Sweet tooth reconsidered: taste responsiveness in human obesity

Taste responses of normal-weight, obese, and formerly obese individuals for sucrose and fat containing stimuli were examined using a mathematical modelling technique known as the Response Surface Method. The subjects accurately rated intensities of sweetness, fatness, and creaminess of 20 different mixtures of milk, cream, and sugar, and no mixture phenomena or inter-group differences were observed. In contrast, hedonic taste responses varied across subject groups, and were affected differentially by the sucrose and lipid content of the stimuli. Normal-weight subjects optimally preferred stimuli containing 20% lipid and less than 10% sucrose. Obese subjects preferred high-fat stimuli (greater than 34% lipid) that contained less than 5% sucrose, while formerly obese subjects showed enhanced responsiveness to both sugar and fat. Hedonic responsiveness as measured by the optimal sugar/fat ratio was negatively correlated with the degree of overweight (body mass index: weight/height). We hypothesize that sensory preferences for dietary sugars and fats aren determined by body-weight status and may affect the patterns of food consumption.     

Neural correlates of individual differences related to appetite

Using neuroimaging technologies to compare normal weight and obese individuals can reveal much about the pathophysiological state of obesity but such comparisons tell us little about what makes some normal weight individuals susceptible to obesity or about important individual differences amongst obese individuals. The current review therefore reviews neuroimaging research on individual difference measures that can illuminate these important topics. After introducing three neuropsychological models of the nature of motivation to approach rewarding stimuli, neuroimaging research on measures of impulsivity, craving, binge eating, restrained eating and disinhibited eating is reviewed. Although neuroimaging research on individual differences measures of brain activity related to appetite is in its infancy, existing studies suggest that such research could enrich the understanding, prevention and treatment of disordered eating and obesity.     

Sugar and fat: sensory and hedonic evaluation of liquid and solid foods

Twenty-five subjects evaluated the sweetness, creaminess and fat content of liquid and solid dairy products containing between 0.1 and 52 g fat/100 g and sweetened with 0-20% sucrose weight/weight. Liquid stimuli included skim milk, whole milk, half and half, and heavy cream, while the solids included cottage cheese and cream cheese, blended and spread "jelly-roll" fashion on slices of white bread. The subjects' ratings of stimulus sweetness, creaminess, and fat content differed sharply between liquids and solids, and the assessment of fat content of solid foods appeared to be impaired. In contrast, acceptability ratings for both sets of stimuli were not substantially different: the subjects optimally preferred equivalent levels of sugar in both liquids and solids, but selected higher fat levels in solid than in liquid foods. Sensory preferences for fat in liquid stimuli may not always be indicative of preferences for fat in solid foods.     

Taste and olfaction in human obesity

Earlier studies have shown differences between normal weight and obese humans in responsivity to external and internal stimuli. This study shows that normal weight and obese subjects do not differ in hedonic response to sucrose (taste) and benzaldehyde (odor). However, a perceptual typing of individuals based upon hedonic response is possible for both gustatory and olfactory processes. Ratings of pleasantness for the sweet taste of sucrose appear to generalize to the food-related odor of bitter almonds. The method of magnitude estimation as applied to the study of taste and olfaction in man may reveal relationship between changes in internal state and hedonic behavior.     

The effects of exercise on the neuronal response to food cues

Increased physical activity is associated with successful long-term weight loss maintenance due to mechanisms likely more complex than simply increased energy expenditure. The impact of physical activity on the central regulation of food intake may be an important mechanism of this effect. The objective of this study was to examine the effects of exercise training and acute exercise on the neuronal response to food cues as well as eating behaviors. fMRI was performed in the fasted state at baseline and again after a 6 month progressive exercise intervention (supervised, 5 days/wk) both with and without an acute exercise bout in 12 overweight/obese (5 women, 7 men; BMI 33 ± 4 kg/m(2)) healthy adults. fMRI data were acquired while subjects were presented with visual stimuli of foods of high hedonic value as compared to neutral control objects. Questionnaires on eating behaviors, ratings of appeal and desire for foods, and ratings of appetite (hunger, satiety, prospective intake) using visual analog scales were also performed at baseline and again after the 6-month exercise intervention. While only a trend was observed for a reduction in body weight (102 ± 5 to 99 ± 6 kg, p=0.09), a significant reduction in fat mass was observed (36.4 ± 2.8 to 33.7 ± 3.2 kg, p=0.04), although as expected changes in fat mass were variable (-10.0 to +3.7 kg). Chronic exercise was associated with a reduction in the neuronal response to food, primarily in the posterior attention network and insula. A significant positive correlation between the change in fat/body mass and the change in insula response to food cues with chronic exercise was observed. An acute exercise bout attenuated the effects of chronic exercise. The exercise intervention, however, did not impact any of the measures of appetitive behavior. In summary, despite no effects on behavioral measures of appetite, chronic exercise training was associated with attenuation in the response to visual food cues in brain regions known to be important in food intake regulation. The insula, in particular, appears to play an important role in the potential exercise-induced weight loss and weight loss maintenance.     

