Body weight: reduction by long-term glycerol treatment

The Rise and Fall of Physiological Theories of the Control of Human Eating Behavior

Kuhns was the first to suggest that theories in science do not develop in small increments but rather in major leaps to paradigms that examine the same question through very different perspectives. Theories on the mechanism responsible for control of human food intake fall into Kuhn’s description. This article describes how the two major theories of the control of food intake in humans, the Glucostatic Theory, and the Lipostatic Theory, showed initial promise as explanations, but later deteriorated with the slow accumulation experimental data. The locus of theories considered eating behavior as a part of physiological system that regulates the storage of energy on the body. We challenge this fundamental belief with data which suggests that we must be ready to accept a major change in the way we think about eating behavior if we are ever to decrease the prevalence of obesity.

Glycerol is synthesized and secreted by adipocytes to dispose of excess glucose, via glycerogenesis and increased acyl-glycerol turnover

White adipose tissue (WAT) produces large amounts of lactate and glycerol from glucose. We used mature epididymal adipocytes to analyse the relative importance of glycolytic versus lipogenic glycerol in adipocytes devoid of external stimuli. Cells were incubated (24/48 h) with 7/14 mM glucose; half of the wells contained ¹⁴C-glucose. We analysed glucose label fate, medium metabolites, and the expression of key genes coding for proteins controlling glycerol metabolism. The effects of initial glucose levels were small, but time of incubation increased cell activity and modified its metabolic focus. The massive efflux of lactate was uniform with time and unrelated to glucose concentration; however, glycerol-3P synthesis was higher in the second day of incubation, being largely incorporated into the glycerides-glycerol fraction. Glycerophosphatase expression was not affected by incubation. The stimulation of glycerogenic enzymes’ expression was mirrored in lipases. The result was a shift from medium glycolytic to lipolytic glycerol released as a consequence of increased triacylglycerol turnover, in which most fatty acids were recycled. Production of glycerol seems to be an important primary function of adipocytes, maintained both by glycerogenesis and acyl-glycerol turnover. Production of 3C fragments may also contribute to convert excess glucose into smaller, more readily usable, 3C metabolites.

Metabolic effects of glycerol supplementation and aerobic physical training on Wistar rats

We evaluated the effects of oral glycerol supplementation on trained rats fed a normal diet. Wistar rats were distributed among 6 groups in a completely randomized 2 × 3 factorial design. The animals were subjected to 6 weeks of aerobic training. In the last 4 weeks, the animals’ diet was supplemented with saline, glucose, or glycerol. Data were subjected to one-way analysis of variance (ANOVA) followed by a Student–Newmann–Keuls test, with values for P < 0.05 considered statistically significant. The change in body mass was lower in the trained groups, and their food and water consumption were higher. Glycerol supplementation resulted in an increase in the levels of triacylglycerol (TAG) and total cholesterol, as well as in the area and diameter of adipocytes. When associated with training, these parameters were similar to those of other trained groups. Levels of low-density lipoprotein + very-low-density lipoprotein cholesterol decreased in the trained animals that received glycerol compared with the non-trained ones. Glycerol consumption caused a reduction in food intake and increased the villous:crypt (V:C) ratio. No changes in glycemia, high density lipoproteins, or density of adipocytes were observed. Supplementation with glycerol together with aerobic physical training promoted beneficial metabolic effects. However, in non-trained rats glycerol increased the diameter and area of adipocytes, as well as the levels of TAG and total cholesterol.

Sensory, digestive and metabolic influences on preference and intake

In the 1960s. scientific work on the mechanisms of appetite began to shift from the effects of deprivation on reinforcement, motivation and sensation (hunger and thirst) to the influences of ingestion itself on what and how much was chosen at a bout of eating and drinking (the sating of appetite). This personal overview starts with John D. Davis's involvement in this trend to studying sensory, digestive and metabolic controls of meal volume, as he and the author each joined the field. The main topic is the diverse series of innovative experiments that Davis carried out in the author's research group in 1970 1971. This work exemplified Davis's technical inventiveness and theoretical precision. It generated influential findings on behavioural roles of dietary flavours, gastric distension and hepatic metabolism.

