP-17–36 alone (0.4 pmol/kg·min) did not lead to a significant change in energy intake. In human subjects the lowest reported infusion rate of GLP-17–36 associated with a change in energy intake is 0.75 pmol/kg·min (21). However, incretin effects have been observed with a lower rate of GLP-17–36 of 0.5 pmol/kg·min (24). In the current study, we found a small but statistically significant reduction in fasting glucose levels and an increase in the fasting insulin concentration with GLP-17–36 infusion.
PYY3–36 infusion (0.4 pmol/kg·min) was associated with a 15% reduction in energy intake, compared with saline infusion, although this did not reach statistical significance. The magnitude of this reduction is in keeping with the larger reduction of 36% reported after higher-rate PYY infusions (0.8 pmol/kg·min) (2).
Coadministration of PYY3–36 with GLP-17–36 was associated with a 27% reduction in energy intake from the buffet meal. The combination was more effective in inhibiting appetite then either peptide alone. Cumulative energy intake over 24 h, including the buffet meal, was reduced by 13% after combined infusion, compared with the saline infusion, and was significantly less than that after all other treatment groups. In contrast with the additive inhibitory effect on energy intake, the incretin effect of GLP-17–36 appeared unchanged with the addition of PYY3–36.
Our study involving coadministration of PYY3–36 and GLP-17–36 suggests an additive inhibitory feeding effect. Interestingly, the GLP-1 receptor agonist, exendin-4, and PYY3–36 have been reported to inhibit food intake synergistically in rodents (31). However, it appears that not all combinations of gut hormones behave in this way. Indeed, it has recently been reported that whereas infusion of CCK-33 or GLP-17–36 inhibited food intake when administered individually, no enhanced inhibition was observed when they were infused together (32). Further work is needed to determine whether the other combinations of gut hormones lead to additive alterations in food intake.
We observed a reduction of endogenous PYY levels with GLP-17–36 infusion. This has been previously demonstrated at a higher dose of peripheral GLP-17–36 administration (32). Inhibition of endogenous PYY could result in a reduction in the anorexic action of GLP-1. In contrast, as in previous work (12), no changes in endogenous GLP-1 concentrations were observed after PYY3–36 infusion.
In summary, we have demonstrated that PYY3–36 and GLP-17–36 may inhibit appetite additively in rodents and man. The mechanism may be through enhanced activation of hypothalamic arcuate neurons. Reduction in energy intake and increase in insulin secretion would make this combination a particularly attractive therapy for patients with type 2 diabetes.
Acknowledgments
We thank Dr. Adam Kennedy, Ms. Tammy Ng, Professor Malcolm Alison, Dr. Chrystalla Orphanides, the human volunteers, and the Sir John McMichael Centre for Clinical Investigation and Research, Hammersmith Hospital.
Footnotes
This work was supported by European Union FP6 Integrated Project Grant (contract LSHM-CT-2003-503041). This publication reflects the authors’ views and not necessarily those of the European Union. The information in this document is provided as is, and no guarantee of warranty is given that the information is fit for any particular purpose. The user thereof uses the information at his/her sole risk and liability. The authors also thank the Medical Research Council and the Wellcome Trust for program grant support. N.M.N., A.J.P., and M.R.D. are Wellcome Trust Clinical Research Training Fellows. C.L.D. was a Diabetes United Kingdom student.
First Published Online September 8, 2005
Abbreviations: AP, Area postrema; ARC, arcuate nucleus; FLI, fos-like immunoreactivity; GLP, glucagon like peptide-1; PB, phosphate buffered; PVN, paraventricular nucleus; PYY, peptide YY; WT, wild type.
Accepted for publication August 22, 2005.
References
Anini Y, Fu-Cheng X, Cuber JC, Kervran A, Chariot J, Roz C 1999 Comparison of the postprandial release of peptide YY and proglucagon-derived peptides in the rat. Pflugers Arch 438:299–306
Batterham RL, Cowley MA, Small CJ, Herzog H, Cohen MA, Dakin CL, Wren AM, Brynes AE, Low MJ, Ghatei, MA, Cone RD, Bloom SR 2002 Gut hormone PYY(3–36) physiologically inhibits food intake. Nature 418:650–654
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 1996 A role for glucagon-like peptide-1 in the central regulation of feeding. Nature 379:69–72
Challis BG, Pinnock SB, Coll AP, Carter RN, Dickson SL, O’Rahilly S 2003 Acute effects of PYY3–36 on food intake and hypothalamic neuropeptide expression in the mouse. Biochem Biophys Res Commun 311:915–919
Halatchev IG, Ellacott KL, Fan W, Cone RD 2004 PYY3–36 inhibits food intake through a melanocortin-4 receptor-independent mechanism. Endocrinology 145:2585–2590
Rodriquez de Fonseca F, Navarro M, Alvarez E, Roncero I, Chowen JA, Maestre O, Gomez R, Munoz RM, Eng J, Blazquez E 2000 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
Dakin CL, Small CJ, Batterham RL, Neary NM, Cohen MA, Patterson M, Ghatei MA, Bloom SR 2004 Peripheral oxyntomodulin reduces food intake and body weight gain in rats. Endocrinology 145:2687–2695
Verdich C, Flint A, Gutzwiller JP, Naslund E, Beglinger C, Hellstrom PM, Long SJ, Morgan LM, Holst JJ, Astrup A 2001 A meta-analysis of the effect of glucagon-like peptide-1 (7–36) amide on ad libitum energy intake in humans. J Clin Endocrinol Metab 86:4382–4389
Szecowka J, Tatemoto K, Rajamaki G, Efendic S 1983 Effects of PYY and PP on endocrine pancreas. Acta Physiol Scand 119:123–126
Yamamoto H, Lee CE, Marcus JN, Williams TD, Overton JM, Lopez ME, Hollenberg AN, Baggio L, Saper CB, Drucker DJ, Elmquist JK 2002 Glucagon-like peptide-1 receptor stimulation increases blood pressure and heart rate and activates autonomic regulatory neurons. J Clin Invest 110:43–52
Sawchenko PE, Swanson LW 1982 The organization of noradrenergic pathways from the brainstem to the paraventricular and supraoptic nuclei in the rat. Brain Res 257:275–325
Batterham RL, Cohen MA, Ellis SM, Le Roux CW, Withers DJ, Frost GS, Ghatei MA, Bloom SR 2003 Inhibition of food intake in obese subjects by peptide YY3–36. N Engl J Med 349:941–948
Wettergren A, Maina P, Boesby S, Holst JJ 1997 Glucagon-like peptide-1 7–36 amide and peptide YY have additive inhibitory effect on gastric acid secretion in man. Scand J Gastroenterol 32:552–555
Gibbs J, Young RC, Smith GP 1973 Cholecystokinin decreases food intake in rats. J Comp Physiol Psychol 84:488–495