Frontal-parietal responses to “oddball” stimuli depicting “fattened” faces are increased in successful dieters: An electroencephalographic study

Here we tested the hypothesis that compared with normal weight non dieting (control) subjects, normal weight successful dieters submitted to a rigorous and continuous monitoring of body weight (i.e. karate athletes) are characterized by an increase of cortical responses to oddball visual stimuli depicting the enlargement of faces or foods, as neural underpinning of attention processes related to the control of weight and eating. Electroencephalographic (EEG) data were recorded in 18 successful dieters (5 females) and 24 non dieting subjects (9 females). The subjects were given frequent (70%) and rare (30%) stimuli depicting faces (FACE), food (FOOD), and landscapes (CONTROL). The task was to click the mouse after the rare stimuli. The rare stimuli depicted the frequent stimuli graphically dilated by 25% along the horizontal axis. Cortical responses accompanying attention processes were probed by the difference between positive event-related potentials peaking around 400-500 ms post-stimulus for the rare minus frequent stimuli (P300). The popular freeware LORETA estimated P300 cortical sources. The results showed that in the FACE condition, the amplitude of left frontal (BA 6) and medial parietal (BA 5) P300 sources was higher in the successful dieters (karate athletes) than non dieting subjects. These results disclose that frontal-parietal responses to “oddball” stimuli depicting enlarged faces (i.e. representing face fattening) are enhanced in successful dieters (karate athletes). Future studies should evaluate this effect in other populations of successful dieters (i.e. boxers, top models etc.).     

Psychophysiological Predictors of Attentional Dysfunction in Children with Congenital Heart Defects

Cardiac responses to non-signal stimuli and to signal stimuli in a vigilance task were examined in children born with congenital heart defects (CHD), and in normal and attention deficit disordered (ADD) subjects. Overall task performance was lower in subjects with heart defects and in the ADD group. Cardiac measures revealed that normal children displayed significantly larger heart rate deceleration to the target stimuli than did either of the clinical groups. Moreover, although no group differences were observed in the cardiac response to non-signal auditory stimuli, exaggerated heart rate deceleration was observed to vibrotactile stimuli in both the clinical groups. Regression analyses revealed that the magnitude of the cardiac response to somatosensory stimuli was predictive of task performance (both within and between subject groups), with larger responses associated with higher error rates and lower perceptual sensitivity. Results were suggestive of a predictive relationship between somatosensory reactivity and neuropsychological maturation.     

The effects of exercise on the neuronal response to food cues

Increased physical activity is associated with successful long-term weight loss maintenance due to mechanisms likely more complex than simply increased energy expenditure. The impact of physical activity on the central regulation of food intake may be an important mechanism of this effect. The objective of this study was to examine the effects of exercise training and acute exercise on the neuronal response to food cues as well as eating behaviors. fMRI was performed in the fasted state at baseline and again after a 6 month progressive exercise intervention (supervised, 5 days/wk) both with and without an acute exercise bout in 12 overweight/obese (5 women, 7 men; BMI 33 ± 4 kg/m²) healthy adults. fMRI data were acquired while subjects were presented with visual stimuli of foods of high hedonic value as compared to neutral control objects. Questionnaires on eating behaviors, ratings of appeal and desire for foods, and ratings of appetite (hunger, satiety, prospective intake) using visual analog scales were also performed at baseline and again after the 6-month exercise intervention. While only a trend was observed for a reduction in body weight (102 ± 5 to 99 ± 6 kg, p = 0.09), a significant reduction in fat mass was observed (36.4 ± 2.8 to 33.7 ± 3.2 kg, p = 0.04), although as expected changes in fat mass were variable (− 10.0 to + 3.7 kg). Chronic exercise was associated with a reduction in the neuronal response to food, primarily in the posterior attention network and insula. A significant positive correlation between the change in fat/body mass and the change in insula response to food cues with chronic exercise was observed. An acute exercise bout attenuated the effects of chronic exercise. The exercise intervention, however, did not impact any of the measures of appetitive behavior. In summary, despite no effects on behavioral measures of appetite, chronic exercise training was associated with attenuation in the response to visual food cues in brain regions known to be important in food intake regulation. The insula, in particular, appears to play an important role in the potential exercise-induced weight loss and weight loss maintenance.     

