The Multiple Faces of Glucagon-Like Peptide-1—Obesity, Appetite, and Stress: What Is Next? A Review

By itself, glucagon-like peptide-1(GLP-1) appears to be an excellent drug for appetite control and the treatment of obesity. Unfortunately, few enzymes, such as IV dipeptidyl peptidase and renal excretin, degrade and render GLP-1 inactive within minutes. A receptor agonist, exendin-4, with a longer biological half-life than GLP-1, has been tried. Subcutaneous injection of exendin-4 or continuous IV injection of GLP-1 warrants further research and investigation.     

Glycaemic response to foods: impact on satiety and long-term weight regulation

Should future nutritional recommendations for the general population take into account the notion of glycaemic index (GI)? This question is all the more legitimate as the glycaemic response to foods seems to be a factor that affects satiety and could therefore affect food intake. The aim of this review was to evaluate whether altering the glycaemic response per se can modulate satiety and to assess the short-term and long-term consequences. A systematic review of human intervention studies was performed. Confounding factors that may influence both GI and satiety were taken into consideration when selecting the studies. Thirty-two studies were thus selected and analysed. There is evidence from the short-term studies (1 day) that low-glycaemic foods or meals have higher satietogenic effect than high-glycaemic foods or meals. This substantiates claims such as 'low-GI foods help one to feel fuller for longer than equivalent high-GI foods'. The mechanisms involved may be the specific effect of blood glucose levels on satiety (glucostatic theory) and other stimuli (e.g. peptides) involved in the control of appetite. In some studies, however it seems difficult to tease out the separate effect of the lowering of postprandial glycaemia per se and fibres. Because of the increasing number of confounding variables in the available long-term studies, it is not possible to conclude that low-glycaemic diets mediate a health benefit based on body weight regulation. The difficulty of demonstrating the long-term health benefit of a satietogenic food or diet may constitute an obstacle to the recognition of associated claims.     

Oxytocin innervation of caudal brainstem nuclei activated by cholecystokinin

The integration of 'long-term' adiposity signaling with the 'short-term' meal-related signal cholecystokinin (CCK) is proposed to involve descending hypothalamic projections to areas of the caudal brainstem (CBS) that regulate the amount of food consumed during a single meal. One such projection extends from cell bodies in the hypothalamic paraventricular nucleus (PVN) to the nucleus tractus solitarius (NTS), where cells that respond to peripheral CCK are concentrated. Candidate neuronal cell types that may comprise this PVN-NTS projection includes those expressing corticotropin-releasing hormone (CRH) or oxytocin. We therefore sought to determine whether oxytocin or CRH axons are preferentially located in close anatomical proximity to neurons of the NTS that are activated by peripheral administration of CCK, as determined by immunocytochemical staining for Fos protein. Rats received injections of either an anorexic dose of CCK (8 nmol/kg, i.p.) or vehicle and were perfused 2 h later with 4% paraformaldehyde. Immunocytochemistry was performed on cryostat sections (14 microm) of caudal brainstem, using a polyclonal antibody to Fos protein and either a monoclonal antibody to oxytocin or a polyclonal antibody to CRH. As expected, CCK administration significantly increased the numbers of Fos-positive neurons by 489% (p<0.01) and 400% (p<0.01), respectively, in the medial and gelatinosus subdivisions of the NTS. These same regions received dense oxytocin axon innervation, whereas CRH immunoreactivity was not as prevalent in these areas. In areas outside the NTS, such as the dorsal motor nucleus of the vagus (DMV), Fos activation was absent despite a dense oxytocin and CRH innervation. To investigate whether CCK-induced reductions of food intake require intact oxytocin signaling, we performed a separate study in which CCK injection was preceded by injection into the fourth ventricle of an oxytocin receptor antagonist [d(CH(2))(5), Tyr (Me)(2), Orn(8)]-vasotocin (OVT). This study showed CCK was 23% and 22% less effective at inhibiting food intake at 30 min (p<0.05) and 1 h (p<0.05) food intake, respectively, in the presence of OVT. Taken together, the data indicate that oxytocin axons within the descending pathway from the PVN to the NTS are anatomically positioned to interact with NTS neurons that respond to vagally mediated peripheral CCK signals such as those that occur following ingestion of a meal. These findings support the hypothesis that oxytocin exerts a tonic stimulatory effect on the response of key neurons within the NTS to CCK and further reduce meal size.     

Cholecystokinin-induced satiety in weanling rats

Intraperitoneal injections fo the synthetic C-terminal octapeptide of cholecystokinin (CCK) into fasted 21-day-old weanling rats produced a significant suppression of intake of solid and liquid diets. Similar injections had no effect on water consumption of thirsty weanlings. The early appearance of CCK-induced satiety is consistent with the idea that this mechanism is responsible for the effectiveness of gastrointestinal factors in determining intake of very young rats.     

