Early phase insulin infusion and muscarinic blockade in obese and lean subjects.

The effect of early phase insulin on postprandial levels of insulin, C-peptide, glucose, and glucagon was investigated in lean (n = 10) and obese (n = 12) subjects. Subjects underwent four conditions during ingestion of a meal (600 kcal): 1) saline infusion; 2) 10-min insulin infusion simultaneously with meal ingestion (0.24 U bolus, 15 mU. m(-2). min(-1)); 3) atropine infusion (0.4 mg/m(2) bolus, 0.4 mg. m(-2). h for 4 h); 4) insulin and atropine infusion. Blood samples were taken for 3.5 h. Insulin infusion had no effect on postprandial insulin levels in either population but significantly reduced postprandial glucose in the obese subjects (P < 0.05). Obese subjects with elevated postprandial glucose levels in the presence of muscarinic blockade exhibited a decline in glucose with insulin supplementation. Atropine reduced postprandial insulin levels in both groups, with a greater attenuation in the obese (P < 0.01), but postprandial glucose levels were also significantly reduced, suggesting that atropine inhibited gastric emptying. Thus the effects of muscarinic blockade on postprandial insulin levels cannot be evaluated. These data suggest that insulin supplementation during the preabsorptive time period may contribute to glucoregulation in the obese population.

Muscarinic blockade inhibits gastric emptying of mixed-nutrient meal: effects of weight and gender.

We compared the vagal contribution to gastric emptying in lean and obese subjects by monitoring gastric emptying of a meal during muscarinic blockade. Lean (n = 6) and obese subjects (n = 6) underwent two treatments: 1) saline infusion and 2) atropine infusion [0.4 mg/m2 bolus, 0.4 mg. (m2)-1. h-1] for 2 h, initiated 30 min before ingestion of a 600-kcal breakfast (64% carbohydrate, 23% fat, 13% protein) composed of orange juice (labeled with Indium-111), egg sandwich (labeled with Technetium-99m), cereal, milk, and banana. Anterior and posterior images were taken every 90 s for 90 min using a dual-headed camera. Atropine significantly delayed emptying of both solid (P < 0.007) and liquid (P < 0.002). Obese subjects exhibited a greater delay in liquid emptying during muscarinic blockade compared with lean subjects (P < 0.02). Female subjects exhibited a slower rate of gastric emptying and were less sensitive to atropine. These data suggest that obese subjects exhibit altered gastric cholinergic activity compared with lean subjects and that gender differences in gastric emptying rate may be due to differences in autonomic tone.

Palatability and dietary restraint: effect on cephalic phase insulin release in women.

The palatability of food has been shown to influence the cephalic phase reflexes. To determine if food palatability affects the magnitude of cephalic phase insulin release (CPIR) in humans, normal-weight women were asked to list foods that they found palatable and unpalatable. Subjects then stayed overnight in the hospital on two separate days. On each morning following an overnight fast, an intravenous line was inserted and arterialized venous blood drawn for the measurement of plasma insulin and glucose. Blood samples were taken prior to and following a modified sham-feed. Subjects sham-fed the palatable or unpalatable foods (that they had previously identified) for a 2-min period. Foods were administered in a counterbalanced order. During the protocol, hunger and food palatability were monitored. The Three Factor Eating Questionnaire was administered to assess eating attitudes. No significant difference in the magnitude of cephalic phase insulin release was found between the two treatments. However, a statistically significant correlation (r = 0.61, p < 0.05) was found between an individual's degree of dietary restraint as measured by the Three Factor Eating Questionnaire and the magnitude of CPIR. These data suggest that the sensory attributes of food may play less of a role in modulating CPIR than an individual's psychological attitude towards food.

Oral sensory stimulation improves glucose tolerance in humans: effects on insulin, C-peptide, and glucagon.

