Biphasic insulin secretion after intravenous but not after intraportal CCK-8 infusion in rats.

The effects of physiological doses of sulfated cholecystokinin-8 (CCK-8) on insulin secretion were investigated in unrestrained unanesthetized rats. The routes of administration were intravenous or intraportal infusion. Intravenous infusion (0.33-5.0 micrograms CCK-8.kg-1.20 min-1) resulted in a biphasic response pattern consisting of a fast 1st-min rise in plasma insulin concentration and a slower second phase that lasted throughout the infusion. The first phase showed the same amplitude with all amounts of CCK-8 administered in this study, whereas the second phase exhibited dose dependency. Blood glucose levels were lowered during all infusions of CCK-8, although the second phase of insulin release was absent with the lowest dose. These results suggest a strong stimulatory effect of CCK-8 on the pancreatic beta-cells, probably by changing the set point for glucose. The described effects of intravenous administration of CCK-8 cannot be produced when the infusion is given into the portal vein. Only very high concentrations of CCK-8 (15 micrograms.kg-1.20 min-1) produced a small increase in plasma insulin levels, indicating a strong CCK-8-eliminating mechanism in the liver. These results indicate that 1) CCK-8 evokes biphasic insulin release and a concomitant drop in glucose levels, and 2) CCK-8 acting on the beta-cell in vivo is not of intestinal origin but is probably released by the pancreatic vagal branch.

Role of the sympathoadrenal system in exercise-induced inhibition of insulin secretion. Effects of islet transplantation.

The present study was designed to investigate the mechanism leading to inhibition of insulin release during exercise. To investigate the influence of circulating epinephrine and norepinephrine, these catecholamines were infused intravenously in resting islet-transplanted and control rats. The role of neural influences on insulin release was investigated by a swimming exercise study in islet-transplanted and control rats, before and after adrenodemedullation. Streptozotocin-induced diabetic Albino Oxford rats received 5 microliters islet tissue into the portal vein, resulting in return of normal basal glucose and insulin levels. Transplanted and control animals were provided with two permanent heart catheters to sample blood and to give infusions. Infusion of epinephrine and norepinephrine did not result in inhibition of plasma insulin levels. Blood glucose levels, as well as nonesterified fatty acids and insulin levels in plasma, were similar in both groups. After the infusion study, the animals were subjected to strenuous swimming. During exercise, plasma insulin levels decreased not only in controls, but also in the islet-transplanted group. Blood glucose and plasma catecholamine responses were identical in both groups. After adrenodemedullation, epinephrine was not detectable and the exercise-induced decrease of insulin was not affected. These results indicate that circulating epinephrine and norepinephrine in physiological concentrations do not cause inhibition of insulin secretion. Since the exercise-induced inhibition of insulin secretion is still present in rats with islet grafts, it seems reasonable to suggest that sympathetic neural influences are responsible for the inhibition of insulin release during exercise and that transplanted islets are sympathetically reinnervated.

Central leptin stimulates corticosterone secretion at the onset of the dark phase.

Leptin, a hormone secreted by adipose tissue in proportion to body adiposity, is proposed to be involved in the central nervous regulation of food intake and body weight. In addition, evidence is emerging that leptin regulates neuroendocrine and metabolic functions as well, presumably via its action in the central nervous system (CNS). To investigate this regulatory effect of leptin, we infused 3.5 microg of human leptin directly into the third cerebral ventricle (i3vt) of lean male Long-Evans rats, 90 min before the onset of their dark phase. Before and after infusion, blood samples were withdrawn through indwelling catheters for assessment of hormonal (plasma corticosterone, insulin, leptin), autonomic (plasma norepinephrine, epinephrine), and metabolic (plasma glucose) parameters. I3vt leptin caused an increase in plasma corticosterone and plasma leptin levels relative to the control condition. The effects of i3vt leptin on corticosterone secretion became particularly apparent after the onset of the dark phase. The results of the present study indicate that i3vt leptin stimulates the hypothalamo-pituitary-adrenal (HPA) axis, particularly when rats normally encounter their largest meals. These results are consistent with the possibility that high circulating leptin levels may underlie the increased activity of the HPA axis that is generally characteristic of human obesity and most animal models of obesity.

The modulatory effect of the hypothalamus on glucagon and insulin secretion in the rat.