Investigation of 3111T/C polymorphism of the CLOCK gene in obese individuals with or without binge eating disorder: association with higher body mass index

Loss of circadian patterning of metabolism-related functions seems to play a role in the pathogenesis of obesity; therefore, it is reasonable to hypothesize that the functional 3111T/C single nucleotide polymorphism (SNP) of the (Circadian locomotor output cycles kaput) CLOCK gene may have a part in the genetic susceptibility to obesity. The aim of this study was to assess the frequencies of 3111T/C CLOCK gene SNP in overweight/obese subjects with or without binge eating disorder (BED) as compared to normal weight healthy controls. A total of 284 Caucasian subjects, including 92 normal weight healthy subjects and 192 overweight/obese patients (107 with BED) participated into the study. Genotype and allele frequencies did not significantly differ between normal weight controls and overweight/obese patients with and/or without BED. However, overweight/obese patients carrying the CC genotype had significantly higher values of body mass index (BMI) as compared to those carrying the CT and/or TT genotypes. Moreover, obese class III individuals had a significantly higher frequency of both the CC genotype and the C allele as compared to individuals with BMI<40 kg/m(2). Present findings show for the first time that the 3111T/C SNP of the CLOCK gene is not associated to human obesity and/or BED, but it seems to predispose obese individuals to a higher BMI.     

Body composition prediction equations based on bioelectrical impedance and anthropometric variables for Japanese obese women

The tetrapolar bioelectrical impedance (BI) technique has been proposed as a convenient, valid approach to estimate body composition in normal, healthy adults. However, the validity of the BI technique for obese individuals is undetermined. The purpose of this study was to determine the validity of body composition estimated from the BI method when compared to underwater weighing with correction for residual volume, and to develop a convenient, useful equation that would sufficiently predict body composition of Japanese obese women. The order of assessing body composition was randomized for each subject. The subjects were 75 moderately or mildly obese women with a body fat content of 25.1–60.1%, aged 18–68 years. Among 12 independent variables which were arbitrarily selected as possible predictors of hydrodensitometrically determined body density (Db), BM Z/S² was the best predictor (r = 0.803, n = 53), where BM = body mass (kg), Z = impedance (ohms), and S = stature (cm). When chronological age was added in a multiple regression analysis, Db was derived with higher accuracy (r = 0.878) with an SEE of 0.0064 g/cm³. The linear multiple regression equation constructed for predicting the Db was, for instance, Db = 1.1171 – 0.0535BM Z/S² – 0.0003 Age – 0.0005T, where T = triceps skinfold thickness (mm). Three equations developed in the present study were then cross-validated in a different sample (n = 22) of obese women. The results indicated that the above equation had the lowest SEE (0.0070 g/cm³). We suggest that body composition of Japanese obese women can be predicted with relatively high accuracy by a combination of BM Z/S² with age and the triceps skinfold thickness. © 1992 Wiley-Liss, Inc.     

A naturalistic study of social influences on meal size among moderately obese and nonobese subjects.

The present study examined the amount of food chosen by moderately obese and nonobese customers in a university cafeteria as a function of whether they were about to eat alone or were explicitly accompanied by others during lunch. It was hypothesized that many overweight individuals are self-conscious about eating due to perceived social pressures, and that obese people would therefore choose less food (i.e., suppress intake) when eating with others than when alone. Food selections of a matched sample of overweight and nonobese individuals were recorded, and subjects were observed as they took a seat to determine if they ate alone or with others. Caloric values of foods chosen for each meal were computed. Results for the number of calories indicated that, as predicted, overweight subjects purchased less food when accompanied than when alone. Nonobese individuals, by contrast, chose more food when with others than when alone. In addition, males chose more food than females and obese subjects chose more than normals. Results were discussed in terms of the impact of social variables on eating behavior, and the theoretical importance of weight consciousness and pereceived social pressures in understanding correlates of obesity. Several alternative explanations were addressed.     