In animals, bypassing the oropharyngeal receptors by intragastric administration of glucose results in glucose intolerance. To determine whether the absence of oral sensory stimulation alters glucose tolerance in humans, we monitored plasma levels of glucose and hormones after intragastric administration of glucose, with and without subjects tasting food. Plasma glucose area under the curve (AUC) was significantly lower after oral sensory stimulation (3,433 +/- 783 vs. 5,643 +/- 1,397 mg.dl-1. 195 min-1; P < 0.03; n = 8). Insulin and C-peptide AUCs were higher during the first one-half of the sampling period (insulin, 5,771 +/- 910 vs. 4,295 +/- 712 microU. ml-1.75 min-1; P < 0.05; C-peptide, 86 +/- 10 vs. 66 +/- 9 ng.ml-1. 75 min-1; P < 0.03) and lower during the second one-half of the sampling period compared with the control condition (1,010 +/- 233 vs. 2,106 microU.ml-1. 120 min-1; P < 0.025; 31 +/- 8 vs. 56 +/- 18 ng.ml-1. 120 min-1; P < 0.05; insulin and C-peptide, respectively). Oral sensory stimulation markedly increased plasma glucagon compared with the control condition (1,258 +/- 621 vs. -2,181 +/- 522 pg.ml-1. 195 min-1; P < 0.002). These data provide evidence in humans that oral sensory stimulation influences glucose metabolism and suggest that the mechanisms elicited by this cephalic stimulation are necessary for normal glucose homeostasis.

Sweet taste: effect on cephalic phase insulin release in men.

To determine whether sweet-tasting solutions are effective elicitors of cephalic phase insulin release (CPIR) in humans, two studies were conducted using nutritive and nonnutritive sweeteners as stimuli. Normal weight men sipped and spit four different solutions: water, aspartame, saccharin, and sucrose. A fifth condition involved a modified sham-feed with apple pie. The five stimuli were administered in counterbalanced order, each on a separate day. In study 1, subjects tasted the stimuli for 1 min (n = 15) and in study 2 (n = 16), they tasted the stimuli for 3 min. Arterialized venous blood was drawn to establish a baseline and then at 1 min poststimulus, followed by every 2 min for 15 min and then every 5 min for 15 min. In both study 1 and study 2, no significant increases in plasma insulin were observed after subjects tasted the sweetened solutions. In contrast, significant increases in plasma insulin occurred after the modified sham-feed with both the 1 min and 3 min exposure. These results suggest that nutritive and nonnutritive sweeteners in solution are not adequate stimuli for the elicitation of CPIR.

Oral sensory stimulation in men: effects on insulin, C-peptide, and catecholamines.

To investigate the effect of oral sensory stimulation on cephalic phase insulin release (CPIR) and other compounds associated with glucose metabolism, a modified sham feed was used in which male subjects (n = 15) tasted, chewed, and then expectorated the food stimulus. Subjects remained fasted, sham fed, or ingested food on separate days over a 5-day period. After four baseline blood samples, poststimulus samples were taken every 2 min for 14 min and then every 15 min for 2.25 h. Increases in plasma insulin and C-peptide were found during the cephalic phase time period (0-10 min poststimulus) on the sham-fed and fed conditions when compared with fasted values. Glucose, norepinephrine, epinephrine, and free fatty acids were not affected by oral sensory stimulation or food ingestion during the cephalic phase period. The magnitude and profile of CPIR were similar after sham feeding and food ingestion. This study demonstrates that oral sensory stimulation can elicit CPIR in humans independently of changes in blood glucose.

Cephalic-phase insulin in obese and normal-weight men: relation to postprandial insulin.