Pyruvate and hydroxycitrate/carnitine may synergize to promote reverse electron transport in hepatocyte mitochondria, effectively 'uncoupling' the oxidation of fatty acids

In a recent pilot study, joint administration of pyruvate, hydroxycitrate (HCA), and carnitine to obese subjects was associated with a remarkable rate of body-fat loss and thermogenesis, strongly suggestive of uncoupled fatty-acid oxidation. Hepatocytes possess an uncoupling mechanism--reverse electron transport--that enables fasting ketogenesis to proceed independent of respiratory control. Electrons entering the respiratory chain at the coenzyme Q (CoQ) level via FAD-dependent acyl coA dehydrogenase, can be driven 'up' the chain by the electrochemical proton gradient to reduce NAD+; if these electrons are then shuttled to the cytoplasm, returning to the respiratory chain at the CoQ level, the net result is heat generation at the expense of the proton gradient, enabling the uncoupled flow of electrons to oxygen. Pyruvate's bariatric utility may stem from its ability to catalyze the rapid transport of high-energy electrons from mitochondria to the cytoplasm, thus stimulating electron shuttle mechanisms. By enabling rapid mitochondrial uptake of fatty acids and thus disinhibiting hepatocyte ketogenesis, HCA/carnitine should initiate reverse electron transport: concurrent amplification of electron shuttle mechanisms by pyruvate can be expected to accelerate this reverse electron transport, thereby decreasing the electrochemical proton gradient. As a result, hepatocytes may be able to convert fatty acids to CO2 and heat with little net generation of ATP. These considerations suggest that it may be feasible to render hepatocytes functionally equivalent to activated brown fat, such that stored fat can be selectively oxidized in the absence of caloric restriction. Other measures which enhance the efficiency of hepatocyte electron shuttle mechanisms may increase the efficacy of this strategy.

Short-term treatment with oleoyl-oestrone in liposomes (Merlin-2) strongly reduces the expression of the ob gene in young rats

Young female rats of 160-180 g were implanted with osmotic minipumps releasing 3.0 micromol/day per kg of oleoyl-oestrone in liposomes (Merlin-2) into the bloodstream for up to 14 days. Merlin-2 induced a loss of appetite in the first days, later recovered, and a decrease in body weight of 7%, which contrasts with the 15% increase in controls during the 2-week period. Neither plasma glucose nor urea was affected by treatment, but liver glycogen increased by 50% in 14 days. Insulin decreased slightly with Merlin-2 treatment. Plasma corticotropin (ACTH) and corticosterone showed a transient increase by day 6 of treatment. The expression of the ob gene in adipose tissue fell during the period studied to practically nil on day 14; circulating leptin levels decreased more than 70% from day 1 to day 14. Oestrone levels increased from 0.3 nM (controls) to a maintained 40-60 nM level for the rest of the experiment. Oleoyl-oestrone levels first increased 4-fold, to decrease again to the initial levels on day 10, increasing later to 100-fold on day 14. The three phases observed in food intake, weight loss and oleoyl-oestrone levels match fairly well, which supports the direct involvement of oleoyl-oestrone in body-weight control. However, the control of oleoyl-oestrone levels seems to be mediated in part by corticosterone. The practical disappearance of leptin synthesis coincides with the massive accumulation of oleoyl-oestrone in plasma. The results presented suggest the involvement of oleoyl-oestrone in the main mechanisms of control of body weight and its regulation by glucocorticoids and leptin.

Normative hematologic and serum biochemical values for adult and infant rhesus monkeys (Macaca mulatta) in a controlled laboratory environment

During the conduct of a long-term toxicity/reproduction study using rhesus monkeys (Macaca mulatta), periodic hematologic and serum biochemical analyses were undertaken on blood obtained from 15 untreated (vehicle control) adult female monkeys over a period of 3 yr, 20 untreated adult males over a period of 6 yr, and 9 infants (5 male, 4 female, whose dams were the vehicle control females) from 12 wk through 2 yr of age. All samples were obtained by femoral puncture. To facilitate handling, ketamine hydrochloride (Ket.HCl) was administered to the males, and for some of the later infant samplings. Complete blood cell counts were performed using electronic cell counters, while reticulocyte counts and leukocyte differentials were assessed manually. The serum biochemistry analyses were performed with automated analyzers. These data were obtained as part of a health monitoring program, with medians and 2.5 to 97.5 percentiles for each parameter indicated. Some monitored parameters were found to be affected by the type of equipment used and by the use of Ket.HCl.