Wren AM, Seal LJ, Cohen MA, Brynes AE, Frost GS, Murphy KG, Dhillo WS, Ghatei MA, Bloom SR 2001 Ghrelin enhances appetite and increases food intake in humans. J Clin Endocrinol Metab 86:5992
Lu XY, Shieh KR, Kabbaj M, Barsh GS, Akil H, Watson SJ 2002 Diurnal rhythm of agouti-related protein and its relation to corticosterone and food intake. Endocrinology 143:3905–3915
Lambert PD, Phillips PJ, Wilding JP, Bloom SR, Herbert J 1995 c-fos expression in the paraventricular nucleus of the hypothalamus following intracerebroventricular infusions of neuropeptide Y. Brain Res 670:59–65
Paxinos GT, Watson C 1998 The rat brain in stereotactic co-ordinates. 4th ed. London: Academic Press
Wren AM, Small CJ, Abbott CR, Jethwa PH, Kennedy AR, Murphy KG, Stanley SA, Zollner AN, Ghatei MA, Bloom SR 2002 Hypothalamic actions of neuromedin U. Endocrinology 143:4227–4234
Cohen MA, Ellis SM, le Roux CW, Batterham RL, Park A, Patterson M, Frost GS, Ghatei MA, Bloom SR 2003 Oxyntomodulin suppresses appetite and reduces food intake in humans. J Clin Endocrinol Metab 88:4696–4701
Gutzwiller JP, Goke B, Drewe J, Hildebrand P, Ketterer S, Handschin D, Winterhalder R, Conen D, Beglinger C 1999 Glucagon-like peptide-1: a potent regulator of food intake in humans. Gut 44:81–86
Flint A, Raben A, Blundell JE, Astrup A 2000 Reproducibility, power and validity of visual analogue scales in assessment of appetite sensations in single test meal studies. Int J Obes Relat Metab Disord 24:38–48
Adrian TE, Ferri GL, Bacarese-Hamilton AJ, Fuessl HS, Polak JM, Bloom SR 1985 Human distribution and release of a putative new gut hormone, peptide YY. Gastroenterology 89:1070–1077
Kreymann B, Williams G, Ghatei MA, Bloom SR 1987 Glucagon-like peptide-1 7–36: a physiological incretin in man. Lancet 2:1300–1304
Ghatei MA, Uttenthal LO, Christofides ND, Bryant MG, Bloom SR 1983 Molecular forms of human enteroglucagon in tissue and plasma: plasma responses to nutrient stimuli in health and in disorders of the upper gastrointestinal tract. J Clin Endocrinol Metab 57:488–495
Batt RA 1983 Decreased food intake in response to cholecystokinin (pancreozymin) in wild-type and obese mice (genotype ob/ob). Int J Obes 7:25–29
Martin NM, Small CJ, Sajedi A, Patterson M, Ghatei MA, Bloom SR 2004 Pre-obese and obese agouti mice are sensitive to the anorectic effects of peptide YY(3–36) but resistant to ghrelin. Int J Obes Relat Metab Disord 28:886–893
Dubuc PU, Wilden NJ, Carlisle HJ 1985 Fed and fasting thermoregulation in ob/ob mice. Ann Nutr Metab 29:358–365
Cone RD, Cowley MA, Butler AA, Fan W, Marks DL, Low MJ 2001 The arcuate nucleus as a conduit for diverse signals relevant to energy homeostasis. Int J Obes Relat Metab Disord 25(Suppl 5):S63–S67
Koda S, Date Y, Murakami N, Shimbara T, Hanada T, Toshinai K, Niijima A, Furuya M, Inomata N, Osuye K, Nakazato M 2005 The role of the vagal nerve in peripheral PYY3–36-induced feeding reduction in rats. Endocrinology 146:2369–2375
Talsania T, Anini Y, Siu S, Drucker DJ, Brubaker PL 2005 Peripheral exendin-4 and peptide YY3–36 synergistically reduce food intake through different mechanisms in mice. Endocrinology 146:3748–3756
Gutzwiller JP, Degen L, Matzinger D, Prestin S, Beglinger C 2004 Interaction between GLP-1 and CCK-33 in inhibiting food intake and appetite in men. Am J Physiol Regul Integr Comp Physiol 287:R562–R567(Nicola M. Neary, Caroline J. Small, Mara)
Abstract
Peptide YY (PYY) and glucagon like peptide (GLP)-1 are cosecreted from intestinal L cells, and plasma levels of both hormones rise after a meal. Peripheral administration of PYY3–36 and GLP-17–36 inhibit food intake when administered alone. However, their combined effects on appetite are unknown. We studied the effects of peripheral coadministration of PYY3–36 with GLP-17–36 in rodents and man. Whereas high-dose PYY3–36 (100 nmol/kg) and high-dose GLP-17–36 (100 nmol/kg) inhibited feeding individually, their combination led to significantly greater feeding inhibition. Additive inhibition of feeding was also observed in the genetic obese models, ob/ob and db/db mice. At low doses of PYY3–36 (1 nmol/kg) and GLP-17–36 (10 nmol/kg), which alone had no effect on food intake, coadministration led to significant reduction in food intake. To investigate potential mechanisms, c-fos immunoreactivity was quantified in the hypothalamus and brain stem. In the hypothalamic arcuate nucleus, no changes were observed after low-dose PYY3–36 or GLP-17–36 individually, but there were significantly more fos-positive neurons after coadministration. In contrast, there was no evidence of additive fos-stimulation in the brain stem. Finally, we coadministered PYY3–36 and GLP-17–36 in man. Ten lean fasted volunteers received 120-min infusions of saline, GLP-17–36 (0.4 pmol/kg·min), PYY3–36 (0.4 pmol/kg·min), and PYY3–36 (0.4 pmol/kg·min) + GLP-17–36 (0.4 pmol/kg·min) on four separate days. Energy intake from a buffet meal after combined PYY3–36 + GLP-17–36 treatment was reduced by 27% and was significantly lower than that after either treatment alone. Thus, PYY3–36 and GLP-17–36, cosecreted after a meal, may inhibit food intake additively.
Introduction
THE GUT HORMONES peptide YY (PYY) and glucagon-like peptide (GLP)-1 are cosecreted from endocrine L cells of the small and large bowel. Endogenous concentrations of both hormones are low in the fasting state and rise within 30 min of a meal (1). However, whereas these hormones have been found to inhibit food intake individually (2, 3), their combined effect has yet to be investigated.
It has been reported that peripheral administration of PYY3–36 inhibits appetite in rodents (4, 5) and man (2). Peripheral administration of GLP-17–36 inhibits food intake in rodents (6, 7) and man (8). GLP-17–36 and the GLP-1 receptor agonist exendin-4 also stimulate insulin release and inhibit circulating glucagon (6). In contrast, no changes in insulin and glucagon concentrations occur after PYY3–36 administration (9).
PYY and GLP-1 may mediate these peripheral effects by altering central nervous system appetite circuits. After ip administration of PYY3–36, an increase in c-fos, a marker of neuronal activation, has been reported in the hypothalamic arcuate nucleus (ARC) (2). In contrast, GLP-1 is reported to act primarily on brain stem neurons. After iv exendin-4, c-fos activation was observed in the area postrema (AP) (10), which, like the ARC, has access to the peripheral circulation. c-fos immunoreactivity was also detected in the ARC and paraventricular nucleus (PVN). This hypothalamic activation could be mediated via neuronal projections from the brain stem to the PVN (11) or directly onto ARC neurons.