Cephalic-phase insulin release (CPIR) and its relation to postprandial insulin release were examined in 18 normal-weight and 15 obese men. When the insulin data were expressed as absolute differences from baseline values, obese subjects exhibited significantly greater CPIR than normal-weight subjects (normals, 8.7 +/- 2.1 microU/mL/10 min; obese, 13.4 +/- 4.3 microU/mL/10 min; P < .01). Obese subjects were then separated into groups depending on their fasting insulin levels. This showed that only those subjects with elevated fasting insulin levels exhibited greater CPIR than normal subjects, and suggested that previous reports of exaggerated CPIR in the obese are merely a reflection of a basal hypersecretion of insulin. However, when insulin values were expressed as percentages of baseline, no significant differences between normal-weight and obese subjects were found, although a trend toward an attenuated response was observed in the obese group as a whole (normals, 81.6 +/- 19.1 microU/mL/10 min; obese, 51.3 +/- 16.1 microU/mL/10 min). A significant correlation between cephalic-phase and postprandial insulin release was found in normal-weight subjects (r = .62, P < .05), but not in obese subjects (r = .02, P < .9).

Cephalic phase insulin release in normal weight males: verification and reliability.

The existence and reliability of cephalic phase insulin release (CPIR) were tested in 20 normal weight males. Each subject was challenged three times with the same food stimulus over a 5-day period. Four baseline blood samples were taken at 5-min intervals before food ingestion and then every 2 min for 16 min postingestion. Significant increases in plasma insulin were found at 4 min postingestion on each trial day. CPIR was found to be highly reproducible between trials (r = 0.83; P less than 0.001). Fifty percent of the subjects exhibited a significant increase of plasma insulin above their own baseline mean on the first trial, whereas 75 and 72% exhibited increases on trials 2 and 3, respectively. Only two subjects (10%) did not demonstrate a response on any trial. A significant decline in plasma glucose was observed at 4 min postingestion on trials 2 and 3. No significant changes in plasma glucagon were found during any trial day. This study confirms a reliable CPIR in normal weight males.

The effect of protein or carbohydrate breakfasts on subsequent plasma amino acid levels, satiety and nutrient selection in normal males.

Normal subjects were fed protein or carbohydrate breakfasts. Both meals were in the form of a chocolate pudding and had similar sensory qualities. At lunchtime subjects were allowed to select from a buffet. The protein breakfast had a greater satiating power than the carbohydrate breakfast, but there was no difference in overall selection of protein or carbohydrate at lunchtime. However, the carbohydrate breakfast did decrease selection of apple, the only pure carbohydrate food available at lunchtime. In a second experiment changes in plasma amino acid levels were studied after subjects received carbohydrate breakfasts containing 0, 4, 8 or 12% protein, or a danish pastry. Only the 0% protein breakfast increased tryptophan availability to the brain. These experiments were performed to test the hypothesis that alterations in brain 5-hydroxytryptamine, brought about by dietary alterations in brain tryptophan, regulate selection of protein and carbohydrate. The results suggest that this mechanism was not operating in our experiments.

Acute effect of protein or carbohydrate breakfasts on human cerebrospinal fluid monoamine precursor and metabolite levels.

Patients with normal pressure hydrocephalus who had three lumbar punctures during 1 week ingested either water, a protein breakfast, or a carbohydrate breakfast 2.5 h before each of the lumbar punctures. The CSF was analyzed for biogenic amine precursors and metabolites. The protein meal raised CSF tyrosine levels, a finding consistent with animal data, but did not alter those of tryptophan or any of the biogenic amine metabolites. The carbohydrate meal increased CSF 3-methoxy-4-hydroxyphenylethylene glycol, an unexplained finding. The carbohydrate meal did not affect CSF tryptophan, tyrosine, 5-hydroxyindoleacetic acid, or homovanillic acid. Our results support the idea that in humans protein or carbohydrate meals do not alter plasma amino acid levels sufficiently to cause appreciable changes in CNS tryptophan levels or 5-hydroxytryptamine synthesis.

Tryptophan availability, 5HT synthesis and 5HT function.