Analysis of the effects of intra-accumbens SKF-38393 and LY-171555 upon the behavioural satiety sequence

The outcome of intra-accumbens infusions of the dopamine D₁ receptor family agonist SKF-38393 and the D₂ receptor family agonist LY-171555 upon measures taken during the behavioural satiety sequence was assessed (0.01 µg, 0.1 µg, 1.0 µg in each case). Each drug was infused either separately, or together as a co-infusion in order to examine the functional relationship between these dopamine receptor subtypes within the nucleus accumbens. Measures of feeding did not change following infusions of SKF-38393 or LY-171555, whether infused separately or together. However, following separate infusion of the lowest dose tested of each drug (0.01 µg), the onset of resting was advanced. Moderate to high doses of SKF-38393 and LY-171555 (0.1 µg, 1.0 µg) infused separately resulted in a marked increase in activity at the expense of resting. Co-infusion of 0.01 µg of each drug also resulted in a dramatic increase in activity. Thus, measures of feeding behaviour were unchanged following excitation of D₁ and D₂ dopamine receptor families within the nucleus accumbens. In marked contrast, locomotor behaviour appeared to be under the potent synergistic control of these receptor families.     

Truncal and hepatic vagotomy reduce suppression of feeding by jejunal lipid infusions

Two experiments investigated mechanisms underlying the decrease in food intake produced by lipid infusions into the jejunum. In Experiment 1, male Sprague-Dawley rats with truncal abdominal vagotomy (TVx), selective hepatic-branch vagotomy (HVx), or sham vagotomy received repeated 7 h infusions of linoleic acid (LA), corn oil (CO), or saline through indwelling jejunal catheters. Cumulative food intake was measured at 1, 3, 6, and 23 h. LA and, to a lesser extent, CO suppressed food intake in excess of the caloric value of the load. This effect was eliminated by TVx, which significantly attenuated the suppression of intake produced by both lipids at 3 and 6 h and also at 23 h when LA was infused. HVx attenuated suppression at 23 h on tests with LA and at 3 and 6 h on CO tests. Experiment 2 showed that jejunal infusion of LA had no effect on multi-unit activity of afferent fibers in the left splanchnic nerve in anesthetized rats. Thus, these results provide further evidence that satiating effects of intestinal lipid infusions are mediated by the vagal fibers, some of which lie within the hepatic branch. However, because significant suppression of food intake remained after TVx, and because of the negative results of Experiment 2, these lipid infusions engage as yet unidentified mechanisms independent of the vagus.     

Hypothalamic and locus coeruleus self-stimulation are decreased by cholecystokinin

Reinforcement from self-stimulation in the lateral hypothalamus has been interpreted as food-related reward. According to the reward-aversion hypothesis, both food and stimulation of the lateral hypothalamus are aversive to the sated animal. Thus, lateral hypothalamic self-stimulation rates should be low when an animal is sated. Because cholecystokinin has been observed to produce behavioral satiety, we measured the effect of intraperitoneal cholecystokinin injections (10–80 Ivy dog units/kg) on self-stimulation rates in rats. Cholecystokinin did decrease lateral hypothalamic self-stimulation rates in a dose-related manner. However, locus coeruleus self-stimulation rates were similarly decreased. Since the locus coeruleus has not been implicated in the control of feeding, these results suggest that satiety decreases the reinforcement value of brain stimulation, regardless of whether the structures stimulated are related to feeding.     

The satiating efficiency of foods

Experiments were undertaken to test the general hypothesis that some foods are more satiating than others, to find a mechanism for their differential satiating efficiencies, and to determine whether certain soups had a high enough satiating efficiency to recommend their addition to a meal as a way of reducing total caloric intake of that meal. In the first experiment it was found that intake of a test meal was lower after a large preload of tomato soup than after a small preload in women, but not in men. However, the total energy intake (soup plus test meal) was no less with meals which included the large soup preload than it was with meals that did not include a preload. Therefore adding a normal portion of tomato soup to a meal would not reduce its total energy intake. We noted the interesting incidential finding that total energy intake (i.e., preload plus test meal) of the meals which contained the larger amount of soup was less than the total energy intake of the meals which contained a combination of crackers, jelly, and juice. In the second experiment we confirmed this finding by showing that when equal weights of tomato soup preloads and a preload of crackers, cheese, and apple juice, which contains more energy, were given, total energy intake was less in meals which included soup. Therefore, substituting tomato soup for a more calorically dense first course could reduce total energy intake of that meal. In the third experiment, the hypothesis suggested by the second was confirmed. Two soups were more satiating than crackers, cheese, and juice. When two calorie levels were used for each preload, it was shown that calorie for calorie, these soups decreased intake of the test meal more than crackers, cheese, and juice. In the fourth experiment we showed that the mechanism for this differential satiating efficiency is not readily attributable to either bulk related factors or fat content. We suggest that the differential satiating efficiencies are related to differences in nutrient dispersion, orosensory cues, or temperature. Finally, reductions in intake were accompanied by reductions in the initial rate of eating and not by increases in the rate of deceleration. This reduction was small but consistent and suggests that foods which are more satiating reduce intake by decreasing desire to eat (i.e., hunger), not by accelerating the onset of meal termination (i.e., satiety). In fact the duration of meals was unaffected by the preloads.