Both PYY3–36 and GLP-17–36 have been administered to humans. Intravenous infusion of PYY3–36 was associated with a 36% reduction in energy intake at a free buffet meal (2). PYY3–36 has also been reported to inhibit food intake in obese subjects (12). A recent metaanalysis reported that infusion of GLP-17–36 was associated with a small dose-dependent reduction in energy intake in both lean and obese subjects (8). Interestingly, when PYY and GLP-1 were coadministered to man, an additive inhibitory effect on pentagastrin-induced gastric acid secretion was observed (13), although their combined effect on appetite was not examined.
Much work has been performed investigating the effect of exogenous administration of individual gut hormones on appetite to determine their potential physiological role in the control of food intake (4, 14, 15). However, after a meal not one but several gut hormones are elevated, and it may be that the interactions between the changes in levels of different hormones influence food intake more than each hormone individually.
The aim of our study was to investigate whether the combined administration of PYY3–36 and GLP-17–36 was more effective in inhibiting food intake than either peptide alone. First, we examined the effects of coadministration of PYY3–36 and GLP-17–36 in lean and obese rodents. Second, to investigate potential mechanisms, we quantified fos-like immunoreactivity (FLI) in the hypothalamus and brain stem after ip PYY3–36 and GLP-17–36 alone and PYY3–36 + GLP-17–36 coadministration. Finally, we performed a randomized, double-blind, placebo-controlled, crossover trial in healthy volunteers to compare the effects of PYY3–36, GLP-17–36, and PYY3–36 with GLP-17–36 on energy intake in man.
Materials and Methods
Materials
Synthetic human PYY3–36 and GLP-17–36 were purchased from Phoenix Pharmaceuticals (Belmont, CA). Unless otherwise stated, all chemicals were purchased from Merck (Poole, Dorset, UK).
Animals
All animal procedures were approved by the British Home Office Animals (Scientific Procedures) Act 1986 (project license no. 70/5516 and 70/5281). Male mice and rats were maintained in individual cages under controlled temperature (21–23 C) and light (12-h light, 12-h dark cycle; lights on at 0700 h) with ad libitum access to standard chow (RM1 diet, SDS Ltd., Witham, Essex, UK) and water. All animals were handled daily for 7 d before the first study, particularly in the light of recent evidence that peripheral PYY3–36 does not inhibit food intake in stressed animals unaccustomed to their experimental conditions but does so in well-handled, acclimatized animals (5). During the acclimatization period, each animal received two saline injections to minimize stress on the study days.
Intraperitoneal injections
Intraperitoneal injections were administered to mice via a 0.5 ml syringe with a 29-gauge needle (maximum injection volume, 0.1 ml) and rats via a 1 ml syringe with a 25-gauge needle (maximum volume, 0.5 ml). In each study animals received one injection only of equivalent volume for each treatment group. For the combination treatment, GLP-17–36 and PYY3–36 were drawn up in a single syringe. In all feeding studies, food intake was measured 1, 2, 4, 8, and 24 h after injection as previously described (7).
Feeding studies in lean mice
C57BL/6 mice (20–25 g) were injected after a 20-h fast. To determine suitable doses for the combined study, dose-response studies were performed for PYY3–36 and GLP-17–36 individually. For the PYY3–36 study, mice were ip injected with 30, 100, 150, and 300 nmol/kg PYY3–36 (n = 11–12). For the GLP-17–36 study, mice were injected with 30, 100, 300, and 900 nmol/kg GLP-17–36 (n = 10). For the combined study, mice were injected with saline, 100 nmol/kg PYY3–36, 200 nmol/kg PYY3–36, 100 nmol/kg GLP-17–36, 200 nmol/kg GLP-17–36, or 100 nmol/kg PYY3–36 + 100 nmol/kg GLP-17–36 (n = 8–9).
Feeding studies in ob/ob and db/db mice
To investigate these effects in obese rodents, PYY3–36 + and GLP-17–36 were coadministered to ob/ob and db/db mice (both on a C57BL/6 background) in a four-way crossover design. After a 20-h fast, ob/ob mice (mean weight 54.5 ± 1.4 g) and their wild-type (WT) littermates (32.5 ± 1.0 g) and db/db mice (41.8 ± 1.12 g) and their WT littermates (31.0 ± 0.5 g) were ip injected with saline, 200 nmol/kg PYY3–36, 200 nmol/kg GLP-17–36, or 100 nmol/kg PYY3–36 + 100 nmol/kg GLP-17–36. Four studies were performed with 72 h between each study so that each mouse received all four treatments (n = 10 per treatment group).
Feeding studies in rats
To examine the effect of PYY3–36 and GLP-17–36 in another species and during another feeding phase, these peptides were coadministered in rats in the dark phase. Rodents are nocturnal feeders and eat the majority of their food during the night (16). Freely feeding rats (150–200 g) were injected in the hour before lights off. For the PYY3–36 dose-response study, animals received ip saline or 1, 3, or 10 nmol/kg PYY3–36 (n = 12–13). For GLP-17–36, rats received saline or 3, 10, 30, or 100 nmol/kg GLP-17–36 (n = 12–13). To investigate the combined effect of low doses, 1 nmol/kg PYY3–36 and 10 nmol/kg GLP-17–36 were chosen for the joint study (see Results). On two evenings, 72 h apart, animals were randomized and injected ip with saline, 10 nmol/kg GLP-17–36, 20 nmol/kg GLP-17–36, 1 nmol/kg PYY3–36, 2 nmol/kg PYY3–36, or 1 nmol/kg PYY3–36 + 10 nmol/kg GLP-17–36 (total n = 19–20 per group) before the dark phase.
Immunocytochemistry
The protocols for c-fos staining and quantification were similar to those described previously (17). Briefly, rats (240–265 g) were each handled for 5 min/d for 1 wk to minimize the stress of the procedure. After an overnight fast, rats received ip injections of saline, 10 nmol/kg GLP-17–36, 1 nmol/kg PYY3–36, or 1 nmol/kg PYY3–36 + 10 nmol/kg GLP-17–36 (n = 5–6 per group). Ninety minutes later, they were terminally anesthetized with ip sodium pentobarbitone (200 mg) (Rhne Merieux, Harlow, UK) and transcardially perfused with 0.1 M PBS followed by 4% phosphate-buffered (PB)-formalin. The brains were removed and postfixed overnight in PB-formalin and then transferred to PB-sucrose (20% wt/vol) for 48 h. Serial 40-μm coronal brain sections were cut on a freezing microtome and stained for FLI using the avitin-biotin-peroxidase method. Fos-positive nuclei were counted using a light microscope (Eclipse E800; Nikon, Tokyo, Japan) by an experienced member of the research team with reference to the Rat Brain Atlas (18), who was unaware of which treatment had been given. The nuclei examined were the PVN, ARC, ventromedial and dorsomedial of the hypothalamus, and the nucleus of the solitary tract and the AP of the brain stem.