Rats were provided both with brain cannulas, to permit infusion of fluids in the ventromedial and lateral hypothalamic areas (resp. VMH and LH), and with two heart catheters. In this way infusions of fluids and withdrawal of blood could be done in unanesthetized free moving animals. Infusion of norepinephrine (NE) in the VMH elicited glucagon release during the whole period of NE infusion whereas insulin levels did not change. This glucagon release could not be suppressed by α- and ß-receptor blockade but it was suppressed by hexamethonium, a blocking agent of signal transmission in peripheral ganglia of the autonomic nervous system. On the other hand, infusion of NE in the LH elicited insulin release during the whole period of NE infusion whereas glucagon levels remained unchanged. It is argued that [1] the sympathetic nervous system is involved in glucagon release, [2] no α- and ß-receptor mechanisms are involved during this glucagon release, [3] the parasympathetic nervous system is involved in insulin release during noradrenergic LH stimulation.

Neuroendocrine mechanisms involved in regulation of body weight, food intake and metabolism.

Body weight regulation is the result of food intake and energy expenditure. The central nervous system (CNS), and in particular, the hypothalamus, controls food intake as well as metabolism, the latter mainly by autonomic effects on the islet of Langerhans, hepatocytes and adipocytes. Body weight, more precisely body fat content, is probably controlled by a feedback mechanism in which insulin, released from the B cell of the islet of Langerhans, plays a key role. The islet of Langerhans is an intricate neuroendocrine unit in which the release of glucagon, insulin, and somatostatin from A, B, and D cells, respectively, is controlled by the CNS via a rich autonomic innervation. In addition, the endocrine cells of the pancreas influence each other by paracrine actions. The CNS control of the islets shapes the plasma insulin and blood glucose profiles during the circadian cycle and thereby regulates the nutrient flow to the different tissues in the body. Thus, the CNS structures involved in regulation of body weight and food intake control also metabolism. The mechanisms contributing to match food intake and the needs of metabolism are discussed.

Overfeeding-induced obesity in rats: insulin sensitivity and autonomic regulation of metabolism.

The metabolic consequences of the development of obesity and the underlying mechanisms were investigated. For this purpose, male rats were overfed for 5 weeks through long-term gastric catheters. Permanent cardiac cannulas implanted before the overfeeding period allowed frequent blood sampling and infusions without disturbing the rats. Hyperalimented rats became grossly obese, displayed elevated basal plasma norepinephrine (NE) concentrations, and developed hyperinsulinemia and insulin insensitivity, but remained normoglycemic and preserved normal intravenous (IV) glucose tolerance. During physical exercise (ie, 15 minutes of swimming), obese rats displayed exaggerated increases in blood glucose concentrations, whereas plasma free fatty acid (FFA) responses were blunted. These alterations were probably due to decreased NE release by the sympathetic nervous system during exercise and to altered tissue responsivity to adrenergic stimulation. The latter was demonstrated by infusions of catecholamines in the resting state. Responses to mild stress were increased in obese animals, as indicated by increased responses of plasma epinephrine (E) and corticosterone during handling and first contact with water. The results of the present study indicate that overfeeding induces changes in the sympathetic control of metabolism and insulin secretion. Whereas elevated NE levels in the basal state probably reflect increased energy expenditure, the pattern of nutrient mobilization during exercise is directed toward sparing of fats.

Hyperinsulinemia and glucose tolerance in obese rats with lesions of the ventromedial hypothalamus: dependence on food intake and route of administration.

This study was performed to investigate the consequences of developing obesity on glucose homeostasis in animals showing hyperphagia plus vagal hyperinsulinemia and rats that were normophagic and hyperinsulinemic. Male rats were lesioned in the ventromedial hypothalamus (VMH) and kept either under ad libitum or absolute (oral or intragastrical) pair-feeding conditions for 4 weeks. Hyperphagic rats, as well as normophagic VMH rats, became obese, but only ad lib-fed obese rats displayed glucose intolerance to intravenous (IV) glucose infusions. Orally pair-fed VMH rats also showed normal oral and intragastric glucose tolerance, but in intragastrically fed VMH animals and controls, oral and intragastric glucose tolerance was decreased. These results indicate that (1) obesity as a consequence of VMH lesions is not dependent on hyperphagia, confirming earlier reports, and also independent of the ingestion of bulk meals. (2) beta-cell release of insulin to IV glucose infusion is not sufficient when hyperphagia and vagally mediated hyperinsulinemia coincide, and is therefore dependent on several factors; and (3) oral glucose intolerance develops when preabsorptive reflexes are blunted, irrespective of whether the animals were hyperinsulinemic or not.