Experimental obesity: a homeostatic failure due to defective nutrient stimulation of the sympathetic nervous system

The basic hypothesis of this review is that studies on models of experimental obesity can provide insight into the control systems regulating body nutrient stores in humans. In this homeostatic or feedback approach to analysis of the nutrient control system, we have examined the afferent feedback signals, the central controller, and the efferent control elements regulating the controlled system of nutrient intake, storage, and oxidation. The mechanisms involved in the beginning and ending of single meals must clearly be related to the long-term changes in fat stores, although this relationship is far from clear. Changes in total nutrient storage in adipose tissue can arise as a consequence of changes in the quantity of nutrients ingested in one form or another or a decrease in the utilization of the ingested nutrients. A change in energy intake can be effected by increased size of individual meals, increased number of meals in a 24-hour period, or a combination of these events. Similarly, a decrease in utilization of these nutrients can develop through changes in resting metabolic energy expenditure which are associated with one of more of the biological cycles such as protein metabolism, triglyceride for glycogen synthesis and breakdown, or maintenance of ionic gradients for Na+ + K+ across cell walls. In addition, differences in energy expenditure related to the thermogenesis of eating or to the level of physical activity may account for differences in nutrient utilization.

Effect of intrahypothalamic hydroxybutyrate on sympathetic firing rate

Following the microinjection of 3-hydroxybutyrate into either the paraventricular or ventomedial nucleus of the hypothalamus, there was a significant increase in firing rate of the sympathetic nerves to brown adipose tissue, with a peak increase in activity occurring three minutes following the injection. The response in both hypothalamic regions was dose-dependent. When injected into both regions at the same time, the increase in firing rate was additive. These studies suggest that ketone bodies in the central nervous system may activate the sympathetic nervous system.

The role of glucose, insulin and glucagon in the regulation of food intake and body weight

Glucose and related pancreatic hormones play a major role in the metabolism of monogastric mammals yet their influence on hunger and/or satiety is, as yet, poorly understood. Glucose, insulin and glucagon rise during a meal and gradually decline to baseline levels shortly after a meal. A sudden drop in plasma glucose as well as insulin have been reported just prior to the onset of a meal but the functional significance of this is not yet clear. Systemic injections of glucose have no acute satiety effects but intraduodenal and intrahepatic infusions reduce food intake and free-feeding and deprived animals respectively. Treatments which decrease cellular glucose utilization directly (2-DG) or indirectly (insulin) increase food intake while exogenous glucagon (which produces hyperglycemia) decreases it. There is considerable evidence that some or all of these effects may be due to a direct central action of glucose, 2-DG, insulin, and glucagon on brain mechanisms concerned with the regulation of hunger and satiety although influences on peripheral "glucoreceptors" have been demonstrated as well. The functional significance of glucoprivic feeding is, however, questioned. The feeding response to 2-DG and related compounds is capricious, and its temporal course does not parallel the hyperglycemic reaction which presumably reflects cellular glucopenia. Moreover, numerous brain lesions which increase, decrease, or have no effect on ad lib intake and often have no effect on the response to deprivation have been shown to severely impair or abolish feeding responses to systemic injections of 2-DG that produce severe central as well as peripheral glucopenia. Feeding responses to insulin are intact after most of these lesions, suggesting that this hormone may influence food intake in a fundamentally different fashion. The mechanism of insulin action is not understood--the classic feeding response is obtained only with doses that are pharmacological when compared to normal plasma levels and there is increasing evidence that lower doses may have opposite, inhibitory effects on food intake and body weight. Relatively small doses of glucagon decrease food intake (although opposite facilitatory effects have been reported after even smaller doses) but the effect does not appear to be due to hepatic mobilization of glucose as initially assumed. Decreases in food intake after intracranial injections of very small doses suggest a direct central action.