Behavioral studies
To determine whether alterations in behavior were associated with treatments of peptides used in the feeding and c-fos studies, an observation study was performed (19). After an overnight fast, rats (150–170 g) were injected ip with saline, 10 nmol/kg GLP-17–36, 1 nmol/kg PYY3–36, 1 nmol/kg PYY3–36 + 10 nmol/kg GLP-17–36, 10 nmol/kg exendin-4, or 1 M lithium chloride (positive control) (n = 10 per group) and their food replaced. Their behavior was assessed for the subsequent 60 min by two observers who were unaware of which treatment had been given. Each observer recorded behaviors of 30 animals (n = 5 per group per observer). Each rat was observed for 15 sec in every 5 min. The 15-sec periods were subdivided into three 5-sec periods. In each or these periods, the observer selected the behavior that most closely resembled what the rat was doing. These options were feeding, drinking, locomotion, grooming, head down, sleeping, and rearing.
In a further experiment, freely feeding rats were ip injected before the dark phase with saline, 10 nmol/kg GLP-17–36, 1 nmol/kg PYY3–36, 1 nmol/kg PYY3–36 + 10 nmol/kg GLP-17–36, or 10 nmol/kg exendin-4. Infrared beam breaks in two planes x-tot (horizontal movement) and z-tot (vertical movement) were counted for 1 h using an Opto-Max system (Columbus Instruments, Columbus, OH) to assess locomotion.
Human randomized, double-blind, controlled trial
The study was approved by the Hammersmith, Queen Charlotte’s, and Chelsea Ethics Committee (reference no. 2002/6261) and performed in accordance with the principles of the Declaration of Helsinki. Healthy lean volunteers were recruited through advertisement. Exclusion criteria were substance abuse, pregnancy, significant medical or psychiatric illness, or regular medication except the oral contraceptive pill. Twelve subjects were recruited, each of whom gave their informed consent to participate in the study. One subject was withdrawn due to violation of the protocol and another was withdrawn due to nausea on a study morning. Thus, 10 subjects (four men and six women) completed the study. Their ages ranged from 22 to 29 with a mean age of 25.6 ± 0.7 yr, and body mass index ranged from 20.5 to 26.4 with a mean of 23.0 ± 0.7 kg/m2.
Each volunteer received four infusions, saline, GLP-17–36, PYY3–36, or PYY3–36 with GLP-17–36, each on a separate day at least 5 d apart. Subjects fasted from 2100 h on the evening before infusions. The infusion days were run as previously described (20).
Both PYY3–36 and GLP-17–36 were sterile on culture, and limulus amoebocyte lysate assay tests for pyrogen were negative. Two vials were used for each infusion (saline + saline, saline + GLP-17–36, saline + PYY3–36, and PYY3–36 + GLP-17–36). Both vials were dissolved in 2.5 ml hemacel (Beacon, Tunbridge Wells, Kent, UK) to minimize peptide adsorption, diluted in 0.9% saline (Bayer, Newbury, Berkshire, UK) and drawn up in a single syringe (total volume 50 ml). The infusion rates chosen were based on the results of previous studies. PYY3–36 at 0.8 pmol/kg·min was associated with a 36% reduction in food intake (2). The lowest rate of GLP-17–36 that has been shown to inhibit food intake is 0.75 pmol/kg·min (21). We aimed for a submaximal effect for each individual peptide and so selected rates of 0.4 pmol/kg·min for both GLP-17–36 and PYY3–36 infusions and a rate of 0.4 pmol/kg·min PYY3–36 + 0.4 pmol/kg·min GLP-17–36 for the combined infusion. Infusions were given for 120 min.
After 90 min infusion, a buffet lunch was served of preweighed food in excess consisting of chicken or vegetable curry, rice, fruit salad, and sweets. Thirty minutes later, the remaining food was removed and weighed and the infusion discontinued. Blood samples (10 ml) were taken 30 min before the start of the infusion, immediately before the infusion, and then every 30 min until 2 h after the meal. Blood was collected into heparin-coated tubes containing 2000 Kallikrein inhibitor units (0.2 ml) of aprotonin (Bayer). After centrifugation, plasma was separated and stored immediately at –70 C until RIA. Subjects were asked to score subjective nausea on a visual analog scales (0–100 mm) (22) at each time point. Pulse and blood pressure were also measured at these time points. Subjects were allowed to leave 2 h after the meal and instructed to eat ad libitum and complete a food diary for the following 24 h.
RIAs
All samples were assayed in duplicate and single assays to eliminate the effects of interassay variation. PYY, GLP-1, and insulin concentrations were quantified using established in-house RIAs and antibodies (23, 24, 25). Glucose concentrations were measured using a YSI-2300STAT analyzer (Yellow Springs Instruments, Yellow Springs, OH).
Statistical analysis
Results are shown as mean values ± SEM. Data from the animal feeding and immunocytochemistry and behavior studies were analyzed using a one-way ANOVA with a Dunnett’s two-sided post hoc test. For the human infusions, a random effect, repeated-measures ANOVA was performed and within-subject effect examined under a Greenhouse-Geisser correction. In all cases P < 0.05 was considered to be statistically significant.
Results
Coadministration of high-dose PYY3–36 and GLP-1 in lean mice
In the PYY3–36 dose-response study, food intake in the first hour was significantly inhibited by all doses of PYY3–36 administered [food intake 0–1 h (grams): saline, 0.67 ± 0.05; PYY3–36, 30 nmol/kg, 0.49 ± 0.04, 100 nmol/kg, 0.44 ± 0.03, 150 nmol/kg, 0.47 ± 0.02, 300 nmol/kg, 0.47 ± 0.04] (P < 0.05 for all doses vs. saline) (Fig. 1A). PYY appeared to cause maximal feeding inhibition from doses of 100 nmol/kg and above, and this dose was selected for the combined study. Inhibition in cumulative feeding was sustained with all doses for up to 8 h after injection (data not shown).
In the GLP-17–36 dose-response study, food intake in the first hour was significantly inhibited by 100, 300, and 900 nmol/kg GLP-17–36 but not significantly by the 30 nmol/kg dose [food intake 0–1 h (grams): saline, 0.85 ± 0.05; GLP-17–36, 30 nmol/kg, 0.72 ± 0.07, 100 nmol/kg, 0.45 ± 0.03, 300 nmol/kg, 0.31 ± 0.05, 900 nmol/kg, 0.20 ± 0.05] (P < 0.05 for 100, 300, and 900 nmol/kg GLP-17–36 vs. saline) (Fig. 1B). In contrast with the PYY study, greater inhibition of feeding was observed with increasing doses of GLP-17–36 (P < 0.05 for 100 vs. 30 nmol/kg GLP-17–36 and for 300 vs. 100 nmol/kg GLP-17–36). A dose of 100 nmol/kg GLP-17–36 was selected for the combined study, which achieved an approximately 50% reduction in food intake in the first hour. Inhibition in cumulative feeding was sustained up to 8 h after injection after 100, 300, and 900 nmol/kg GLP-17–36.