Metabolic and hormonal responses to adrenoceptor antagonists in 48-hour-starved exercising rats.

The influence of 48 hours of starvation on sympathoadrenal regulation of nutrient utilization was investigated in rats. To assess the role of alpha- and beta-adrenoceptors, rats were studied during alpha- and beta-blockade. Energy metabolism was measured using indirect calorimetry before, during, and after moderate swimming exercise (approximately 60% maximal O2 consumption [VO2max]). Additionally, blood samples were taken for determination of nutrient and hormone concentrations. In 48-hour-starved rats, under baseline conditions, there was a reduction in energy expenditure (EE) accompanied by a shift toward fat oxidation (fat-ox) in comparison to fed rats. Exercise-induced responses in EE, fat-ox, and carbohydrate oxidation (CHO-ox) did not differ from those in fed rats. In starved rats, a stronger response to exercise of the sympathoadrenal system was observed. In comparison to control 48-hour-starved rats, blockade of alpha- and beta-adrenoceptors led to a reduction in the exercise-induced increase in EE and fat-ox. The rate of CHO-ox was slightly reduced after blockade of either adrenoceptor type. Alpha-blockade prevented the exercise-induced increase in blood glucose. Plasma free fatty acid (FFA) was not affected. Blood lactate, plasma insulin, norepinephrine (NOR), and epinephrine (EPI) were increased after alpha-blockade. Due to beta-blockade, exercise-induced increases in glucose and FFA were prevented. Blood glucose even declined below the baseline value. EPI showed an exaggerated increase, and NOR showed a smaller increase. Results obtained in starved rats support the idea that alpha-adrenoceptor blockade-induced changes in energy metabolism are the result of a diminished oxygen supply due to diminished circulation.(ABSTRACT TRUNCATED AT 250 WORDS)

Metabolic and hormonal responses to adrenoceptor antagonists in exercising rats.

alpha- and beta-adrenoceptors play a key role in the regulation of nutrient supply to working muscles during exercise. To assess their influence in the regulation of substrate utilization, rats were studied during alpha- or beta-adrenoceptor blockade. Energy metabolism was studied by means of indirect calorimetry before, during, and after moderate swimming exercise. Blood samples were taken for the determination of nutrient and hormone concentrations. In addition, central venous blood samples were withdrawn for determination of blood gases, pH, and total hemoglobin concentration (c/Hb). alpha- and beta-adrenoceptor blockade decreased the rates of energy expenditure (EE) and fat oxidation (fat-ox) during and after swimming in comparison to swimming without adrenoceptor blockade. The oxidation of carbohydrates (CHO-ox) was increased in both cases. alpha-Blockade prevented the exercise-induced increase in blood glucose, plasma free fatty acids (FFA) were not affected, and plasma insulin, norepinephrine (NOR), epinephrine (EPI), and lactate were markedly increased. beta-adrenoceptor blockade prevented the exercise-induced increases in blood glucose and FFA. EPI increased slightly more than and NOR less than in the control experiment. The exercise-induced decrease in insulin was more pronounced after beta-blockade. alpha-Blockade caused a less pronounced decrease in venous oxygen saturation (SO2) and tension (PO2) than in the control experiment. The exercise-induced increase in carbon dioxide tension (PCO2) was almost absent. After beta-blockade, venous SO2 and PO2 decreased more and PCO2 increased more than in the control experiment. It is concluded that both alpha and beta-blockade restrict the rate of EE during exercise.(ABSTRACT TRUNCATED AT 250 WORDS)

Noradrenergic and cholinergic reinnervation of islet grafts in diabetic rats.