Effect of palatable diet on growth, caloric intake and endocrine-metabolic profile in weanling rats with dorsomedial hypothalamic lesions

Weanling male Sprague-Dawley rats received bilateral electrolytic lesions in the dorsomedial hypothalamic nuclei (DMNL rats); sham-operated rats served as controls. All animals were fed lab chow for 15 postoperative days. At that time they were subdivided into two groups each. One DMNL and one control group continued to be fed lab chow until the termination of the experiment on postoperative day 116. A second DMNL and control group were fed a high-fat diet and 32% sucrose solution (HF/SS diet). All DMNL rats showed reduced body weight and linear growth, but the HF/SS diet depressed these parameters further below the levels of the chow-fed groups. Both DMNL and control rats fed HF/SS had more carcass fat, heavier epididymal fat pads, more carcass fat per calories eaten, higher plasma levels of glucose, glycerol and free fatty acids but lower insulin levels than chow-fed DMNL rats and controls. This occurred in the face of lower body weights and caloric intake. Neither growth hormone nor insulin showed lesion effects. Rats with DMNL exhibited the same inverse relationship between plasma insulin and free fatty acids as controls. The data indicate that DMNL rats respond to the HF/SS diet essentially like sham-operated controls, i.e., they develop dietary obesity. Although they do show some small deficits, their lipogenic capacity is actually significantly greater than that of HF/SS-diet fed controls.

Food intake in the domestic fowl: effect of intrahepatic lipid and amino acid infusions

Five experiments were conducted to determine the effect of hepatic and jugular infusions of amino acid and lipid solutions on short-term food intake of cockerels. Neither jugular nor hepatic infusions of amino acids had any effect on food intake of Single-Comb White Leghorn (SCWL) cockerels. Lipid infusions had no effect on SCWL cockerels when administered intrajugularly but decreased food intake significantly when infused intrahepatically. In contrast, no effect on food intake was observed when Rock-Cornish (RC) cockerels were infused intrahepatically with lipid. Thus, it appears that the liver of the SCWL cockerel is not responsive to plasma amino acid concentrations but is sensitive to circulating lipid levels. This lipid sensitive mechanism may be important in the peripheral control of food intake in a chicken such as the SCWL which has been bred for egg production but may be desensitized in fowl such as the RC broiler which has been selected for large body size and feed consumption.

Endogenous sugar acid derivative acting as a feeding suppressant

Evidence suggests that endogenous sugar acids 3,4-dihydroxybutanoic acid (2-deoxytetronic acid, 2-DTA) and 2,4,5-trihydroxypentanoic acid (3-deoxypentonic acid, 3-DPA) may participate in the regulation of feeding. To study the effect of 2-buten-4-olide, a 2-DTA synthetic derivative, on food intake, male Wistar rats were subjected to various applications. Intraperitoneal administration of 2-buten-4-olide in doses of 30 to 100 mg/kg, decreased food intake dose-dependently by reducing meal frequency, meal size and eating rate, and prolonging meal duration, latency to eat the first meal after injection and post-prandial intermeal intervals. Drinking patterns and locomotor activity were not significantly affected. Administration of 2-buten-4-olide intragastrically in doses of 50 to 300 mg/kg, and intra-third cerebroventricularly in doses of 1.2 to 5.0 mumol/rat, dose-dependently reduced food intake. This and previous evidence suggest that: 2-DTA and its derivatives that share its bioactive components suppress food intake in the rat; They might represent a new category of potential therapeutic agents for hyperphagia and obesity.

Regulation of food intake by hepatic oxidative metabolism

It has been proposed that the turnover of metabolic fuels affects feeding. In the present study the effects of different metabolites (glycerol, L-malate, D-3-hydroxybutyrate, L-lactate) or their oxidation products (dihydroxyacetone, oxaloacetate, acetoacetate, pyruvate) on food intake were investigated under various feeding conditions. The results of these experiments increase our knowledge about the nature and origin of metabolic signals in the energostatic control of food intake. In the experiments, glycerol, malate, 3-hydroxybutyrate, lactate or pyruvate (4.5-7.7 mmoles/kg metabolic body weight injected subcutaneously reduced food intake in rats, whereas dihydroxyacetone, acetoacetate or oxaloacetate did not. The effect of glycerol disappeared with high levels of dietary protein, whereas the effects of lactate and pyruvate disappeared with high levels of dietary fat. The hypophagic effects of glycerol, malate, 3-hydroxybutyrate, lactate and pyruvate were abolished by selective hepatic vagotomy. The results suggest that vagally mediated signals originating from hepatic oxidation of metabolic fuels contribute to the regulation of food intake.