Role of Peripheral Blood Mononuclear Cells in the Predisposition of Obese Individuals to Inflammation and Infection

ObjectiveTo compare the production of pro- and anti-inflammatory cytokines by peripheral blood mononuclear cells (PBMC) from obese but otherwise healthy individuals to that of normal-weight volunteers.Methods25 healthy normal-weight subjects and 41 obese individuals were enrolled. Weight and height were measured twice. PBMC were examined for their capacity to generate pro-inflammatory (TNF-α, IFN-γ, IL-1β, IL-6, IL-2) and anti-inflammatory IL-10 and IL-1ra) cytokines.ResultsPBMC from obese individuals, compared to those from subjects with normal weight showed an increased production of the pro-inflammatory cytokines IL-2 (6.7 ± 0.4. vs. 4.9 ± 0.3 ng/ml; p = 0.003), TNF-α (505 ± 45 vs. 277 ± 32 pg/ml; p = 0.001), and IFN-γ (93.8 ± 6.0 vs. 73.9 ± 2.7 ng/ml; p = 0.0016). However, PBMC from obese individuals produced a lower amount of the anti-inflammatory cytokine IL-10 (651 ±72 pg/ml) versus those from subjects with normal weight (951 ± 133 pg/ml; p = 0.039).ConclusionsThe findings imply that obese individuals are in a ‘low-grade inflammatory state’, presumed to be connected with metabolic and cardiovascular co morbidities. The surplus of pro-inflammatory cytokines produced by circulating mononuclear cells of obese individuals, together with those secreted by adipocytes and non-fat cells in the adipose tissue, may contribute to the predisposition of obese patients to inflammation and infections.Key Words: Obesity, Peripheral blood mononuclear cells, Cytokines, Inflammation, Infection     

Pregnancy and lactation in the obese rat: Effects on maternal and pup weights

Lister hooded female rats, fed palatable high energy foods and chow, weighed significantly more than chow-fed control rats before mating. A smaller proportion of the obese rats became pregnant, and they lost more litters in lactation. When litters survived (7±1 pups), maternal weight changes differed between groups during lactation. The controls gained 6.2±3.2 g, whereas the obese rats lost variable amounts of weight despite the continued availability of the palatable diet. The rats that were heaviest at mating and parturition and which showed the largest non-fetal weight gains in pregnancy (i.e., the “large weight loss group”) lost 60.6±4.8 g, while less obese rats which showed similar non-fetal gains to controls (i.e., the “small weight loss group”) lost 24.6±3.2 g. Thus the weights of all groups converged and were similar after three weeks of lactation, but diverged again after weaning. During lactation the total energy intakes and amounts of protein consumed by the obese rats were significantly below those of controls, and total fat intake was significantly elevated. Although litter size and pup weights did not differ significantly at birth, pups of obese mothers weighed significantly less than those of controls at weaning. Maternal obesity in lactation appears to influence both body weight regulation and lactational performance.     

Investigation of 3111T/C polymorphism of the CLOCK gene in obese individuals with or without binge eating disorder: Association with higher body mass index

Loss of circadian patterning of metabolism-related functions seems to play a role in the pathogenesis of obesity; therefore, it is reasonable to hypothesize that the functional 3111T/C single nucleotide polymorphism (SNP) of the (Circadian locomotor output cycles kaput) CLOCK gene may have a part in the genetic susceptibility to obesity. The aim of this study was to assess the frequencies of 3111T/C CLOCK gene SNP in overweight/obese subjects with or without binge eating disorder (BED) as compared to normal weight healthy controls. A total of 284 Caucasian subjects, including 92 normal weight healthy subjects and 192 overweight/obese patients (107 with BED) participated into the study. Genotype and allele frequencies did not significantly differ between normal weight controls and overweight/obese patients with and/or without BED. However, overweight/obese patients carrying the CC genotype had significantly higher values of body mass index (BMI) as compared to those carrying the CT and/or TT genotypes. Moreover, obese class III individuals had a significantly higher frequency of both the CC genotype and the C allele as compared to individuals with BMI < 40 kg/m². Present findings show for the first time that the 3111T/C SNP of the CLOCK gene is not associated to human obesity and/or BED, but it seems to predispose obese individuals to a higher BMI.     

Associations between repression,general maladjustment,body weight,and body shape in older males: The normative aging study

We examined relationships between repression, general maladjustment, body mass index (BMI), and waist-to-hip ratio (WHR). The participants were 1,081 healthy older men from the Normative Aging Study. Repression and General Maladjustment Scales of the Minnesota Multiphasic Personality Inventory were composite measures of personality. Repression was associated with lower BMI and WHR, and maladjustment with higher BMI and WHR. However, associations between WHR and personality dimensions were no longer significant when controlling for BMI, but associations between BMI and personality dimensions remained significant when controlling for WHR. These effects were explained by differing relationships between WHR, repression, and maladjustment for normal weight, overweight, and obese individuals. Specifically, associations between repression, maladjustment, and body shape were significant for normal weight and overweight individuals, but not for obese individuals. Health behaviors including smoking did not mediate relationships between repression, maladjustment, and body shape, but might be considered in future studies as mechanisms underlying links between personality and body shape.     