The doses for the coadministration study were as follows: saline, 100 nmol/kg PYY3–36, 200 nmol/kg PYY3–36, 100 nmol/kg GLP-17–36, 200 nmol/kg GLP-17–36 and 100 nmol/kg PYY3–36 + 100 nmol/kg GLP-17–36. All peptides inhibited feeding in the first hour, compared with saline. Coadministration of PYY3–36 with GLP-17–36 led to significantly greater feeding inhibition than all other groups [food intake 0–1 h (grams): saline, 0.74 ± 0.07; 100 nmol/kg GLP-17–36, 0.33 ± 0.05; 200 nmol/kg GLP-17–36, 0.33 ± 0.07; 100 nmol/kg PYY3–36, 0.60 ± 0.03; 200 nmol/kg PYY3–36, 0.51 ± 0.04; 100 nmol/kg PYY3–36 + 100 nmol/kg GLP-17–36, 0.07 ± 0.07: P < 0.05 for combined groups vs. all other groups] (Fig. 2). Feeding inhibition after the combination treatment was sustained up to 24 h after injection [food intake 0–24 h (grams): saline, 7.0 ± 0.2; 100 nmol/kg GLP-17–36, 6.5 ± 0.2; 200 nmol/kg GLP-17–36, 7.2 ± 0.2; 100 nmol/kg PYY3–36, 6.9 ± 0.2; 200 nmol/kg PYY3–36, 6.8 ± 0.2; 100 nmol/kg PYY3–36 + 100 nmol/kg GLP-17–36, 6.1 ± 0.3: P < 0.05 for combined group vs. saline].
Coadministration of PYY3–36 and GLP-17–36 in ob/ob and db/db mice
Mice were randomized to receive saline, 200 nmol/kg GLP-17–36, 200 nmol/kg PYY3–36, or 100 nmol/kg PYY3–36 + 100 nmol/kg GLP-17–36. Coadministration of PYY3–36 and GLP-17–36 in ob/ob mice and their WT littermates led to significant inhibition of feeding, compared with saline and both peptides alone [food intake 0–1 h (grams), ob/ob mice: saline, 0.48 ± 0.09; 200 nmol/kg GLP-17–36, 0.25 ± 0.03; 200 nmol/kg PYY3–36, 0.32 ± 0.05; 100 nmol/kg PYY3–36 + 100 nmol/kg GLP-17–36, 0.14 ± 0.03; P < 0.05 for combined group vs. all other groups; food intake 0–1 h (grams), WT mice: saline, 1.2 ± 0.08; 200 nmol/kg GLP-17–36, 0.68 ± 0.09; 200 nmol/kg PYY3–36, 0.78 ± 0.069; 100 nmol/kg PYY3–36 + 100 nmol/kg GLP-17–36, 0.33 ± 04; P < 0.05 for combined group vs. all other groups]. The percentage reduction in food intake with the coadministration treatment, compared with saline, was similar in ob/ob (–74 ± 4%) and WT mice (–70 ± 6%) (Fig. 3A). Reduction in cumulative food intake was sustained for up to 24 h in the WT mice and up to 8 h in the ob/ob mice (data not shown).
Coadministration of PYY3–36 and GLP-17–36 was also associated with significant reductions in food intake in db/db mice and their WT littermates, compared with saline and both peptides alone [food intake 0–1 h (grams), db/db mice: saline, 0.59 ± 0.10; 200 nmol/kg GLP-1, 0.45 ± 0.05; 200 nmol/kg PYY3–36, 0.36 ± 0.04; 100 nmol/kg PYY3–36 + 100 nmol/kg GLP-1, 0.20 ± 0.04; P < 0.05 for combined group vs. all other groups; food intake 0–1 h (grams), WT mice: saline, 1.17 ± 0.09; 200 nmol/kg GLP-17–36, 0.99 ± 0.05; 200 nmol/kg PYY3–36, 0.76 ± 0.07; 100 nmol/kg PYY3–36 + 100 nmol/kg GLP-17–36, 0.41 ± 0.07; P < 0.05 for combined group vs. all other groups]. The percentage reduction in food intake with the coadministration treatment, compared with saline, was similar in db/db (–65 ± 7%) and WT mice (–63 ± 6%) (Fig. 3B). Reduction in cumulative food intake was sustained for up to 8 h in both WT and db/db mice (data not shown).
Coadministration of low-dose PYY3–36 and GLP-17–36 in rats
PYY3–36 and GLP-17–36 dose-response studies were performed in freely feeding rats in the dark phase. In the PYY3–36 study, food intake in the first hour of the dark phase was significantly inhibited by 3 and 10 nmol/kg PYY3–36 but not significantly by the 1 nmol/kg dose [food intake 0–1 h (grams): saline, 4.9 ± 0.2; PYY3–36, 1 nmol/kg, 4.5 ± 0.5; 3 nmol/kg, 3.8 ± 0.3; 10 nmol/kg, 3.5 ± 0.3; P < 0.05 for 3 and 10 nmol/kg PYY3–36 vs. saline] (Fig. 4A). In the GLP-17–36 study, food intake in the first hour was significantly inhibited by 30 and 100 nmol/kg GLP-17–36 [food intake 0–1 h (grams): saline, 5.0 ± 0.3; GLP-17–36, 3 nmol/kg 4.9 ± 0.5; 10 nmol/kg, 4.5 ± 0.3; 30 nmol/kg, 4.1 ± 0.3; 100 nmol/kg, 3.8 ± 0.3; P < 0.05 for 30 and 100 nmol/kg GLP-17–36 vs. saline] (Fig. 4B).
The doses of peptides selected for the combined study were 3-fold lower than the minimum doses found to inhibit food intake significantly in the first hour. Rats received ip injections of saline, 1 nmol/kg PYY3–36, 2 nmol/kg PYY3–36, 10 nmol/kg GLP-17–36, 20 nmol/kg GLP-17–36, or 1 nmol/kg PYY3–36 + 10 nmol/kg GLP-17–36. Food intake was significantly inhibited only in the PYY3–36 + GLP-17–36 group (Fig. 5) [food intake 0–1 h (grams): saline, 4.1 ± 0.2; 10 nmol/kg GLP-17–36, 3.9 ± 0.3; 20 nmol/kg GLP-17–36, 4.2 ± 0.3; 1 nmol/kg PYY3–36, 4.2 ± 0.3; 2 nmol/kg PYY3–36, 4.1 ± 0.2; 1 nmol/kg PYY3–36 + 10 nmol/kg GLP-17–36, 2.8 ± 0.2]. There was significant inhibition of food intake with the combined treatment, compared with all other groups (P < 0.02). This effect was most marked in the first hour of the dark phase. Cumulative food intake was inhibited in the coadministration group for up to 4 h, compared with saline (P < 0.05) [food intake 0–4 h (grams): saline, 9.8 ± 0.3; 10 nmol/kg GLP-17–36, 9.7 ± 0.4; 20 nmol/kg GLP-17–36, 10.0 ± 0.2; 1 nmol/kg PYY3–36, 9.9 ± 0.5; 2 nmol/kg PYY3–36, 10.1 ± 0.4; 1 nmol/kg PYY3–36 + 10 nmol/kg GLP-17–36, 8.9 ± 0.4]. At these low doses, no differences in food intake were observed between groups at later time points (data not shown).