Selective hepatic vagotomy eliminates the hypophagic effect of different metabolites

To test the hypothesis that vagally mediated metabolic signals from the liver affect feeding, we investigated the influence of selective hepatic vagotomy on the anorectic effect of different metabolites in rats. Subcutaneous injection of glycerol (GLY), D-3 hydroxybutyrate (DHB), L-malate (MAL), L-lactate (LAC), or pyruvate (PYR) (5.7 mmol/kg metabolic body weight = b.wt.0.75, each) reduced food intake in sham-vagotomized rats. Selective hepatic vagotomy eliminated the hypophagic effect of all metabolites. In an additional experiment, LAC retained its potency to inhibit feeding in sham-vagotomized rats after an intraperitoneal injection of atropine methylnitrate (5 mg/kg b.wt.0.75). The results demonstrate that the hypophagic effect of GLY, DHB, MAL, LAC and PYR originates in the liver and is mediated by hepatic vagal afferents. This implicates the existence of adequate hepatic receptors which might contribute to the control of food intake.

Different metabolites might reduce food intake by the mitochondrial generation of reducing equivalents

To test the possibility that hypophagic effects of injected metabolites are linked to a particular oxidative step in their metabolism, cumulative food intake in rats after subcutaneous injection of various metabolites or their immediate oxidation products was investigated. Glycerol was compared with dihydroxyacetone, L-malate with oxaloacetate, and L-lactate with pyruvate. Subcutaneous injection of 4.5 mmol/kg metabolic body weight (BW0.75) of glycerol or malate reduced food intake significantly, whereas the same doses of dihydroxyacetone or oxaloacetate did not. Lactate or pyruvate (4.5 mmol/kg BW0.75, each) both reduced food intake significantly. However, even higher doses (7.7 mmol/kg BW0.75) of lactate or pyruvate failed to affect food intake when rats were fed a high fat (HF) diet, which is known to decrease the activity of the enzyme pyruvate dehydrogenase. In an additional experiment, subcutaneous injection of lactate or pyruvate (7.7 mmol/kg BW0.75, each) increased plasma levels of lactate and pyruvate 2 h after the injection in HF-rats more than in rats fed the usual high-carbohydrate diet. The data are consistent with the hypothesis that the increased generation of reducing equivalents in the mitochondria, brought about by the oxidation of injected glycerol, malate, lactate, or pyruvate, reduces food intake in rats.

Endogenous sugar acid control of hypothalamic neuron activity and gastric acid secretion in rats

The feeding related endogenous sugar acids, 2-deoxytetronic acid, 2-DTA and 3-deoxypentonic acid, 3-DPA, were investigated for their peripheral and central (hypothalamic) control of gastric acid secretion, and effects on activity of lateral hypothalamic (LHA) neurons in rats. Peripheral gastric acid secretion was not affected by either 2-DTA or 3-DPA. Slight gastric acid secretion was elicited by 3-DPA only when it was applied directly into the gastric related site of the LHA. Gastric acid secretion induced by 2-deoxy-D-glucose (2DG) was suppressed by application of 2-DTA in the LHA. 3-DPA had no effect on 2DG induced secretion. Electrophoretic application of 2-DTA significantly inhibited the activity of both gastric and non-gastric type glucose-sensitive neurons in the LHA, and 3-DPA significantly excited both types of glucose-sensitive neurons. The results agree with previous reports that 2-DTA and 3-DPA are endogenous satiety and hunger factors, respectively, and act by modulating hypothalamic control of gastric acid secretion which is mediated through gastric type and non-gastric type glucose-sensitive neurons.

Feeding suppression induced by a fecal anorexigenic substance (FS-T)

Intraperitoneal (IP) injection of a fecal anorexigenic substance (FS-T) induced significant suppression of feeding and this suppression recovered on the second day. At 2 hr after IP injection, at the time of maximum feeding suppression, plasma glucose, insulin and free fatty acid (FFA) levels did not change but amino acid level decreased. Intra-third cerebral ventricle (ICV) infusion of FS-T induced parallel but more potent feeding suppression. Analysis of meal patterns demonstrated that suppression of feeding after ICV treatment continued into the second day. FS-T was applied electrophoretically to glucose-sensitive and non glucose-sensitive neurons in the lateral hypothalamic area (LHA) and to glucoreceptor and non glucoreceptor neurons in the ventromedial hypothalamic nucleus (VMH). It significantly inhibited glucose-sensitive neurons but not non glucose-sensitive neurons, and excited both neuron types in the VMH. FS-T might thus work directly through the hypothalamic feeding control centers to suppress feeding. Even after pronase treatment of FS-T, a non-dialysable fraction of large molecular weight, consisting of protein and carbohydrate, maintained the original anorexigenic activity.