1. Tryptophan increases 5HT synthesis, but the extent to which it increases 5HT release and therefore 5HT function is unclear. 2. The possibility that increased 5HT levels will lead to increased 5HT release is enhanced when 5HT neurons are firing at a higher rate. The rate of firing of 5HT neurons is increased as the level of behavioral arousal increases. Thus, altered tryptophan levels will be more likely to influence brain function at higher levels of arousal. 3. In the rat, tryptophan administration increased CSF 5HT appreciably when the animals were aroused by being put in the dark, but not when they were left in a lighted room. 4. In monkeys, the level of behavioral arousal does seem to influence the effect of altered tryptophan levels on aggression. This is consistent with the fact that altered tryptophan levels had no effect on aggression in normal subjects, but that tryptophan had a therapeutic effect in pathologically aggressive patients. 5. The confusing literature on the antidepressant effect of tryptophan can, to some extent, be explained by considering the circumstances in which tryptophan administration will lead to increases in 5HT release as well as increases in 5HT synthesis. 6. Although in some circumstances tryptophan can decrease pain perception by activation of spinal 5HT pathways, when it was given to postoperative patients it attenuated morphine analgesia by activation of a 5HT pathway in the brain. 7. The effect of altered tryptophan levels depend critically on the circumstances in which it is given.

The effect of lowering plasma tryptophan on food selection in normal males.

The effects of a tryptophan deficient amino acid mixture on food selection were studied using a double-blind counterbalanced crossover design in normal male subjects. The subjects ingested tryptophan deficient or nutritionally balanced amino acid mixtures in the morning after an overnight fast. Five hours after the tryptophan deficient amino acid mixture plasma tryptophan was only 19% of the level found five hours after the nutritionally balanced amino acid mixture. After both mixtures subjects were allowed to select lunch from a buffet. The tryptophan-deficient mixture was associated with a modest but significant decline in protein selection with no significant alteration in selection of carbohydrate, fat or total kcal. Our results suggest that 5-hydroxytryptamine is involved in the control of protein selection in humans.

Effects of carbohydrate and protein administration on rat tryptophan and 5-hydroxytryptamine: differential effects on the brain, intestine, pineal, and pancreas.

We compared the acute effects of intragastric administration of protein and carbohydrate on tryptophan and 5-hydroxytryptamine (5HT) in rat brain, pineal, intestine, and pancreas. Protein decreased and carbohydrate increased brain indoles relative to water-infused controls. These effects were due to competition between the large neutral amino acids for entry into the brain. This competition does not exist in the pineal. The macronutrients had no effect on pineal tryptophan metabolism. In the intestine, protein resulted in higher tryptophan levels as compared to controls, owing to absorption of tryptophan in the protein. However intestinal 5HT levels were influenced by factors other than precursor availability. Pancreatic indoles were affected in a similar manner to the brain indoles. Competition between the large neutral amino acids for entry into the pancreas was also indicated by the finding that valine administration lowered brain and pancreatic tryptophan, but not the levels in the intestine and pineal. It remains to be seen whether the decrease in pancreatic 5HT after a protein meal and the increase after carbohydrate modulate the release of insulin and glucagon.

Serotonin in cisternal cerebrospinal fluid of the rat: measurement and use as an index of functionally active serotonin.

A simple, selective reverse-phase HPLC-fluorometric method is described for the determination of serotonin (5HT) in cisternal CSF of the rat. The mean (+/- SE) value of CSF 5HT observed in control adult rats was 457 +/- 83 pg/ml (N = 16). In an attempt to validate the measure as an index of extracellular, or functionally active, 5HT, groups of animals were treated with fenfluramine, amitriptyline, pargyline, pargyline plus tryptophan, and 5-hydroxytryptophan plus carbidopa. In all cases CSF 5HT appeared to reflect well the presumed effects of the agents on extracellular levels of 5HT. CSF 5HT was superior in this regard to brain 5HT, brain 5HIAA, or CSF 5HIAA levels. The measurement of cisternal CSF 5HT would appear to offer a convenient index of functionally active 5HT.