Activity-induced GLP-1 release in lean and obese subjects

The aim of the study was to determine whether physical activity stimulates GLP-1 release on the short-term in normal weight and in obese subjects compared to rest and, furthermore, whether modest weight loss affects GLP-1 release or sensitivity in the obese. Normal weight (n=28; 12 males, 16 females; BMI 22.9+/-1.4; age 35+/-12.7), as well as obese subjects (n=27; 21 males, 6 females; BMI 30.9+/-2.7; age 47.1+/-11.86) were tested in a resting and a physical activity condition. Obese subjects were matched over two groups for a weight loss period of 3 months. After weight loss, the tests were repeated. The area under the curve (AUC pmol/lxmin) for GLP-1 concentrations was significantly increased in the physical activity condition compared to rest in lean subjects (P=0.05) as well as in the obese subjects after weight loss (P<0.05), but not in the obese subjects before weight loss. Physical activity-stimulated GLP-1 release in lean and obese subjects after a weight loss period supports the idea of a neuroendocrine loop in addition to distal-intestinal stimulation of GLP-1 release. Modest weight loss might be effective for increasing GLP-1 sensitivity to acute stimulation.     

Effects of diet composition on food intake and carcass composition in rats

Since most of the weight-reduced obese humans are in a protein deficit state, this study was designed to examine whether a high protein diet (HP) enhances the restoration of lean body mass and facilitates the maintenance of weight loss. Obesity in rats was produced by 16 weeks of high fat diet (HF) feeding. In the 17th week, all HF-fed obese rats were fed a limited amount of control diet to normalize their body weights, but they still had more body fat content. The HF-fed rats were then divided into subgroups with different diets offered for 5 weeks: HP, HF or chow diet. A control group was fed the chow diet throughout the study. HP feeding maintained normal body weight and carcass composition in weight-reduced obese rats by reducing feeding efficiency levels to within normal ranges. Weight-reduced rats fed a chow diet, however, had more fat mass than controls and HF feeding stimulated weight gain again. Therefore, a HP diet has a higher probability of enhancing weight loss maintenance in weight-reduced obese subjects than does a usual well-balanced diet.     

Multidimensional scaling of musks.

Twenty subjects discriminated 14 commercial musks which varied in chemical structure. With the possible exception of 4 stimuli, the 14 musks could be discriminated from one another at a level greater than chance. Confusabilities and similarities were analyzed by the Guttman-Lingoes nonmetric multidimensional scaling procedure, SSAI, yielding a three-dimensional space. The multidimensional space achieved arranged the macrocyclic and nitro musks in separate regions of the space. Multidimensional scaling of confusabilities and similarties proved to be a more stable means of characterizing the differences between the 14 musks than ratings on adjective scales.     

Operant responding in rats with dietary obesity: A systematic replication

Two groups of female rats were first trained to bar press for food reward available on fixed ratio (FR) schedules ranging from 1–64. Then one group was fed a variety of palatable supermarket foods for 45 days while the controls were fed laboratory chow. Six of the 8 rats fed supermarket foods became moderately obese, gaining 3 times as much weight as controls. In tests conducted after the differential feeding treatment these moderately obese rats earned fewer reinforcements than controls on each of the FR schedules (1–64). Thus pretraining, which attenuates the deficit in operant performance for food displayed by dynamic stage rats with VMH lesions did not eliminate the deficit in food-motivated performance by rats that became obese by eating palatable foods.     

Emotional labeling and overeating in obese and normal weight individuals.

It was hypothesized that obese individuals respond to unlabeled high arousal by overeating, while no such response was predicted for labeled high arousal states. Obese and normal weight subjects were led to believe that they were hearing their own heart beat, and that it was either fast or slow. A label for this heart rate either was or was not provided, and subjects' eating behavior was measured unobtrusively. The results supported these hypotheses: aroused obese subjects ate more when they could not identify the cause of their arousal than when a label was known. When obese subjects were calm, the presence or absence of a label did not affect their eating. Furthermore, obese subjects showed significant affect reduction following eating. Normal weight subjects were not affected by the presence or absence of an arousal label. Instead, they ate more when they called themselves calm than when anxious, and more when hungry than when full.