Effects of endogenous sugar acids on the ventromedial hypothalamic nucleus of the rat

Effects of certain endogenous sugar acids such as 2-deoxytetronic acid (2-DTA) and 3-deoxypentonic acid (3-DPA) were investigated on the neuronal activity of the ventromedial hypothalamic nucleus (VMH) in the rat. Electrophoretically applied 2-DTA significantly and specifically facilitated the activity of glucoreceptor neurons in the VMH, while 3-DPA suppressed their activity. However, non-glucoreceptor neurons were not affected by these sugar acids. These results reinforces the previous suggestion that 2-DTA and 3-DPA also might act as endogenous satiety and hunger substances, respectively.

[Satiation mechanism]

Food intake in man and animals occurs as meals. Because changes in daily food intake in laboratory animals are often associated with corresponding changes in meal size, while meal frequency is less affected, food intake appears to be regulated mainly by the satiety mechanism. Meal size is controlled by feedback or satiety signals originating from the stomach, the small intestine, the liver and the fat depots. Both neural and hormonal (Bombesin?, Cholecystokinin?) feedback signals are elicited in the stomach and intestine. The feedback signals of the liver apparently originate from hepatic chemoreceptors, whose activity is modulated by certain metabolites, in particular glucose. The satiety effect of pancreatic glucagon, whose plasma concentration in the portal vein increases during a meal, also appears to be mediated by the hepatic chemoreceptors. Unmyelated nerve endings are probably the morphological substrate of the hepatic chemoreceptors. There is some evidence that circulating fat metabolites (fatty acids, ketone bodies, glycerol), whose plasma levels are related to the size of the fat depots, function as lipostatic feedback signals. The signal transfer from the gastrointestinal chemo- and mechanoreceptors and from the hepatic chemoreceptors to the central nervous system occurs mainly through vagal afferents. The first relay for this afferents is the nucleus tractus solitarii in the medulla oblongata. The next relay is at the pons, from where neurons project to the hypothalamus and other forebrain areas. The hypothalamus has important integrative functions in the control of food intake.

Short-chain polyhydroxymonocarboxylic acids as physiological signals for food intake

In order to clarify the effects of endogenous organic acids on short and long-term feeding behavior, ingestive behavior was monitored for 2 hr before and after intra-third ventricular infusions of 3,4-dihydroxybutyric acid (2-deoxytetronic acid, 2-DTA), 2,4,5-trihydroxypentanoic acid (3-deoxypentonic acid, 3-DPA), and 3-hydroxybutyric acid (3-HBA). In addition, meal patterns were recorded for 2 days before and after the ventricular infusions. 2-DTA suppressed both short and long-term feeding by decreasing meal size (MS). 3-DPA elicited transient feeding behavior, but caused no change in long-term feeding. 3-HBA initially stimulated feeding, but subsequently suppressed long-term feeding by decreasing MS and prolonging postprandial intermeal interval (IMI). The suppressive effects of 3-HBA on feeding behavior lasted about 24 hr longer than those of 2-DTA. Based upon these observations as well as our previous reports, it appears that some of the processes affecting hunger and satiation are mediated by changes in central and peripheral concentrations of these organic acids.

The integration of visceral information in the control of feeding

Electrophysiological evidence exists for sensitivity to many aspects of nutrient depletion and repletion at all levels of the viscera and beyond. Behavioral evidence suggests that feeding is sensitive to manipulations at all of these levels, while at the same time showing that no one site or nutrient is required for feeding regulation. This suggests that one must look at the sequence and interdependence of events occurring from ingestion to ultimate tissue distribution in order to understand the control of food intake.

Early streptozotocin diabetes and hunger

The initial effect of streptozotocin diabetes is not hyperphagia, but reduced food intake. Diabetic hyperphagia reaches maximum only after many days. Utilization of body fat may account for the delayed appearance of diabetic hyperphagia; this effect may be mediated by plasma fat metabolites. Plasma levels of free fatty acids (FFAs), ketone bodies, and glycerol are greatly elevated following STZ treatment, but return nearly to normal as diabetic hyperphagia appears.