Grafted islets become denervated due to the islet transplantation procedure. The aim of the present study was 1) to examine whether islet grafts in the liver, the spleen, and under the kidney capsule in rats become reinnervated following the transplantation and experimental procedures used in our laboratory, 2) whether there is any difference in reinnervation at these different sites, and 3) how these results relate to previous physiological experiments. Isogeneic isolated islets were transplanted into diabetic Albino Oxford rats, resulting in normoglycaemia. After at least 5 wk, graft-receiving organs were removed and several antibodies were employed to detect insulin, neuron-specific proteins, and cholinergic and noradrenergic nerve fibers. Islets in all three receiving organs contained viable insulin-positive B-cells. Neuron-specific enolase (NSE) as well as the growth-associated protein B-50 was observed at all sites. The cholinergic marker choline acetyltransferase (ChAT) was localized in islets grafts at all sites, but with the lowest density in the spleen. Staining for the noradrenergic markers tyrosine hydroxylase (TH) and dopamine-beta-hydroxylase (DBH) was observed in islet grafts at all sites with the lowest density in grafts under the kidney capsule. All these neurochemical substances were most frequently observed in fibers associated with blood vessels, which may be the route along which nerves grow into the graft. It can be concluded that 1) islet grafts in the liver, in the spleen and under the kidney capsule become reinnervated; 2) the innervation pattern of the islet grafts differs only slightly from that in the control pancreatic islets; and 3) in combination with our previously physiological data, we can conclude that these nerve fibers are, at least partly, functionally active.

Control of insulin secretion and glucose homeostasis in exercising diabetic rats with intrasplenic or kidney subcapsular islet grafts.

This study was designed 1) to investigate mechanisms of insulin secretion during exercise after transplantation of islets in the spleen and under the kidney capsule, and 2) to compare these organs as transplantation site regarding an adequate portal or systemic delivery of insulin and glucose homeostasis during exercise. Diabetic rats were provided with 5 microL isogenic islet tissue in the spleen or under the kidney capsule, which results in normoglycemia, and were submitted to a swimming test. Portal plasma insulin levels were higher than simultaneously sampled systemic insulin levels in the control and in the intrasplenic islet grafted group, but not in the kidney subcapsular islet-grafted group. Plasma portal and systemic insulin levels decreased, and glucose levels increased during exercise in all groups. The exercise-induced increase in levels of catecholamines was larger in systemic than in portal plasma, suggesting catecholamine extraction by the lungs or intestines. The experiments were repeated after removing of adrenal medulla, resulting in nondetectable or very low plasma adrenaline levels. Despite these low adrenaline levels, insulin levels decreased during exercise. The results indicate that 1) the exercise-induced reduction of insulin secretion is not mediated by circulating adrenaline, but is probably under control of the sympathetic nervous system, which could be the result of reinnervation of the transplanted islets. 2) Although a portal-systemic insulin gradient was absent in rats with kidney subcapsular islet grafts, the absence of a difference in glucose homeostasis during exercise between the sites revealed that all investigated sites are preferential to transplant islets.

Hypothalamically-induced insulin release and its potentiation during oral and intravenous glucose loads.

Male Wistar rats were provided with bilateral cannulas in the lateral hypothalamic area (LHA) and cannulas in the left and right jugular vein. Freely moving rats provided in this way with cannulas were infused with transmitters in the LHA and with various substances in the blood circulation during simultaneous sampling of blood without disturbing the animals. Infusion of norepinephrine (NE) in the LHA resulted in increased insulin levels while plasma glucagon and blood glucose were nearly not affected. This LHA mediated insulin release was suppressed by atropine injection in the blood circulation suggesting a vagal contribution to the observed phenomenon. Administration of either an oral or i.v. glucose load during noradrenergic stimulation of the LHA elicited an exaggerated insulin response when compared to their controls. This LHA potentiated insulin response during an oral and i.v. glucose load could be suppressed by atropinization of the rats. It is concluded that meal-related stimuli are relayed to the NE-stimulated area of the LHA and that these stimuli modulate the output from this area of the LHA that is concerned with the release of insulin.

The course of paraventricular hypothalamic efferents to autonomic structures in medulla and spinal cord.

By application of the anterograde transport technique of Phaseolus vulgaris leuco-agglutinin the descending autonomic projection of the paraventricular hypothalamic nucleus was investigated. The Phaseolus lectin technique allowed the detection of the cells of origin in the paraventricular PVN, the precise position of two distinct descending axon pathways and the detailed morphology of terminal structures in midbrain, medulla oblongata and spinal cord.

Hormonal and metabolic effects of paraventricular hypothalamic administration of neuropeptide Y during rest and feeding.