Immunocytochemistry
To elucidate a possible mechanism for the additive inhibitory feeding effect of PYY and GLP-1, c-fos was quantified after low dose individual peptide and combined peptide administration. Doses of PYY and GLP-1 that had been found not to alter food intake individually but had led to significant reduction in food intake after coadministration were chosen. Rats received ip injections of saline, 10 nmol/kg GLP-17–36, 1 nmol/kg PYY3–36, or 1 nmol/kg PYY3–36 + 10 nmol/kg GLP-17–36. In the hypothalamus, there were no significant differences in the number of FLI counts in the PVN, and ventromedial and dorsomedial of the hypothalamus between treatment groups (data not shown). At these low doses, no significant differences were seen in the ARC with either PYY3–36 or GLP-17–36 alone, compared with saline. However, an 85% increase in the number of FLI counts was observed in the ARC after the coadministration treatment, compared with saline (P < 0.05) (Fig. 6). Moreover, after the combined PYY3–36 + GLP-17–36 treatment, there were significantly more FLI than after PYY3–36 alone (P < 0.05) and a trend for more FLI after GLP-17–36 alone (P = 0.05) (FLI-positive neurons in the ARC per section: saline, 35.4 ± 6.18; GLP-17–36, 42.9 ± 1.2; PYY3–36, 36.2 ± 4.4; PYY3–36 + GLP-17–36, 65.6 ± 10.0).
In the brain stem, no significant differences in FLI counts were observed between these low-dose treatments in the nucleus of the solitary tract or AP (data not shown).
Behavioral studies
To determine whether behavioral alteration were associated with doses of PYY3–36 and GLP-17–36 at which an additive feeding effect was observed, an observation study was performed. Fasted rats were ip injected with saline, 10 nmol/kg GLP-17–36, 1 nmol/kg PYY3–36, 1 nmol/kg PYY3–36 + 10 nmol/kg GLP-17–36, 10 nmol/kg exendin-4, or 1 M lithium chloride (positive control) (n = 10 per group). The groups treated with exendin-4 and lithium chloride had significantly fewer feeding episodes and significantly more head-down episodes, compared with saline (P < 0.05) (Fig. 7). However, no significant differences in behavior were observed after PYY3–36, GLP-17–36, or PYY3–36 + GLP-17–36.
In further study, locomotion was assessed in freely feeding rats before the dark phase after ip saline, 10 nmol/kg GLP-17–36, 1 nmol/kg PYY3–36, 1 nmol/kg PYY3–36 + 10 nmol/kg GLP-17–36, or 10 nmol/kg exendin-4 (n = 9–10 per group). No significant differences in horizontal movement (x-tot) were observed after the combined PYY3–36 + GLP-17–36 treatment, but there was a slight reduction in vertical movement (z-tot), compared with saline (P = 0.04) (Table 1). There were significant reductions in both horizontal and vertical movement after exendin-4, compared with saline (P < 0.01 for both comparisons).
Coadministration of PYY3–36 and GLP-17–36 in human volunteers
Infusion of PYY3–36 (0.4 pmol/kg·min) was associated with a 15% reduction in energy intake at the buffet meal, compared with saline, although this reduction achieved only borderline statistical significance (P = 0.07). There was no significant difference in energy intake with GLP-17–36 infusion (0.4 pmol/kg·min), in keeping with previously published results (8). With the combined PYY3–36 and GLP-17–36 infusion, caloric intake at the buffet meal was reduced by 27% (buffet meal energy intake (kilojoules): saline, 4174 ± 285; GLP-17–36, 3978 ± 340; PYY3–36, 3551 ± 441; PYY3–36 + GLP-17–36, 3033 ± 279; percentage change from saline day: GLP-17–36, –5 ± 5%; PYY3–36, –15 ± 9%; PYY3–36 + GLP-17–36, –27 ± 3%) (Fig. 8). There was a statistical difference in energy intake among the four infusion groups (P < 0.01). The buffet caloric intake on the PYY3–36 + GLP-17–36 day was significantly less than on all other infusion days (P < 0.05).
Cumulative energy intake over 24 h (buffet + food diary) was significantly reduced with the combined PYY3–36 + GLP-17–36, compared with all the other groups (cumulative 24-h energy intake (megajoules): saline, 13.7 ± 0.6; GLP-17–36, 13.6 ± 1.3; PYY3–36, 13.5 ± 0.9; PYY3–36 + GLP-17–36, 12.0 ± 0.7, P < 0.05 for PYY3–36 + GLP-17–36 vs. all other groups; percentage change from saline day: GLP-17–36, –2 ± 7%; PYY3–36, –1 ± 5%; PYY3–36 + GLP-17–36, –13 ± 5%). Analysis of the visual analog scales revealed no significant differences in nausea scores between infusion days [nausea score after 90 min infusion (millimeters): saline, 9.4 ± 0.83; GLP-17–36, 1.2 ± 0.05; PYY3–36, 3.8 ± 0.18; PYY3–36 + GLP-17–36, 6.3 ± 2.2, P=NS].
There were significant increases in plasma PYY levels with PYY3–36 and PYY3–36 + GLP-17–36 infusions and plasma GLP-1 levels with GLP-17–36 and PYY3–36 + GLP-17–36 infusions (Tables 2 and 3). Interestingly, our GLP-17–36 infusion inhibited circulating PYY (PYY picomoles per liter at 90 min: saline infusion, 10.3 ± 1.2; GLP-17–36 infusion, 7.7 ± 0.8; P < 0.05). However, no significant changes in circulating GLP-1 were observed during PYY3–36 infusion.
Fasting plasma insulin was increased by GLP-17–36 infusion at 90 min, compared with saline, and no changes in insulin secretion were observed with PYY3–36 alone. Whereas the coadministration PYY3–36 + GLP-17–36 infusion enhanced insulin secretion, there was no additional secretion, compared with GLP-17–36 alone (insulin picomoles per liter at 90 min: saline, 53.2 ± 4.9; GLP-17–36, 63.0 ± 3.5; PYY3–36, 51.2 ± 5.5; PYY3–36 + GLP-17–36, 66.7 ± 4.3; P < 0.05 for GLP-17–36 and combined PYY3–36 + GLP-17–36 infusions vs. saline) (Table 4). Consistent with these changes in insulin levels, fasting plasma glucose was reduced by GLP-17–36 infusion and the joint coadministration infusion but unchanged by PYY3–36 alone (glucose millimoles per liter at 90 min: saline, 4.6 ± 0.1; GLP-17–36, 4.2 ± 0.1; PYY3–36, 4.5 ± 0.1; PYY3–36 + GLP-17–36, 4.1 ± 0.1; P < 0.01 for GLP-17–36 and combined PYY3–36+GLP-17–36 infusions vs. saline) (Table 5). Postprandial insulin and glucose levels were not measured due to the variability of energy intake at the buffet meal.