Increases in plasma glycerol levels precede the hypophagia following subcutaneous glycerol injection in rats

Cumulative food intake and plasma glycerol levels following subcutaneous glycerol injections were investigated in rats, because increases in plasma glycerol have been suspected to contribute to glycerol-induced hypophagia. Besides plasma glycerol, plasma non-esterified fatty acids (NEFA), plasma D-(-)-3-hydroxybutyrate (3-HB), blood glucose, and liver glycogen content were also measured. Two daily injections of 80 mg/kg body wt. glycerol did not affect food intake and failed to increase plasma glycerol 30 min after the injection. A single injection of 660 mg/kg body wt. glycerol reduced food intake and increased plasma glycerol 1-2 hours after the injection. This glycerol dose also increased liver glycogen content 1-2 hours after the injection but did not affect plasma NEFA and 3-HB. Rats injected with 660 mg/kg body wt. glycerol did not reduce feeding within the first 2 hours following the injection, when plasma glycerol levels were increased. Inhibition of feeding began at about 3 hours and continued up to 6 hours although plasma glycerol levels had declined to control values 5-6 hours after the injection. It is concluded that metabolic consequences of elevated plasma glycerol levels rather than increases in plasma glycerol levels per se elicit the food intake reduction following exogenous glycerol loads in rats.

Plasma fat metabolites and hunger

In order to test the hypothesis that the fat metabolites are the blood-borne signals which suppress hunger during recovery from reversible obesity, experiments were designed to manipulate plasma fat metabolite levels directly. In order to elevate plasma glycerol levels, glycerol was infused intravenously into relatively unrestrained rats for 36 hours; this treatment greatly increased plasma glycerol levels but reduced voluntary food intake only slightly. Similar results were obtained when glycerol was mixed with powdered rat food. These results suggest that glycerol is not the "lipostatic hormone" although it may contribute to regulation. Similar experiments with a synthetic precursor of the ketone bodies (1,3 butanediol), suggest that the ketone bodies contribute to the decrease in food intake after reversible obesity, but cannot be a complete explanation. Dietary fat consumption raised plasma free fatty acid (FFA) levels to the range seen during recovery from reversible obesity, suggesting that plasma FFAs may be a blood-borne signal of fat utilization in both cases. Intralipid, a synthetic triglyceride emulsion designed for intravenous administration, also increased plasma FFA levels but suppressed food intake by less than predicted. However, Intralipid may tend to cause spuriously high plasma FFA readings for reasons which are discussed. These results suggest that plasma fat metabolites, especially FFAs, may be blood-borne signals which contribute to the voluntary dieting after reversible obesity.

Plasma metabolites and food intake reduction following heparinoid injection in rats

Behavioral and physiological consequences of heparinoid injections, which increase plasma levels of nonesterified fatty acids (NEFA) by releasing lipoproteinlipase into the circulation, were studied in rats, because, according to the lipostatic hypothesis of food intake control, changes in plasma metabolite levels should influence food intake. Subcutaneously injected 55 mg/kg body wt. of the heparinoid Na-pentosanpolysulfate (PPS) significantly reduced feeding in rats with ad lib access to either a high carbohydrate (HC) or a high fat (HF) diet. However, the reduction of food intake in HF-rats exceeded that in HC-rats. Moreover, the food intake reduction following PPS-injections was associated with greater increases in plasma levels of NEFA and free glycerol in HF-rats, whereas PPS increased plasma levels of D-(-)-3-hydroxybutyrate only in HF-rats, and did not alter blood glucose concentrations. Although PPS inhibited blood coagulation, it did not affect the hematocrit. Furthermore, PPS-injections that reduced feeding, failed to produce a conditioned taste aversion in a two bottle preference test. The data therefore strongly suggest that elevated levels of plasma NEFA and/or free plasma glycerol decrease food intake in rats. These metabolites might contribute to a lipostatic mechanism of feeding control.

Inverse relationship between brown fat thermogenesis and meal size: the thermostatic control of food intake revisited

This study examines a new hypothesis whereby heat production from brown fat in response to eating may serve as a feedback signal for satiety. To test this hypothesis, in vitro respiration rate of brown adipose tissue (BAT) was determined in relation to the voluntary caloric intake of the preceding test meal. This relationship was examined as a function of meal composition and of obesity. It was found that in rats fed a high fat diet, as well as in two types of obese rats (VMH and Zucker), respiration rate per 100 mg tissue was significantly reduced, and energy intake of the preceding test meal increased compared to rats receiving a low fat diet or to respective lean rats. These data lend support to a brown fat mediated thermostatic hypothesis for the control of food intake.