To investigate the role of neuropeptide Y (NPY) in the paraventricular nucleus of the hypothalamus (PVN) in the regulation of autonomic outflow, hormonal (plasma insulin and catecholamines), metabolic (blood glucose and plasma free fatty acids) and cardiovascular (heart rate and main arterial pressure) indices were measured before, during, and after bilateral infusion of NPY (1.0, 0.2, 0.04 micrograms in 1 microliter synthetic CSF) into the PVN of conscious resting rats. Administration of the highest dose (1.0 microgram/microliter) caused bradycardia and reduced circulating norepinephrine levels without effecting circulating fuels, insulin or epinephrine. In a second experiment, feeding-induced changes in hormonal and metabolic indices were assessed after NPY administration (1.0 microgram/microliter) into the PVN. During and after feeding, NPY enhanced the feeding-induced insulin response (P < 0.01) and attenuated the feeding-induced norepinephrine response (P < 0.05). The results of the present study suggest that stimulation of NPY receptors in the PVN decreases sympathetic activity and increases parasympathetic activity in resting conditions, and that these effects are potentiated during feeding.

Structural characteristics of endogenous sugar acids and relations to feeding modulation.

Structural specificity among short-chain organic acids for effects on feeding behavior, blood glucose and insulin was investigated by infusion of 1 exogenous and 6 endogenous derivatives into the rat third cerebral ventricle. Glyceric acid (GEA) (1.0 mumol), 3,4-dihydroxybutanoic acid gamma-lactone (3,4-DB) and 3,4,5-trihydroxypentanoic acid gamma-lactone (3,4,5-TP) (2.50 mumol) decreased food intake for, at most, 24 h. These acids depressed the size of the first meal after infusion, but did not affect latency to the first meal, eating speed, drinking or ambulation. Infusion of 2,4-dihydroxybutanoic acid gamma-lactone (2,4-DB) (1.25 mumol), 2,4,5-trihydroxypentanoic acid gamma-lactone (2,4,5-TP), and an exogenous compound, 2,4,5,6-tetrahydroxyhexanoic acid gamma-lactone (2,4,5,6-TH) (2.50 mumol), induced transient initial feeding which was not necessarily accompanied by periprandial drinking. Ambulation was concomitantly increased. Of these organic acids, 3,4-DB and 2,4,5-TP were most potent in their effects on feeding. Hyperglycemia was induced by 2.50 mumol 3,4-DB leaving insulin unaffected; 2.50 mumol 2,4,5-TP caused hypoglycemia, with a persistent but not significant rise in insulin. The results suggest that slight structural differences of endogenous organic acids, in particular the positions of hydroxyl groups on the lactone ring of 4-butanolide, may be important in feeding modulation by conveying intrinsically reciprocal signals to neurons involved in feeding and satiety.

Insulin in the arcuate nucleus of the hypothalamus reduces fat consumption in rats.

Data are accumulating that insulin acting in the central nervous system is a physiological regulator of food intake and body weight, presumably via its effect in the hypothalamus. The present study investigated whether infusion of a small dose of insulin into two major hypothalamic insulin-binding areas also has an effect on diet selection and behavior. At the beginning of the dark period, rats received local bilateral infusions of 4 microU of insulin or vehicle during 34 min into the arcuate (ARC) or paraventricular (PVN) nucleus of the hypothalamus. Consumption of carbohydrate (C)-, protein (P)-, and fat (F)-enriched food and time spent on certain behaviors (drinking, resting, grooming, rearing, exploring/sniffing) were assessed during the first nocturnal hour. In addition, 21-h diet selection was assessed. The percentage contribution of macronutrients (C/P/F) to total energy content of the C-, P-, and F-enriched diets was 71.9/17.2/10.9, 45.8/43.4/10.8, and 47.1/17.5/35.4, respectively. During the first hour, infusion of insulin into the PVN increased grooming behavior compared to infusion of the vehicle. Although infusion of insulin had no effect on diet selection during the first hour, insulin infused in the ARC caused a reduction in F-enriched food consumption and total intake of F (as a macronutrient) over the 21-h period without altering total food intake. Infusion of a higher dose of insulin (10 microU) into the third ventricle had no effect on any of the assessed parameters. The data are explained to indicate that insulin (being an indicator of a positive energy balance) adjusts body weight homeostasis by modulating the preference for fat, at least at the level of the ARC, but not at the PVN.

Involvement of nitric oxide in neuroglycopenia-induced insulin and glucagon secretion in the mouse.