Discussion
In this paper we investigated the effects of the coadministration of PYY3–36 and GLP-17–36 on feeding in rodents and humans.
PYY3–36 and GLP-17–36 inhibited food intake individually in mice and rats. PYY3–36 appeared to cause maximal feeding inhibition at doses of 100 nmol/kg and above. In contrast, feeding inhibition associated with GLP-17–36 was similar to that seen with CCK-8, with increasing inhibition seen with increasing doses (14). In the dose-response studies with low doses, the minimum doses of PYY3–36 and GLP-17–36 found to inhibit food intake were 3 and 30 nmol/kg, respectively. Both these results are consistent with our previously published data (2, 7). In both high- and low-dose coadministration studies, PYY3–36 and GLP-17–36 administered together inhibited food intake significantly more than either peptide alone or more than double the dose of either peptide. Thus, PYY3–36 and GLP-17–36 appeared to inhibit feeding additively.
PYY3–36 and GLP-17–36 were coadministered in genetically obese mice. Examination of the food intake in the saline-injected groups revealed that the ob/ob and db/db mice ate less in the first hour of refeeding than their WT littermates. This somewhat surprising finding is consistent with previously published work (26, 27) and suggests that thermogenic adaptation to fasting reduced the subsequent energy requirement of the obese mouse models (28). Coadministration of PYY3–36 and GLP-17–36 treatment led to significantly greater reduction in feeding than twice the dose of either peptide alone in the ob/ob and db/db and WT mice. Thus, PYY3–36 and GLP-17–36 inhibited food intake additively in obese as well as lean rodents.
We quantified c-fos afterPYY3–36, GLP-17–36, and PYY3–36 + GLP-17–36 administration to determine a possible mechanism for their additive effect on feeding. There was a significant increase in c-fos in the ARC after low-dose combination PYY3–36 + GLP-17–36 treatment but no changes in the other treatment groups. Therefore, PYY3–36 and GLP-17–36 may inhibit food intake through additive stimulation of ARC neurons. However, c-fos mapping cannot identify the pathway(s) through which neuronal activation was mediated. The ARC neuronal activation observed in our study could have been mediated through the blood-brain barrier via the median eminence (29) or the vagus nerve and brain stem (30).
There were no significant differences in observed behavior in rats after PYY3–36, GLP-17–36, or PYY3–36 + GLP-17–36 at doses at which the combination inhibited feeding. In the locomotion study, no significant differences in horizontal beam breaks were observed after the combined PYY3–36 + GLP-17–36 treatment, but there were marginally fewer vertical beam breaks. In their recent publication, Talsania et al. (31) suggested that low-dose exendin-4 together with PYY3–36 may increase the suppression of food intake in rodents without inducing significant side effects. There were no differences in nausea scores between infusions in our human study. Taken together these results suggest that behavioral change is unlikely to account for the feeding inhibition observed with coadministration of PYY3–36 and GLP-17–36. Thus, these peptides may have a specific effect on feeding independent of behavioral change.
In our human study, GLP-17–36 alone (0.4 pmol/kg·min) did not lead to a significant change in energy intake. In human subjects the lowest reported infusion rate of GLP-17–36 associated with a change in energy intake is 0.75 pmol/kg·min (21). However, incretin effects have been observed with a lower rate of GLP-17–36 of 0.5 pmol/kg·min (24). In the current study, we found a small but statistically significant reduction in fasting glucose levels and an increase in the fasting insulin concentration with GLP-17–36 infusion.
PYY3–36 infusion (0.4 pmol/kg·min) was associated with a 15% reduction in energy intake, compared with saline infusion, although this did not reach statistical significance. The magnitude of this reduction is in keeping with the larger reduction of 36% reported after higher-rate PYY infusions (0.8 pmol/kg·min) (2).
Coadministration of PYY3–36 with GLP-17–36 was associated with a 27% reduction in energy intake from the buffet meal. The combination was more effective in inhibiting appetite then either peptide alone. Cumulative energy intake over 24 h, including the buffet meal, was reduced by 13% after combined infusion, compared with the saline infusion, and was significantly less than that after all other treatment groups. In contrast with the additive inhibitory effect on energy intake, the incretin effect of GLP-17–36 appeared unchanged with the addition of PYY3–36.
Our study involving coadministration of PYY3–36 and GLP-17–36 suggests an additive inhibitory feeding effect. Interestingly, the GLP-1 receptor agonist, exendin-4, and PYY3–36 have been reported to inhibit food intake synergistically in rodents (31). However, it appears that not all combinations of gut hormones behave in this way. Indeed, it has recently been reported that whereas infusion of CCK-33 or GLP-17–36 inhibited food intake when administered individually, no enhanced inhibition was observed when they were infused together (32). Further work is needed to determine whether the other combinations of gut hormones lead to additive alterations in food intake.
We observed a reduction of endogenous PYY levels with GLP-17–36 infusion. This has been previously demonstrated at a higher dose of peripheral GLP-17–36 administration (32). Inhibition of endogenous PYY could result in a reduction in the anorexic action of GLP-1. In contrast, as in previous work (12), no changes in endogenous GLP-1 concentrations were observed after PYY3–36 infusion.
In summary, we have demonstrated that PYY3–36 and GLP-17–36 may inhibit appetite additively in rodents and man. The mechanism may be through enhanced activation of hypothalamic arcuate neurons. Reduction in energy intake and increase in insulin secretion would make this combination a particularly attractive therapy for patients with type 2 diabetes.
Acknowledgments
We thank Dr. Adam Kennedy, Ms. Tammy Ng, Professor Malcolm Alison, Dr. Chrystalla Orphanides, the human volunteers, and the Sir John McMichael Centre for Clinical Investigation and Research, Hammersmith Hospital.
Footnotes
This work was supported by European Union FP6 Integrated Project Grant (contract LSHM-CT-2003-503041). This publication reflects the authors’ views and not necessarily those of the European Union. The information in this document is provided as is, and no guarantee of warranty is given that the information is fit for any particular purpose. The user thereof uses the information at his/her sole risk and liability. The authors also thank the Medical Research Council and the Wellcome Trust for program grant support. N.M.N., A.J.P., and M.R.D. are Wellcome Trust Clinical Research Training Fellows. C.L.D. was a Diabetes United Kingdom student.
First Published Online September 8, 2005
Abbreviations: AP, Area postrema; ARC, arcuate nucleus; FLI, fos-like immunoreactivity; GLP, glucagon like peptide-1; PB, phosphate buffered; PVN, paraventricular nucleus; PYY, peptide YY; WT, wild type.
Accepted for publication August 22, 2005.