Glycerol is not a physiologic signal in the control of food intake in rats

Rats were given two daily subcutaneous injections of either 40, 80 or 160 mg/kg of glycerol or control injections of glucose. Glycerol injections did not reliably affect body weight gain at any dose tested whether animals were maintained in constant light or a 12 hr light/dark cycle. Only injections of 160 mg/kg of glycerol given twice daily significantly reduced daily food intake; however, this dose also raised plasma glycerol concentrations briefly to an unphysiologic level. Continuous intravenous infusions of 390 or 780 mg/rat/day of glycerol had no effect on food intake, even though plasma glycerol levels increased markedly within the normal range. These findings indicate that reductions of food intake and body weight after administration of glycerol are probably responses to a treatment which is not physiologic.

Sustained intracerebroventricular infusion of brain fuels reduces body weight and food intake in rats

Long-term infusion of glucose, beta-hydroxybutyrate, and glycerol into the third ventricle of the rat brain caused a stabilization of body weight at a lower than normal level. Among the glucose- and glycerol-treated animals this weight loss was caused in part by temporary hypophagia. Among the animals treated with beta-hydroxybutyrate the weight loss was unaccompanied by a reduction in food intake. The results are consistent with the view that the systems controlling food intake and body weight are sensitive to the availability of brain fuels. They are not consistent however, with the view that these control systems monitor calories independently of their source.

Cellular basis of direct insulin action in the central nervous system

The in vivo radioautographic method has been applied to elucidate the mechanism of direct peptide hormone "feedback" action in the CNS. Using this method we have identified the circumventricular organs of the brain as general endocrine target tissues for a variety of blood-borne polypeptide hormones, including insulin. In the arcuatemedian eminence region of the hypothalamus blood-borne insulin directly interacts with receptive nerve terminals, suggesting that insulin acts to influence the electrical activity of select hypothalamic nerve circuits at the level of synaptic transmission. Recent results obtained from preliminary surgical and chemical lesion studies of brain indicate that insulin-receptive nerve terminals in the arcuate-median eminence region arise from neurons intrinsic to the medial basal hypothalamus. This has lead us to propose the concept of the hypothalamic tuberoinfundibular insulin-receptive neuron and its axon collaterals as a pathway for the centripetal flow of insulin "signals" in the form of electrical impulses. We envisige that the neuroanatomic pathway, provided by the hypothalamic tuberoinfundibular neuron, functions to link changes in body metabolic activity, as reflected in changing levels of circulating insulin, to the neuronal process of elaborating specific central metabolic-regulatory programs. This pathway could be of key importance in understanding and combating metabolic disease.

Effect of glycerol on weight loss and hunger in obese patients

The effectiveness of oral glycerol as a dietary component or as a supplement to a 1000-kcal/day diet was examined in two studies involving obese patients. Glycerol did not differ from an equicaloric dose of glucose in its effect on hunger ratings, diet compliance or overall weight loss. We conclude that oral glycerol is not a useful adjunct to weight reduction programs.

Food and water deprivation: changes in rat feeding, drinking, activity and body weight

Groups of male rats totally food deprived for periods up to 96 hr, water ad lib do not totally regain body weight lost through deprivation. Degree of body weight defence was estimated by fitting statistically a curve to the predeprivation baseline data and extrapolating the curve over the deprivation and postdeprivation period. From group data, body weight is not fully defended. However, marked individual differences in degree of body weight defence were noted. During food deprivation the daily drinking cycle persists for approximately 48 hr, then a reduction in nocturnal drinking appears. General activity increases but only in the light period. After food restitution, the main hyperphagia is confined to the first post-deprivation day. Additional, subtle long term effects of overeating in the light period are also found. In contrast to food deprived groups, water deprived groups accurately defend body weight. During water deprivation daily food intakes decrease immediately whereas nocturnal activity decreases after approximately 72 hr of deprivation has elapsed. After the termination of water deprivation the immediate response is gross hyperdipsia, but again subtle long term effects are apparent. Total daily food intake increases for many days but the increase is not confined to either light or dark periods. Water intake is similarly affected. These results are evaluated in relation to the notion of a body weight set-point regulating feeding behavior. Some limitations on the value of the findings in terms of species differences and lack of ecological validity are also discussed.