Neuroglycopenia induced by administration of 2-deoxy-D-glucose is known to stimulate the secretion of both insulin and glucagon in mice by a mechanism that is dependent on neural activity. In the present study, we examined whether the neurotransmitter nitric oxide (NO) is involved in this process. Therefore, 2-deoxy-D-glucose (500 mg/kg) was injected intravenously alone or together with the inhibitor of NO synthase, NG-nitro-L-arginine methyl ester (50 mg/kg) to conscious mice. It was found that NG-nitro-L-arginine methyl ester inhibited the increased plasma levels of both insulin (by 26%; P = 0.039) and glucagon (by 45%; P < 0.001) at 10 min after injection of 2-deoxy-D-glucose. Similarly, the NO synthase inhibitor, NG-nitro-L-arginine, which is devoid of the anticholinergic property of NG-nitro-L-arginine methyl ester, inhibited the responses of both insulin (by 53%; P = 0.026) and glucagon (by 57%; P = 0.003) to 2-deoxy-D-glucose. In contrast, the stereoisomer of NG-nitro-L-arginine methyl ester, NG-nitro-D-arginine methyl ester, which is devoid of NO synthase inhibitory activity, was without effect on 2-deoxy-D-glucose-induced insulin and glucagon secretion. Plasma levels of adrenaline and noradrenaline after administration of 2-deoxy-D-glucose were also reduced by NG-nitro-L-arginine methyl ester. In contrast, the insulin and glucagon secretory responses to intravenous injection of arginine (250 mg/kg), glucose (500 mg/kg) or the cholinergic agonist, carbachol (30 micrograms/kg), were not influenced by NG-nitro-L-arginine methyl ester, NG-nitro-D-arginine methyl ester or NG-nitro-L-arginine.(ABSTRACT TRUNCATED AT 250 WORDS)

Effects of long-term d-fenfluramine treatment on energy metabolism in rats.

The effects of chronic intragastric administration of the anorectic agent d-fenfluramine on energy metabolism and nutrient concentrations were investigated at rest and during swimming. Rats were provided with permanent cannulas for blood sampling and intragastric administration of d-fenfluramine or saline. Energy expenditure and nutrient and hormone concentrations were determined. Under baseline conditions, d-fenfluramine increased carbohydrate utilization (14.2 vs. 7.0 mg/kg.min) and decreased fat oxidation (2.8 vs. 5.5 mg/kg.min). Plasma free fatty acid concentration was decreased (0.29 vs. 0.55 mmol/l) and lactate and insulin concentrations were increased after d-fenfluramine treatment (0.64 vs. 0.37 mmol/l and 61 vs. 33 mU/l, respectively). The shift in nutrient utilization also occurred during swimming. The exercise-induced increase in blood glucose was reduced after d-fenfluramine (+0.8 vs. +2.0 mmol/l). During swimming, free fatty acid, lactate and insulin concentrations were similar in the two groups. It is hypothesized that chronic d-fenfluramine treatment increases in the oxidation of carbohydrates and decreases the oxidation of fat as a result of a decrease in the transport of fatty acids over the mitochondrial membrane.

Neuroendocrine factors regulating blood glucose, plasma FFA and insulin in the development of obesity.

A number of neurotransmitters and neuropeptides in the hypothalamus play a role in the control of food intake, metabolism, and body weight. Particularly, noradrenergic mechanisms in several areas of the hypothalamus are involved. Control of peripheral metabolism by the hypothalamus is achieved via autonomic modulation of the function of hepatocytes, adipocytes, and the endocrine cells in the islets of Langerhans. The autonomic control mechanisms ultimately lead to an appropriate shaping of blood glucose, plasma FFA, and insulin profiles to guarantee an adequate flow of nutrients under different physiological situations. Peripheral insulin and glucose can penetrate into the brain where they might affect the function of those brain structures involved in control of food intake, metabolism, and body weight.

Blood glucose levels in portal and peripheral circulation and their relation to food intake in the rat.

Rats weighing about 450 g were provided with permanent catheters in the portal vein and the right auricle. This method allows blood sampling from the portal and peripheral circulation at the same moment in the nondisturbed unanesthetized rat. In the ad lib condition the portal glucose level was higher than that in the general circulation before, during, and after the meal. After a fast of 22 hr premeal portal vein levels were equal to those of the general circulation. During the meal the portal glucose levels rose to about 150 mg per 100 ml whereas those of the general circulation did not exceed 130 mg/100 ml. Experiments with glucose infusions systemically and intraportally show that, under conditions of mild deprivation, the level of glucose in the portal vein plays no or only a very minor role in the termination of feeding.