References
Anini Y, Fu-Cheng X, Cuber JC, Kervran A, Chariot J, Roz C 1999 Comparison of the postprandial release of peptide YY and proglucagon-derived peptides in the rat. Pflugers Arch 438:299–306
Batterham RL, Cowley MA, Small CJ, Herzog H, Cohen MA, Dakin CL, Wren AM, Brynes AE, Low MJ, Ghatei, MA, Cone RD, Bloom SR 2002 Gut hormone PYY(3–36) physiologically inhibits food intake. Nature 418:650–654
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 1996 A role for glucagon-like peptide-1 in the central regulation of feeding. Nature 379:69–72
Challis BG, Pinnock SB, Coll AP, Carter RN, Dickson SL, O’Rahilly S 2003 Acute effects of PYY3–36 on food intake and hypothalamic neuropeptide expression in the mouse. Biochem Biophys Res Commun 311:915–919
Halatchev IG, Ellacott KL, Fan W, Cone RD 2004 PYY3–36 inhibits food intake through a melanocortin-4 receptor-independent mechanism. Endocrinology 145:2585–2590
Rodriquez de Fonseca F, Navarro M, Alvarez E, Roncero I, Chowen JA, Maestre O, Gomez R, Munoz RM, Eng J, Blazquez E 2000 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
Dakin CL, Small CJ, Batterham RL, Neary NM, Cohen MA, Patterson M, Ghatei MA, Bloom SR 2004 Peripheral oxyntomodulin reduces food intake and body weight gain in rats. Endocrinology 145:2687–2695
Verdich C, Flint A, Gutzwiller JP, Naslund E, Beglinger C, Hellstrom PM, Long SJ, Morgan LM, Holst JJ, Astrup A 2001 A meta-analysis of the effect of glucagon-like peptide-1 (7–36) amide on ad libitum energy intake in humans. J Clin Endocrinol Metab 86:4382–4389
Szecowka J, Tatemoto K, Rajamaki G, Efendic S 1983 Effects of PYY and PP on endocrine pancreas. Acta Physiol Scand 119:123–126
Yamamoto H, Lee CE, Marcus JN, Williams TD, Overton JM, Lopez ME, Hollenberg AN, Baggio L, Saper CB, Drucker DJ, Elmquist JK 2002 Glucagon-like peptide-1 receptor stimulation increases blood pressure and heart rate and activates autonomic regulatory neurons. J Clin Invest 110:43–52
Sawchenko PE, Swanson LW 1982 The organization of noradrenergic pathways from the brainstem to the paraventricular and supraoptic nuclei in the rat. Brain Res 257:275–325
Batterham RL, Cohen MA, Ellis SM, Le Roux CW, Withers DJ, Frost GS, Ghatei MA, Bloom SR 2003 Inhibition of food intake in obese subjects by peptide YY3–36. N Engl J Med 349:941–948
Wettergren A, Maina P, Boesby S, Holst JJ 1997 Glucagon-like peptide-1 7–36 amide and peptide YY have additive inhibitory effect on gastric acid secretion in man. Scand J Gastroenterol 32:552–555
Gibbs J, Young RC, Smith GP 1973 Cholecystokinin decreases food intake in rats. J Comp Physiol Psychol 84:488–495
Wren AM, Seal LJ, Cohen MA, Brynes AE, Frost GS, Murphy KG, Dhillo WS, Ghatei MA, Bloom SR 2001 Ghrelin enhances appetite and increases food intake in humans. J Clin Endocrinol Metab 86:5992
Lu XY, Shieh KR, Kabbaj M, Barsh GS, Akil H, Watson SJ 2002 Diurnal rhythm of agouti-related protein and its relation to corticosterone and food intake. Endocrinology 143:3905–3915
Lambert PD, Phillips PJ, Wilding JP, Bloom SR, Herbert J 1995 c-fos expression in the paraventricular nucleus of the hypothalamus following intracerebroventricular infusions of neuropeptide Y. Brain Res 670:59–65
Paxinos GT, Watson C 1998 The rat brain in stereotactic co-ordinates. 4th ed. London: Academic Press
Wren AM, Small CJ, Abbott CR, Jethwa PH, Kennedy AR, Murphy KG, Stanley SA, Zollner AN, Ghatei MA, Bloom SR 2002 Hypothalamic actions of neuromedin U. Endocrinology 143:4227–4234
Cohen MA, Ellis SM, le Roux CW, Batterham RL, Park A, Patterson M, Frost GS, Ghatei MA, Bloom SR 2003 Oxyntomodulin suppresses appetite and reduces food intake in humans. J Clin Endocrinol Metab 88:4696–4701
Gutzwiller JP, Goke B, Drewe J, Hildebrand P, Ketterer S, Handschin D, Winterhalder R, Conen D, Beglinger C 1999 Glucagon-like peptide-1: a potent regulator of food intake in humans. Gut 44:81–86
Flint A, Raben A, Blundell JE, Astrup A 2000 Reproducibility, power and validity of visual analogue scales in assessment of appetite sensations in single test meal studies. Int J Obes Relat Metab Disord 24:38–48
Adrian TE, Ferri GL, Bacarese-Hamilton AJ, Fuessl HS, Polak JM, Bloom SR 1985 Human distribution and release of a putative new gut hormone, peptide YY. Gastroenterology 89:1070–1077
Kreymann B, Williams G, Ghatei MA, Bloom SR 1987 Glucagon-like peptide-1 7–36: a physiological incretin in man. Lancet 2:1300–1304
Ghatei MA, Uttenthal LO, Christofides ND, Bryant MG, Bloom SR 1983 Molecular forms of human enteroglucagon in tissue and plasma: plasma responses to nutrient stimuli in health and in disorders of the upper gastrointestinal tract. J Clin Endocrinol Metab 57:488–495
Batt RA 1983 Decreased food intake in response to cholecystokinin (pancreozymin) in wild-type and obese mice (genotype ob/ob). Int J Obes 7:25–29
Martin NM, Small CJ, Sajedi A, Patterson M, Ghatei MA, Bloom SR 2004 Pre-obese and obese agouti mice are sensitive to the anorectic effects of peptide YY(3–36) but resistant to ghrelin. Int J Obes Relat Metab Disord 28:886–893
Dubuc PU, Wilden NJ, Carlisle HJ 1985 Fed and fasting thermoregulation in ob/ob mice. Ann Nutr Metab 29:358–365
Cone RD, Cowley MA, Butler AA, Fan W, Marks DL, Low MJ 2001 The arcuate nucleus as a conduit for diverse signals relevant to energy homeostasis. Int J Obes Relat Metab Disord 25(Suppl 5):S63–S67
Koda S, Date Y, Murakami N, Shimbara T, Hanada T, Toshinai K, Niijima A, Furuya M, Inomata N, Osuye K, Nakazato M 2005 The role of the vagal nerve in peripheral PYY3–36-induced feeding reduction in rats. Endocrinology 146:2369–2375
Talsania T, Anini Y, Siu S, Drucker DJ, Brubaker PL 2005 Peripheral exendin-4 and peptide YY3–36 synergistically reduce food intake through different mechanisms in mice. Endocrinology 146:3748–3756
Gutzwiller JP, Degen L, Matzinger D, Prestin S, Beglinger C 2004 Interaction between GLP-1 and CCK-33 in inhibiting food intake and appetite in men. Am J Physiol Regul Integr Comp Physiol 287:R562–R567(Nicola M. Neary, Caroline J. Small, Mara)